Predictions of Public Opinion from the Mass Media: Computer Content Analysis and Mathematical Modeling (Contributions to the Study of Mass Media and Communications)

PREDICTIONS OF PUBLIC OPINION FROM THE MASS MEDIA Computer Content A n a l y s i s and Mathematical Modeling

D A V I D P. F A N

C o n t r i b u t i o n s to the S t u d y of M a s s M e d i a a n d C o m m u n i c a t i o n s , N u m b e r 12

G R E E N W O O D PRESS NEW YORK • WESTPORT, CONNECTICUT •

LONDON

L i b r a r y of C o n g r e s s C a t a l o g i n g - i n - P u b l i c a t i o n

Data

F a n , D a v i d P. Predictions o i public o p i n i o n from the m a s s media : c o m p u t e r c o n t e n t a n a l y s i s a n d m a t h e m a t i c a l m o d e l i n g / D a v i d P. F a n . p.

c m . — ( C o n t r i b u t i o n s to t h e s t u d y of m a s s m e d i a a n d

c o m m u n i c a t i o n s , I S S N 0732-4456 ; n o . 12) Bibliography:

p.

Includes indexes. I S B N 0-313-26296-9 (lib. b d g . : a i k . p a p e r ) 1. M a s s m e d i a — U n i t e d S t a t e s — I n f l u e n c e — D a t a p r o c e s s i n g . media—United States—Influence—Mathematical models. opinion—United States—Data processing. States—Mathematical models. Forecasting. I

Title

2. M a s s

3. P u b l i c

4. P u b l i c o p i n i o n — U n i t e d

5. P u b l i c o p i n i o n — U n i t e d S t a t e s -

6. C o n t e n t a n a l y s i s ( C o m m u n i c a t i o n ) — D a t a p r o c e s s i n g ,

II. Series.

MN90.M3F36

1988

303.3 ' 8 - d c l 9

88-5683

B r i t i s h L i b r a r y C a t a l o g u i n g i n P u b l i c a t i o n D a t a is a v a i l a b l e . C o p y r i g h t © 1988 by D a v i d P. F a n A l l rights r e s e r v e d . N o p o r t i o n of t h i s b o o k m a y be reproduced, by a n y process or technique, without the e x p r e s s w r i t t e n c o n s e n t of t h e p u b l i s h e r . L i b r a r y of C o n g r e s s C a t a l o g C a r d N u m b e r : 88-5683 ISBN:0-313-26296-9

First p u b l i s h e d in 1988 G r e e n w o o d Press, Inc. 88 Post R o a d W e s t , W e s t p o r t , C o n n e c t i c u t 06881 P r i n t e d i n t h e U n i t e d S t a t e s of A m e r i c a

T h e paper used in this book complies with the P e r m a n e n t P a p e r S t a n d a r d i s s u e d by the N a t i o n a l Information S t a n d a r d s O r g a n i z a t i o n (Z39.48-1984). 10 9 8 7 6 5 4 3 2

1

DEDICATION To M ary se

Copyrighte

Contents

Kill

A CK NOW

I.

« V I* I*

3

Ouüine

il

Organization

7

9 1.1

Strategies Used in Formulating Ideodynamics

1.2

Nature of the Population

10

1 3

Nature of Persuasion

11

1.4

Nature o f Persuasive Messages

12

1.5

Relationships between ldeodynamic Structures

16

1.6

Overview of Opinion Calculations

16

1.7

Details of Opinion Calculations for the A wares

17

1.8

Details of Opinion Calculations for the Unaw ares

22

1.9

Time Scale of ldeodynamic Analyses

22

Figures 1.1-1.3

24

9

VIII

Contents

2.1

Significant Features of Ideodynamics

27

2.2

Model Comparisons

32

CHAPTER

3: D A T A F O R

CALCULATING

PUBLIC

OPINION

37

3.1

T i m e Series o f O p i n i o n Polls

37

3.2

Relevant Persuasive Messages in the Associated Press

39

Figure 3.1

43

CHAPTER

4; C O M P U T E R T E X T A N A L Y S I S RV OF SUCCESSIVE FILTRATIONS

METHOD 45

4.1

General Text Analysis Programs

45

4.2

Strategy for Content Analysis Using Successive Filtrations

46

4.3

Sketch of Filtration and Scoring Computer Program Runs

47

4.4

Text Analyses for Defense Spending

49

4.5

Text Analysis for Troops i n Lebanon

50

4.6

Democratic Primary

52

4.7

Text Analysis for the Economic Climate

53

4.8

Text Analysis for Unemployment versus Inflation

53

4.9

Text Analysis for Contra A i d

53

4.10

Summary Features of Text Analysis by Successive Filtrations

54

4.11

Extensions o f the Text Analysis Procedure

56

CHAPTER 5:

PROJECTIONS

OF PUBLIC OPINION

57

5.1

Opinion Predictions for Defense Spending

57

5.2

O p i n i o n Predictions for Troops in Lebanon

62

5.3

Opinion Predictions for the Democratic Primary

64

5.4

Opinion Predictions for the Economic Climate

65

5.5

Opinion Predictions for Unemployment versus Inflation

66

5.6

Opinion Predictions for Contra A i d

66

Contents

Lx,

5.7

Summary o f Constants Used i n Poll Projections

67

5.8

Summary of Statistics for Poll Projections

68

Tables 5.1-5.3

69

Figures 5.1-5,45

72

CHAPTER

6: M E T H O D O L O G I C A L OF W O R K

SIGNIFICANCE 117

6.1

Validation o f Ideodynamics

H8

6.2

Data and Issues for Successful Ideodynamic Calculations

119

6.3

Positions for W h i c h Persuasive Messages Are Scored

121

6.4

Computer Text Scoring

121

6.5

Ideodynamic Calculations o f Opinion Time Trends

123

6.6

Insensitivity o f Predictions to the Starting Opinion Values

127

6.7

Interpretations for A l l Ideodynamic Parameters

127

6.8

Significance of No Opinion Change

127

6.9

Analysis o f Persuasive Messages Acting on Public Opinion

128

C H A P T E R 7: S I G N I F I C A N C E O F W O R K T O OF OPINION FORMATION

THEORIES 129

7.1

Mass Media Messages and Opinion Leadership

129

7.2

Reinforcing Role of Persuasive Messages

131

7.3

Cumulative Effects of Information Rather than M i n i m a l Effects o f the Media

132

7.4

Caveats for Laboratory Experiments

134

7.5

L a w o f the 24-Hour Day

134

7.6

Interpretations o f Ideodynamic Parameters

7.7

Nature o f Effective Persuasive Messages i n the Mass Media

APPENDIX

A.2

A:

MATHEMATICS

Structure of the Population

OF

IDEODYNAMICS

T35

136

141

141

JC

Contents

A.3

Structure o f Messages

142

A.4

Nomenclature Simplification

144

A.S

Infon Properties

144

Ai)

I n f o n Persuasive Force

144

A.7

Information Influencing the Unawares

144

A.8

Information Influencing the A wares

145

A.9

Effect o f Information on the Population

147

A. 10 Modifications for AP Infons Assuming No Unawares

149

A.11

152

Comparison w i t h U n i f o r m Distribution

A. 12 Modifications for AP Infons Assuming Non-negligible Unawares

153

A. 13 Extensions to Very L o n g Times

154

A . 14 Models w i t h N o Dependence on Subpopulations

154

o PIN^ (\\* r H AN f IF

L A T I N F I

L55

R.l

Defense SpcndinR--1977-1984

155

B .2

Troops in Lebanon-1983-1984

156

B.3

Democratic Primary-1983-1984

[56

B.4

Economic CUmate-1980-1984

157

B.5

Unemployment versus Inflation-1977-1980

157

B .6

Contra Aid--1983-1986

157

T a h i t i * _1-tt.fi

159

APPENDIX

C:

SUMMARIES

OF

TEXT

ANALYSES

165

C. 1

Strategy for Content Analysis by Successive Filtrations

165

C.2

Text Analysis for Defense Spending-Including Detailed Example

165

C.3

Text Analysis for Troops i n Lebanon

172

C.4

Text Analysis for the Democratic Primary

173

C.5

Text Analysis for the Economic Climate

173

Contents

xi

C.6

Text Analysis for Unemployment versus Inflation

174

C.7

Text Analysis for Contra A i d

174 116

P R O J K C T I O N ^ ^

PIR' ir

OPINION

^

D.l

Computations o f Persuasive Forces

183

D.2

Population Conversion Models

183

D.3

O p i n i o n Projections

184

AUTHOR

1NDFX

193

SUBJECT

INDEX

122

Tables and Figures

Figure 1.1

Example o f persuasive forces o f infons.

24

Figure 1.2

Population conversion model for defense spending.

25

Figure 1.3

Illustration o f the impact o f a single persuasive infon favoring m o r e defense spending.

26

Figure 3.1

Poll data for defense spending.

43

Table 5.1

Statistical comparisons for opinion projections.

69

Table 5.2

Candidate name counts in dispatches retrieved for the Democratic primary.

70

Table 5.3

O p t i m a l constants for opinion projections.

71

Figure 5.1

Persuasive forces o f A P infons scored for favoring more, same, and less defense spending.

72

Persuasive forces o f A P infons scored for favoring more and less defense spending.

73

Opinion on defense spending from dispatches scored to favor more, same, and less spending.

74

Figure 5.4

Constant optimization curves for defense spending.

75

Figure 5.5

Opinion f r o m a subset o f A P dispatches scored to favor more, same, and less defense spending.

76

Opinion f r o m another subset o f A P dispatches scored to favor more, same, and less defense spending.

77

Opinion on defense spending assuming the entire population favored more spending at the time o f the first scored A P infon in January 1977.

78

Figure 5.2

Figure 5.3

Figure 5.6

Figure 5.7

Copyrighte u

111 d

i d idi

xiv

List of Tables and

Figures

Opinion on defense spending from dispatches scored to favor more and less defense spending only.

79

Persuasive forces o f A P infons f r o m 1977 to 1986 scored for favoring more, same, and less defense spending.

80

Figure 5.10

Opinion on defense spending from 1977 to 1986.

81

Figure 5.11

Constant optimization curve for the contributions from paragraphs favoring less defense spending.

82

Effect o f stories on waste and fraud on public opinion on defense spending.

83

Figure 5.13

Persuasive forces for troops i n Lebanon from A P infons only.

84

Figure 5.14

Persuasive forces for troops in Lebanon from A P infons w i t h and w i t h o u t a truck bombing infon favoring more troops

85

Population conversion model for actions o f infons favoring more, same, and less troops in Lebanon.

86

Optimizations for the modified persuasibility constant, the weight for paragraphs favoring less troops, and the value o f the truck bombing infon favoring more troops.

87

Optimization curves far the persistence half-life and the value o f the truck bombing infon favoring less troops.

88

Figure 5.18

O p i n i o n on troops i n Lebanon assuming o n l y A P infons.

89

Figure 5.19

Opinion on troops in Lebanon w i t h a truck bombing infon favoring more troops.

90

Comparison o f opinion projections w i t h (solid line) or without (dotted line) the truck bombing infon favoring more troops.

91

Persuasive forces from A P infons w i t h and without a truck bombing infon favoring less troops.

92

Opinion projections w i t h and w i t h o u t a truck bombing infon favoring less troops.

93

Persuasive forces favorable to Democratic presidential candidates from AP paragraphs scored using bandwagon words

94

Figure 5.8

Figure 5.9

Figure 5.12

Figure 5.15

Figure 5.16

Figure 5.17

Figure 5.20

Figure 5.21

Figure 5.22

Figure 5.23

Figure 5.24

Persuasive forces unfavorable to Democratic presidential candidates from A P paragraphs scored using bandwagon words

Figure 5.25 Figure 5.26

Persuasive forces o f A P infons scored by name count only. Population conversion model for actions o f infons scored using bandwagon words.

95 96

97

Copyrighte

List of Tables and Figures

Figure 5.27

Figure 5.28

Figure 5.29

xv

Population conversion model for actions o f infons scored by name count only.

98

Optimization curves for constants for the Democratic primary.

99

Opinion on Democratic candidates when infons were scored by the bandwagon analysis.

100

Opinion on Democratic candidates when infons were scored by name count only.

101

Persuasive forces from A P paragraphs favoring better, same, and worse economic conditions.

102

Population conversion model for actions o f infons favoring better, same, and worse economic conditions.

103

Figure 5.33

Optimization curves for constants for the economic climate.

104

Figure 5.34

Opinion on economic climate.

105

Figure 5.35

Persuasive forces o f A P infons favoring unemployment more important equal importance, and inflation more important.

106

Population conversion model for actions o f infons favoring unemployment more important, equal importance, and inflation more important.

107

Optimization curves for the modified persuasibility constant and the infon weighting constants for unemployment versus inflation.

108

Optimization curve for the persistence constant for unemployment versus inflation.

109

Opinion favoring unemployment more important, equal importance, or inflation more important.

110

Persuasive forces o f A P infons scored by the author as favoring and opposing Contra aid.

111

Persuasive forces o f A P infons scored by S w i m , Miene, and French as favoring and opposing Contra aid.

112

Population conversion model for actions o f infons favoring and opposing Contra aid.

113

Figure 5.43

Constant optimization curves for Contra aid.

114

Figure 5.44

Opinion favoring and opposing Contra aid using infon scores by the author.

115

Opinion favoring and opposing Contra aid using infon scores by S w i m , Miene, and French.

116

Polls on the desirability o f increasing defense spending.

159

Figure 5.30

Figure 5.31

Figure 5.32

Figure 5.36

Figure 5.37

Figure 5.38

Figure 5.39

Figure 5.40

Figure 5.41

Figure 5.42

Figure 5.45

Table B. 1

Copyrighte U

I I ICLlwl l u i

xvi

List of Tables and

Table B.2

Figures

A B C News Poll on the stationing o f American troops i n Lebanon.

160

Table B.3

A B C News Poll on the Democratic primary.

161

Table B.4

A B C News Poll o n the economic climate.

162

Table B.5

N B C News Poll on the importance o f unemployment versus inflation.

163

Table B.6

Polls on the desirability o f sending Contra aid.

164

Table C. 1

Summary o f text analysis for defense spending.

176

Table C.2

Summary o f text analysis for defense waste and fraud.

177

Table C.3

Summary o f text analysis for troops i n Lebanon: scoring for more, same, and less troops.

178

Summary o f text analysis for Democratic primary: scoring for bandwagon words.

179

Summary o f text analysis for economic climate: scoring for better, same, and worse.

180

Summary o f text analysis for unemployment versus inflation: scoring for unemployment more important, equal importance, and infladon more important.

181

Summary of text analysis for Contra aid: scoring for infons favoring and opposing aid.

182

Table C.4

Table C.5

Table C.6

Table C.7

Copyrighte

Acknowledgments

In acknowledging the principal influences o n this book, I w o u l d like to begin w i t h m y father, Professor Hsu Y u n Fan, a physicist w h o persuaded me to obtain a bachelor's o f science degree i n physics rather than pursue m y o w n interests i n the social sciences. His viewpoint was that physics was easy to study given its precision and concreteness w h i l e the social sciences were m u c h more d i f f i c u l t w i t h their uncertainties and ambiguities. This background i n physics is seen i n the mathematical m o d e l o f ideodynamics at the core o f this book. That model uses equations i n the tradition o f such areas in physics as dynamics and thermodynamics. Next, I w o u l d like to acknowledge Professor Cyrus L e v i n t h a l , the thesis advisor for m y P h . D . studies i n biology. H e showed me how t o formulate assumptions w h i c h c o u l d both reflect problems o f interest t o me and be translated i n t o mathematical equations. Indeed, the equations o f ideodynamics are very similar t o those underlying my Ph.D. thesis on the metabolism o f the messenger ribonucleic acids involved i n the expression o f genes. Throughout m y studies i n the physical sciences, 1 have been struck by the fact that important advances often depended on the discoveries o f elementary particles such as protons i n nuclei, nuclei i n atoms, and atoms i n molecules, t a c h o f these particles has a l i m i t e d number o f well-defined and quantifiable characteristics. S i m i l a r l y , the rapid progress i n biology i n recent years has been based on the concept o f the gene, w h i c h is a discrete unit o f inheritance w i t h a simple structure and a small number o f properties. These are but some examples o f how the understanding o f complex natural phenomena were advanced by analyses o f discrete elemental units. The direct consequence o f this viewpoint is m y postulate that persuasive messages can be coded as infons. M y studies in biology included the use o f both genetic and biochemical techniques. The thought patterns i n these areas f o r m the bases o f the n e w method o f content analysis i n this book. F r o m biochemistry, 1 learned that the study o f complicated materials frequently benefits f r o m a series o f purification steps, each one removing extraneous components to y i e l d progressively more homogeneous preparations enriched i n relevant materials. This logic led to the strategy o f successive "filtrations" during the text analyses. The detailed strategies f o r the text filtrations and final scoring are derived i n large part f r o m the principles o f gene expression i n genetics. I n more recent times, m y biological research has turned t o w a r d the study o f the k i l l e r T cells o f the immune system involved i n protection against viral infections and rejection o f organ transplants. For those studies, I realized that the analyses c o u l d be greatly aided by a mathematical model to describe the k i l l i n g activity and a computer program based o n the model t o process the data. This need l e d me t o return to mathematical modeling, w h i c h I had not performed since m y Ph.D. days. I also learned computer languages i n order to write the programs needed for the data analyses. These exercises i n mathematical modeling and computer p r o g r a m m i n g gave additional

Copyrights

xviii

Acknowledgments

impetus to m y desire to examine the social sciences f r o m a mathematical and computational perspective. I had maintained an active interest i n social phenomena from before m y university days. Remaining at the general level, I w o u l d l i k e to thank the Graduate School o f the University o f Minnesota for the funds used for the data gathering and the hiring o f the graduate assistants who scored the text for the case o f the Contras i n Nicaragua. I n addition, i t is necessary to note the crucial role o f the tenure system at A m e r i c a n universities w h i c h permitted me to explore the social sciences using the techniques and thought patterns from my home department in the biological sciences. It was also very useful for my university to permit a sabbatical leave, during which I performed much o f the research i n this book. A number o f h e l p f u l investigators have also contributed i m p o r t a n t l y i n ways directly useful for this book. M a n y o f these individuals were associated w i t h the University o f Minnesota. W i t h o u t being exhaustive, and i n order o f topic rather than importance, I w o u l d like to note i n particular: Professors M i c h a e l Simmons, James Curtsinger, and Frank Enfield o f m y o w n Department o f Genetics and Cell Biology, Dennis Cooke o f the Department o f A p p l i e d Statistics, and Hans Weinberger o f the Institute o f A p p l i e d Mathematics w i t h w h o m I discussed many o f the details o f the mathematical modeling and statistical concerns. Thanks also t o Professor Donald McTavish o f the Department of Sociology, who graciously permitted me to explore his M C C A computer content analysis program; the late Professor F. Gerald K l i n e o f the School o f Journalism and Mass Communications, w h o made very useful suggestions, such as the use o f the Associated Press to represent the A m e r i c a n mass media; Professors John Sullivan o f the Department o f Political Science, and Eugene Borgida o f the Department o f Psychology w i t h w h o m I discussed the relationships between my w o r k and those o f others i n the social sciences; and Professor John Freeman o f the Department of Political Science who gave this book a careful and critical reading. I w o u l d also l i k e to thank Professor Bruce Russett o f Y a l e U n i v e r s i t y and the anonymous reader for the Greenwood Press f o r the useful comments m a d e b e f o r e completion o f this book. A m o n g other colleagues not at Minnesota, I w o u l d like to give special thanks t o Professors B e n j a m i n Page and Robert Shapiro and their associates at the National O p i n i o n Research Center i n Chicago for extremely helpful discussions, and particularly for access to their many time series o f p o l l data f r o m which I chose several to analyze. Those p o l l data were absolutely indispensible for the studies. Obviously, none o f these acknowledgements i m p l y anyone else's responsibility for this book. Clearly, that responsibility is totally mine.

Copyrighte

Predictions of Public Opinion from the Mass Media

Copyrighte

Introduction

This book concerns the power o f information o n society. T h e central thesis is that public o p i n i o n can be swayed in a predictable fashion by messages acting o n the populace. W h e n the b u l k o f the relevant messages are i n the press, then the press becomes the p r i n c i p a l determinant o f society's attitudes and beliefs. Although previous w o r k has suggested that the press is able to set the agenda f o r public discussions, this book is unusual i n demonstrating that the press is also able to mold o p i n i o n w i t h i n agenda items. The importance o f the press o n opinion has long been recognized. This is seen i n the concept o f governmental press censorship, w h i c h was invented long ago. However, the assignment o f the preeminent role o f the press in opinion formation i n a free democracy is i n apparent conflict w i t h a sizable body o f literature describing the " m i n i m a l effects o f the media." W i t h this shield, journalists and editors could w o r k without feeling that every one o f their daily choices was affecting opinion. However, the conflict between press importance and its m i n i m a l effect is more apparent than real. As summarized i n Chapter 7, the impact o f a piece o f news is most appropriately assessed quantitatively. I n other words, messages i n the mass media should be given numerical strengths. A l t h o u g h any one news story, or restricted group o f media messages, can have effects ranging f r o m very small through very large, opinions can frequently be computed f r o m the cumulative effect o f all news stories, most o f w h i c h can indeed have relatively m i n i m a l effects individually. Therefore, i n general, the concept o f the " c u m u l a t i v e effects o f i n f o r m a t i o n " comprising mainly mass media information for many issues—is more useful than the law o f m i n i m a l effects. T h i s idea o f the c u m u l a t i v e impact o f i n f o r m a t i o n s t i l l permits w o r k i n g members o f the press t o proceed w i t h o u t constantly w o r r y i n g about the effects o f their every w o r d . I n d i v i d u a l news items are themselves still likely to have small impact. However, over the long term, all the effects accumulate and the totality o f press messages is capable o f being the major influence on o p i n i o n . Thus society should realize that i n d i v i d u a l messages can indeed have m i n i m a l effects, but w i t h long-term trends being o f great importance. As j u s t noted, this book does not propose that the press is always the dominant force i n o p i n i o n f o r m a t i o n . Rather, the hypothesis is that i t is the totality o f relevant information w h i c h w i l l shape opinion. Therefore, the press w i l l only be the primary influence i f other messages are o f minor importance. Obviously, the importance o f the press is related to its credibility. This trust has n o direct relationship to whether the public ranks the press as credible in opinion polls. I t is only essential that the public as a whole uses no alternate sources o f

Copyrighte U

4

Introduction

information for polled issues discussed i n the mass media. For example, the press i n closed societies is not l i k e l y to be the main determinant o f attitudes i f its reputation is so l o w that sizable portions o f the populace rely on rumor and the underground press. I n an open society like that i n the United States, trust in the press is likely t o be greater. It was to explore the domain o f media dominance i n opinion formation that studies i n this book were performed on a variety o f topics. Issues were chosen f r o m both the domestic and foreign policy arenas. The two issues w i t h the clearest foreign policy implications concerned whether more troops should be sent to Lebanon (1983¬ 1984) and whether U.S. aid should be sent to the Contra rebels in Nicaragua (1983¬ 1986). The domestic topics included those on governmental policy and economic issues. The policy question was whether or not more should be spent for national defense (1977-1986). The two economic issues focused on whether unemployment or inflation was the more important problem (1977-1980), and whether the economic climate was i m p r o v i n g (1981-1984). The remaining domestic issue was voter preference for the best candidate in the Democratic presidential primary (1983-1984). For all these cases, the mass media has the p r i n c i p a l role in influencing public opinion. For the Democratic primary, the press could not be expected to be the dominant influence i f there were important additional informational sources such as campaign advertising. It was to avoid the complication o f such alternative information sources that the Democratic primary was studied before the I o w a caucuses, a time when national media stories should have been the most important source o f persuasive messages for opinion nationwide. A t these early times, campaign advertising was negligible countrywide w h i l e o p i n i o n p o l l results were obtained f r o m this large population base. F r o m the discussion above, a quantitative analysis is able to reconcile the m i n i m a l effects o f the media w i t h the cumulative effects o f information. Obviously, such quantitative assessments imply a mathematical analysis, and, indeed, this book describes the new mathematical model of ideodynamics for calculating the impact o f information on the population. This model was constructed on the premise that time trends o f opinion percentages could be predicted from the relevant messages available to the public. This model also has the important feature that i t can unify many seemingly c o n f l i c t i n g results. A useful analogy is the story o f b l i n d men reporting on an elephant; the man studying the leg could report that the elephant was like a tree trunk w h i l e the man examining the tail could find that the elephant was most like a rope. The contradiction vanishes when the entire elephant is considered in overview w i t h both the leg and tail being special cases o f the more general model w h i c h is the elephant. In the same way, the cumulative effects o f information can encompass both individual groups o f mass media messages having m i n i m a l effects and the totality o f the media having major effects. The unifying power o f ideodynamics derives importantly f r o m its quantitative nature. By g i v i n g numerical values to the contributions o f different phenomena, there is no need to assert or i m p l y that certain phenomena are always more or less important than others. Instead, the question becomes the relative importance o f different phenomena under specific circumstances. For instance, this book demonstrates that o p i n i o n f o r m a t i o n is frequently affected rather l i t t l e by reinforcement o f previous opinion due to the resolution o f cognitive dissonance i n the direction of favorable information. This statement does not deny the existence o f opinion reinforcement and does not assert that such reinforcement is never important. In fact, such reinforcement is e x p l i c i t l y included i n ideodynamics. Instead, the statement is merely that such reinforcement is small relative to the forces in the mass media causing opinion change for cases like the six studied in this book. The elephant analogy can be extended to the emphasis i n this book on the global behavior o f the population. The concern is less with the behavior o f subpopulations and selected media messages than on the effect o f the totality o f messages on attitudes w i t h i n the entire population. In the analogy, the theory is less concerned w i t h the

Introduction

5

behavior o f the parts o f the elephant during locomotion than w i t h the path taken by the elephant as a whole. There is no i m p l i c a t i o n that the elephant's path is more important to study than the effects o f the legs, for example, on elephant m o v e m e n t The analogy is only pursued to highlight the fact that this book is mainly about the macro effects o f the totality o f information on overall attitudes without a systematic dissection o f a l l c o n t r i b u t i n g factors, even t h o u g h some such dissections are performed The theory is also formulated w i t h very few parameters so that i t can be tested e m p i r i c a l l y . E m p i r i c a l testability means that confidence i n the model c o u l d be derived f r o m f i n d i n g that stories i n the Associated Press c o u l d give good time trend predictions o f public opinion percentages over time spans ranging from three months to nine years. The success o f applications to real data is crucially important because i t can demonstrate that approximations and calculated population parameters are reasonable, even though some might seem heroic at first glance. A m o n g the parameters examined, the most interesting lead to the conclusions that there is no lag before the onset o f persuasion and that the impact o f a mass media message decreases exponentially w i t h a half-life o f only one day. This means that the effect is entirely dissipated within a week. These results argue that there is no two-step transfer o f i n f o r m a t i o n f r o m the press to the populace v i a o p i n i o n leaders. Rather, the people are influenced directly by the mass media. Examination o f the equations also shows w h y the big lie can be effective i n propaganda, w h y the causes o f fringe groups can be helped by terrorism, and w h y the political Left and Right can both accuse the press o f unfair bias. T o be consistent w i t h the previous discussion on the importance o f quantitative assessments, these parameters might have other values i n future studies, resulting i n different implications for other circumstances. The mathematical predictability o f opinion indicates a large public malleability in the hands o f the mass media. This malleability is l i k e l y to arise f r o m the law o f the 2 4 - h o u r day w h i c h is first i n t r o d u c e d i n Chapter 1. T h i s l a w s i m p l y acknowledges that the public is constantly bombarded by new information, w i t h so much being available that a person can only reflect carefully on a small fraction. As a result, most information is taken at face value. T h i s importance o f superficial information is at the very heart o f the words reputation and prejudice. These words both i m p l y d e c i s i o n m a k i n g based on observations or information from the past. B y b r i n g i n g such prior information to bear, an i n d i v i d u a l is spared the time and effort needed to make a careful detailed examination o f the current details o f the issue. I n fact, the time needed to make careful evaluations o f all current information simply may not be available. These considerations stress another o f the recurring themes i n this book, the importance o f real t i m e for examinations o f social issues. I t is not enough to describe pathways and sequences for social changes w i t h o u t an appreciation o f the time spent i n each step. For example, i t has already been mentioned that real-time constraints lead to superficiality i n decisionmaking for the population as a whole. Such superficiality is not likely to be observed when people are forced to ponder issues carefully i n laboratory studies, focus groups, and interviews where people are asked to reconstruct their states of mind. O b v i o u s l y , superficial thoughts are simpler to analyze mathematically than complex ones. Therefore, i t is reasonable t o use straightforward mathematical equations t o calculate public opinion f r o m persuasive messages. For a wide variety o f issues, like those mentioned above, persuasion is further due to i n f o r m a t i o n largely confined to the mass media. Throughout this book, the emphasis is on message impact w i t h little discussion o f message generation. This emphasis certainly does not mean that message senders w o r k i n a vacuum, oblivious to other factors, including actual or anticipated public o p i n i o n . I n fact, both Chapters 2 and 7 discuss how message generation i n the model can be dependent o n o p i n i o n . However, the interdependence o f o p i n i o n formation and message generation do not exclude these t w o phenomena f r o m being studied separately.

Cop*

6

Introduction

In the same way, even though a nuclear w a r can i n v o l v e an exchange o f missiles, it is s t i l l possible to study separately missile damage and missile launch. Indeed, a thorough understanding of missile impact w i l l aid in a complete analysis of nuclear war. By analogy, a complete description of the persuasive process can benefit f r o m a careful study of the impact o f communications on the populace. A n accurate description o f message effect can then be used as a f i r m base from which to continue the analysis o f message generation. The key to uncoupling missile launch f r o m missile impact is a valid description of the pertinent properties o f the missile, namely its trajectory and megatonnage. Once these properties are recognized, it is possible to model both missile launch and impact i n terms o f these parameters. Given the appropriate parameters, the analyst o f missile impact can predict the devastation w i t h o u t regard for the factors influencing the launch. I n the same manner, once persuasive messages are coded in terms o f the equivalents of trajectory and megatonnage, knowledge o f the sender's motives is not important i n considering message effect. A n important goal o f this book is t o develop and validate parameters which are sufficient to describe a persuasive message without regard for the message sender. A later analysis could then turn to message generation, w i t h the messages coded in the same terms. W h e n both message generation and impact are understood, then the trade o f messages in a persuasive process can be explored i n the same way that an exchange o f missiles can be examined for a nuclear war. The w o r k i n this book focused o n message i m p a c t rather than message generation because impact was likely to be more predictable. The law o f the 24-hour day argues that o p i n i o n w i l l usually reflect messages. I n contrast, o p i n i o n is more frequently o n l y one factor rather than the sole factor i n influencing the message sender. I n addition to o p i n i o n , new discoveries and facts can greatly affect the messages broadcast I f not, nothing new w o u l d ever be disclosed by the mass media since the very novelty o f a discovery must mean that very few people are aware o f it and hence that there is very little opinion favoring the dissemination o f this rare event. Therefore, a thorough analysis o f the dissemination o f mass media messages must include n o t o n l y an examination o f o p i n i o n but new events w h i c h are unpredictable by their very nature. The foreseeable response o f the populace to information is clearly important for understanding social behavior for issues as t r i v i a l as fads and as profound as w a r and peace. For instance, the predictability and consequent superficiality o f information absorption suggest that the average member of the public i n modem democracies may make no more carefully reasoned decisions for most issues than persons i n more p r i m i t i v e societies. Furthermore, since the model should apply to all societies, the predictions should be as v a l i d i n dictatorships as i n democracies so long as a l l the information available to the public can be coded.

OUTLINE A s discussed above, this book explores the new mathematical m o d e l o f ideodynamics describing social responses to information. A l t h o u g h the outlines have already been published, the model has been m o d i f i e d as a consequence o f its application to empirical data, the focus o f this book. Therefore, this book begins w i t h a presentation o f ideodynamics followed by an examination o f the ability o f the model to incorporate previous theories (Chapters 1 and 2). Then data applications are considered (Chapters 3 to 5). A t the end (Chapters 6 and 7), there is a discussion o f the conclusions to be drawn from the w o r k . The e m p i r i c a l testing o f the model involves its use to predict p u b l i c o p i n i o n f r o m i n f o r m a t i o n i n the mass media. The computed o p i n i o n is i n the f o r m o f percentage support o f polled positions w i t h the values appearing as continuous time trends calculated every six or twenty-four hours. Therefore, to the extent that p o l l s are like snapshots, the trends from these new computational methods are like moving

Copyrighte i

Introduction

7

pictures, capable o f filling in the gaps between actual p o l l points and extending opinion estimates t o times when polls have not yet been taken. The studies i n this book show (hat media exemplified by Associated Press ( A P ) s t o r i e s - c a n be used t o predict the o p i n i o n percentages published by reputable national polling organizations such as A B C News, one o f the major p o l l sources for this book. Successful projections were made for all six o f die issues analyzed. For each issue, the procedure consisted of: (1) gathering the texts o f A P dispatches relevant to the issue, (2) scoring each story for the extent to w h i c h i t supported different positions w i t h i n the issue, (3) using these scores i n the equations o f ideodynamics to compute o p i n i o n time trends, and (4) comparing the computed time trends w i t h published poll data. Each study used the new I n f o T r e n d ^ methods. The first step i n the InfoTrend procedures relies o n a previously unreported computer procedure for scoring A P stories. T h e second step employs computer solutions f o r the equations o f ideodynamics.

mass

messages-as

ORGANIZATION Chapter 1 describes the deduction o f ideodynamics f r o m know n phenomena i n the area o f persuasion. A n appreciation o f this chapter is essential for understanding the opinion computations. Ideodynamics considers persuasive messages to have structures similar to that o f M I R V e d missiles. The analogs o f the independent warheads ( M u l t i p l e Independent Reentry Vehicles) are message components, each one able to have an impact o n appropriate target subpopulations. These message components are called infons. For example, a persuasive message relevant to defense s p e n d i n g could have one i n f o n or component favoring more spending, another infon favoring same spending, and yet another favoring less spending. L i k e M I R V e d missiles, a l l the infons are bundled together i n the same persuasive message and launched at the p o p u l a t i o n . T h i s chapter models mathematically the effects o f infons on the population. Chapter 2 discusses the m a i n features o f ideodynamics i n the context o f previous models for the impact o f information on society, especially those in the area o f public o p i n i o n . Therefore, readers p r i m a r i l y interested i n the new methodology can skip this chapter. Chapter 3 describes the data used for the calculations and therefore should be read. Chapter 4 describes the new InfoTrend computer method for obtaining infon scores for the messages discussed i n Chapter 3. Readers less interested i n computer content analysis than opinion projections need not read Chapter 4. This chapter is free-standing, describing a general technique o f content analysis able to score any text for the extent to w h i c h different ideas are favored. The methodology is not restricted to generating i n f o n scores s u p p o r t i n g different positions. F o r example, i t is also possible to use this text analysis f o r other purposes such as assessing whether a letter o f recommendation comments favorably on specific traits for a person being discussed. Since the major function o f Chapter 4 is to produce infon scores, i t is possible to bypass this chapter, for o p i n i o n projection studies, by using alternate scoring procedures. The most straightforward way w o u l d be to ask human judges to score the persuasive messages. However, the computer methods do have distinct advantages: the critical features o f the persuasive text are explicitly defined; large amounts o f text can be scored; a l l scoring criteria are applied u n i f o r m l y to the entire body o f text examined. I n f o T r e n d is a registered trademark for i n f o r m a t i o n a l analysis b y I n f o T r e n d I n c .

8

introduction

describes h o w infon scores for persuasive messages are used to compute expected public opinion and is the heart o f this book f r o m the standpoint o f opinion projections. Each o f the six studies is considered i n detail. M o s t o f the results fall into four major categories: 1. A set o f graphs describing the t i m e trends o f persuasive i n f o r m a t i o n favoring different positions, 2. A set o f graphs comparing published o p i n i o n - p o l l results w i t h o p i n i o n calculated on the basis o f infon scores and the first set o f published opinion percentages, 3. A set o f graphs showing the optimizations o f the various constants i n the Chapter

5

ideodynamic equations, and

4.

A table showing the goodness o f fît based on the squares o f the differences between the poll projections and the opinion percentages i n published polls. Chapter 6 examines the implications o f the studies and further applications o f the method. Strictly speaking, this chapter need not be read by those interested only in the technical aspects o f the methodology. However, this chapter is useful even for methodological considerations since i t examines both the strengths and robustness o f the techniques as w e l l as their weaknesses and limitations. Chapter 7 discusses the broader significance o f the w o r k i n this book to theories o f effective persuasion and examines the procedures by w h i c h ideodynamics can be extended to include theories o f message generation. Appendices. This book is written so that the reader can f o l l o w the main thrust of the arguments without a detailed study o f either the mathematics of ideodynamics or the computer text analyses. However, both o f these technical areas are explained more f u l l y i n the appendices: Appendix A for the mathematics o f ideodynamics, A p p e n d i x C for the computer text analyses, and A p p e n d i x D for the computer calculations o f opinion based on ideodynamics. The primary data for the analyses are also presented i n Appendix B . References are made to the appendices throughout the text Further technical details o f the procedures and computer programs used for this book are given in a pending patent application.

Copyrightec

1

Formulation of Ideodynamics

The m a i n thesis o f this book is that information controls public o p i n i o n . For many issues i n a free democracy, the d r i v i n g force f o r o p i n i o n change is persuasive messages i n the mass media. Support for this thesis derives f r o m the ability t o use messages i n the press to calculate time trends o f public opinion. The calculations are performed by computer using the new InfoTrend methods, and are divided into t w o main areas, those for text analysis and those f o r assessing the impact o f information on o p i n i o n . The studies are grounded in a general model for information impact applicable to the adoption o f both behaviors and attitudes. I n this book, however, the discussion w i l l focus o n attitudes since the applications are restricted to p u b l i c o p i n i o n . Appendix A presents an improved version o f the mathematical model w h i c h has already been c a l l e d i d e o d y n a m i c s ( F a n , 1984, 1985a, 1985b). T h e name ideodynamics is d r a w n f r o m idea w h i c h refers t o ideas, and dynamics, which emphasizes changes w i t h time. I n order t o present the arguments w i t h o u t undue distractions, this chapter discusses the formulation o f ideodynamics w i t h m i n i m u m reference to alternative models. Relationships to other models are discussed i n Chapter 2. Ideodynamics was developed to explain a number o f k n o w n features concerning the formation o f public o p i n i o n . Therefore, the model is deduced f r o m phenomena which needed to be explained and shares the deductive approach used by other workers like Downs (1957) i n An Economic Theory of Democracy.

1.1 S T R A T E G I E S U S E D I N F O R M U L A T I N G

IDEODYNAMICS

One o f the essential considerations i n f o r m u l a t i n g ideodynamics was that the model should be testable using data f r o m observations. This condition is important since, as w i t h any mathematical model, simplifying approximations are needed. The p r e d i c t i v e powers o f a model p r o v i d e a good test o f its soundness. I f the approximations are v a l i d for a large number o f circumstances, the model should successfully predict measured values for many cases. I f the approximations are appropriate f o r o n l y a small number o f examples, then the predictions f r o m the model should frequently fail. Therefore, empirical testability provides a method f o r assessing the validities o f the approximations. A s w i t h a n y set o f s i m p l i f i c a t i o n s , i t is a l w a y s possible t o i m a g i n e complications w h i c h w i l l lead to failure o f the approximations. Nevertheless, the model can succeed i n a large number o f instances i f the complications usually make o n l y m i n o r contributions w i t h i n the total constellation o f relevant phenomena. The usefulness o f the simplifications w i l l depend on the extent to w h i c h the resulting

Copyright

10

Predictions

of Public

Opinion from

the Mass

Media

predictions are accurate. T o guard against the possibility that the accuracy is fortuitous, the model can be tried under a variety o f conditions. The calculations w i l l gain in robustness as acceptable predictions continue to be obtained. The important advantage of empirical testability is that a crucial criterion for a model's success can be the predictions obtained. I n this way, the v a l i d i t y need not rely solely o n the plausibility o f the argument. T e s t a b i l i t y , however, is clearly a two-edged s w o r d . A l t h o u g h accurate predictions can argue that imagined concerns are o f m i n o r importance, consistently inaccurate predictions would also force abandonment of the model. Once a model can be tested empirically, it is both possible and desirable to be bold in postulating s i m p l i f y i n g approximations. After a l l , i f the simplifications are too extreme, then the model w i l l fail to give useful predictions. Therefore, a reasonable strategy is to begin w i t h the m i n i m a l model i n v o l v i n g the smallest number o f parameters. More complicated approximations involving more parameters w o u l d only be added i f the m i n i m a l model d i d not eive good predictions. Another important advantage o f using simple approximations is that the mathematics and resulting computations are less complicated. Not only w o u l d the procedure be simpler to understand, but fewer errors w o u l d be made i n f o r m u l a t i n g the mathematical theory and i n p e r f o r m i n g the resulting calculations. W i t h these considerations, ideodynamics was developed using q u i t e simple approximations. For many public issues, the population was assumed to f o l l o w blindly the information in the mass media. Interestingly, this simple model d i d give reasonable opinion projections suggesting that the media is not o n l y responsible for setting the agenda (Cook et al., 1983; E r b r i n g , Goldenberg, and M i l l e r , 1980; Funkhouser, 1973a. 1973b; Funkhouser and McCombs, 1972; Iyengar, Peters, and K i n d e r , 1982; M c C o m b s and Shaw, 1972; M a c K u e n , 1 9 8 1 , 1984) but is also the key agent i n determining opinion. 1.2 N A T U R E O F T H E P O P U L A T I O N Since one of the major requirements was empirical testability, ideodynamics was structured so that tests could be applied using readily available data, namely those f r o m public opinion polls. The starting point for any o p i n i o n p o l l is a question relating to a particular issue. In ideodynamics, the issues are defined as they are in o p i n i o n surveys. In particular, issues are topics on w h i c h members o f the populace can each h o l d o n l y one o f t w o or more m u t u a l l y exclusive positions or ideas. For instance, the first issue i n this book concerns American public opinion on funding for military defense. This issue was defined as having only three ideas or positions, favoring m o r e , same, or less spending since these were the positions i n several published polls. Since public opinion polls divide people into subpopulations, each holding a different viewpoint, ideodynamics also divides the population into subpopulations along the same lines. However, the model makes a distinction between individuals unaware o f the issue and persons aware o f the topic and h o l d i n g an o p i n i o n . "Unawares" comprise a portion o f the N o O p i n i o n or Don't K n o w groups i n opinion polls. " A w a r e s " are subdivided into those holding each o f the permitted answers to polled questions. The Don't K n o w s might also include some w h o are aware but are undecided. The defense spending analysis i n this book ignored the N o Opinions, including both the unawares and the awares but undecided, because the N o Opinions were few i n number, usually comprising less than 10 percent o f the total population. That left three subpopulations of awares supporting more, same, or less spending. The differences in treatment between awares and unawares are discussed in Appendix A and later i n this chapter.

Copyrighte<

Formulation

of Ideodynamics

11

1.3 N A T U R E OF P E R S U A S I O N H a v i n g defined issues, positions, and subpopulations as is done for o p i n i o n p o l l s , i t is possible t o turn to the manner b y w h i c h ideodynamics analyzes persuasion and calculates o p i n i o n percentages. For these analyses, ideodynamics notes that persuasion occurs i n t w o steps: (1) message generation by senders, and (2) message impact upon receivers. The vital links between these t w o steps are the messages themselves. One o f the key s i m p l i f y i n g assertions o f ideodynamics is that the messages, when properly coded, can include all the relevant information about the senders needed for the accurate assessment o f message impact. Such coding simplifies the analysis by enabling separate analyses for message impact and message generation. Therefore, i t becomes possible t o study messages and their effects w i t h o u t simultaneously considering how the messages were created. I n the i n t r o d u c t i o n , an analogy was d r a w n between persuasive actions and intercontinental ballistic missiles. T o assess the effects o f missiles, i t is unnecessary to k n o w where they came from so long as certain key features such as trajectory and megatonnage are available. The effects w i l l be as devastating regardless o f whether the missiles were sent by accident or by design. I n the same way that missile effect can be determined w ithout k n o w i n g anything about the missile sender—so long as the relevant traits o f the missile are k n o w n , i t should be possible t o compute the effects o f persuasive messages o n receivers w i t h o u t k n o w i n g anything about the source. I t is o n l y c r u c i a l that the pertinent traits o f the message be coded. Clearly, message broadcast and message impact are not independent events. Some message senders are l i k e l y to be receivers and vice versa. A l s o , message sources are l i k e l y to change their messages after interaction w i t h receivers. For example, Rogers (1983) stresses the importance o f message receivers asking senders for more information. However, any interaction between message sender and receiver w i l l still proceed b y w a y o f messages. Therefore, i t is s t i l l possible to separate the analysis o f persuasion into the t w o distinct portions o f message generation and message e f f e c t so long as the messages themselves can be captured and analyzed. After separating message generation f r o m message effect, i t is still possible to study interactions between message senders and receivers. For example, i f a receiver transmits a question to a sender, then that question, itself, is a message. The person receiving the question can then become a sender and send a message i n response. T o examine the behavior o f this person, i t is sufficient to k n o w that a message was received i n the f o r m o f a question. The impact is to cause the sending o f the answer, another message. Interactions between message generation and impact are discussed further i n the final chapter o f this book. H a v i n g argued that proper coding obviates the need for further information about message senders, i t is necessary to consider i n detail the message coding scheme used i n ideodynamics. This coding must take into account the fact that different messages w i l l have different effects o n different subpopulations. Differences i n effects have been examined by many authors i n considering cognitive dissonance and the " m i n i m a l effects" o f the media. These t w o topics are related t o each other, w i t h c o m m u n i c a t i o n theorists dating back to Lazarsfeld. Berelson, and Gaudet (1944) proposing that the principal effects o f the media are not to convert opinion but to reinforce i t . In other words, people are l i k e l y to suppress dissonant messages w h i l e preferentially selecting information favoring their position to reinforce their current viewpoint. This concept of the " m i n i m a l effects' o f the media still has adherents (Chaffee, 1975; Klapper, 1960; Kraus and Davis, 1976; M c G u i r e , 1986; Rogers, 1983) although this viewpoint is not uncontested (Graber, 1984; Noelle-Neumann, 1984; Page, Shapiro, and Dempsey, 1987; Wagner 1983). A n y model w h i c h can account for opinion reinforcement needs to account for an important l o g i c a l consequence o f the resolution o f dissonance and the resulting r e i n f o r c e m e n t o f previous o p i n i o n : the p o p u l a t i o n m u s t be d i v i d e d i n t o

Copyright

12

Predictions

of Public

Opinion from the Mass

Media

subpopulations holding different positions, just the type o f subdivision identified by polls and used i n ideodynamics. T h a t is because i n f o r m a t i o n favoring a position should o n l y reinforce opinion among people supporting that position. I n the defense spending case, for example, a message supporting more spending s h o u l d o n l y reinforce opinion in the subpopulation already favoring this idea. It should further be noted that the concept o f minimal effects does not mean "no effects" o f the media. Thus the media is permitted to have some-perhaps s m a l l effect in changing minds w i t h i n a population. Similarly, cognitive dissonance is not assumed t o result invariably i n reinforcement. O n rare occasions, dissonance can be resolved i n f a v o r o f the dissonant i n f o r m a t i o n . Consider again the case o f information favoring more defense spending. Although the major effect might be to reinforce those already s u p p o r t i n g this p o s i t i o n , i t is possible that the same information might also have a weak conversion effect and increase the number o f people f a v o r i n g this position. The people converted must have held some other viewpoint earlier, such as that supporting same spending. Even though i t may be small relative to reinforcement, the conversion effect can be very important. I n fact, the essential question f r o m the standpoint o f o p i n i o n is whether reinforcement is so strong that no change occurs at a l l . I n the absence o f change, o p i n i o n w i l l stay static and invariant, a situation w h i c h is k n o w n to be false for a large n u m b e r o f issues. F o r any issues w h e r e o p i n i o n s do change, reinforcement cannot be so o v e r w h e l m i n g as to block all shifts. I f changes can occur, the c r u c i a l element i n determining public o p i n i o n is the residual amount o f persuasive force, however s m a l l , w h i c h can override the reinforcing i n f o r m a t i o n , since that is the effect w h i c h w i l l cause opinion alterations. The critical role o f factors overriding reinforcement is reminiscent o f the work o f Granovetter (1973, 1978, 1980) o n the "strength-of-weak-ties." L o o k i n g at the sources from w h i c h people learned about the jobs they took, Granovetter f o u n d that the most useful sources were frequently those w i t h w h i c h the job-seeker had relatively little interaction. Reinforcement i n the present j o b d i d not l e a d to j o b changes, so the reinforcing interactions were unimportant regardless o f their frequency o r intensity. The key element was i n f o r m a t i o n about new j o b s , even i f that information was rare. I f there was sufficient reinforcement for the original j o b that a person d i d not change employment, then that non-change w o u l d not have been recorded in die Granovetter studies. The finding that weak ties are very important to changes is i n agreement w i t h ideodynamics, w h i c h argues that the mass media may have as its main function the reinforcement o f a person's v i e w p o i n t , but projections o f public o p i n i o n must focus on an analysis o f the few factors w h i c h d o induce change.

1.4 N A T U R E O F P E R S U A S I V E M E S S A G E S The previous section has discussed how the same message can affect different subpopulations differently, w i t h most o f those i n favor o f a message's position being reinforced and a few of those opposed being converted. Messages in ideodynamics are coded to account for both opinion reinforcement and conversion. For s i m p l i c i t y , consider the issue o f A m e r i c a n a i d to the Contra rebels in Nicaragua (1983-1986). For this issue, there were o n l y the t w o positions o f " p r o " (favoring continued aid) and " c o n " (opposing continued aid). I f a message favors only the pro position, that message should reinforce the opinion o f the pros and might convert a few o f the cons. I f a con message arrives at the same time, then this con message should reinforce the cons and might convert some o f the pros. W i t h these t w o messages, both the pro and con positions w o u l d be reinforced. Simultaneously, there could also be opinion conversions i n both directions. Logically, it is plausible that the result should be the same i f the two messages were not i n separate communications but were part o f a single m i x e d message. Returning to the analogy o f messages being ballistic missiles, the model is that a single launched message could split into a number of M u l t i p l e Independent Reentry

Copyrighte U

Formulation

of Ideodynamics

13

Vehicles. A message w i t h two components favoring the pros and cons respectively w o u l d be like a M I R V e d rocket w i t h t w o warheads, one favoring the pro position and the other the con position. The assumption is that the target w o u l d be indifferent to whether the warheads were sent i n one missile or i n t w o separate missiles. The end effect would be as devastating. Therefore, the analysis can be made i n terms o f message subunits f a v o r i n g different positions, each one being analogous to a single Independent Reentry Vehicle and being treatable as a persuasive entity i n its o w n r i g h t For ease o f reference, die w o r d " i n f o n " was defined to refer to the concept o f "a message component favoring one o f the possible positions being considered." The first part o f this term is f r o m the w o r d information and refers to persuasive messages. The ending o f i n f o n is the same as that o f terms for elemental entities such as electrons and introns i n the physical and biological sciences. In fact, ideodynamics postulates that infons are also elemental i n that the entire persuasive power o f a message can be coded i n the properties o f its infons. Infons are categorized in four dimensions: 1. The first dimension is the position favored by the i n f o n . Since each infon favors a specific position, an infon is o n l y defined once Ihe positions under consideration have been specified. I n fact, the same message can have infons defined differently for different issues. I n a message discussing both defense spending and Contra aid, for example, the infons for the defense spending analysis w o u l d be defined i n terms o f favoring more, same, or less spending. The same message used in a Contra aid study w o u l d be coded as having different infons, this time either favoring or opposing aid. 2. The second dimension o f the infon refers to whether the infon directly or indirectly supports its position. This distinction is useful because persons aware o f the issue and its associated arguments can d r a w an inference from indirect data while an unaware individual can o n l y be persuaded about an issue i f there is a direct statement i n the message about the issue. For instance, someone unaware that defense and domestic programs competed for the same funds c o u l d not connect a deficiency i n a domestic program w i t h defense spending while someone aware of the association might. 3. The third dimension refers to the sender or source o f the i n f o n . U s i n g this dimension, information favoring more defense spending f r o m t w o different sources-such as the President o f the U n i t e d States and Congress—could be assigned to t w o different infons. T h i s d i s t i n c t i o n is made i n case some sources have more persuasive powers than others, as has been studied extensively by Page, Shapiro, and Dempsey (1987). 4 . The fourth dimension gives the index number o f the message containing the infon. Therefore, all infons f r o m the message labeled as message 1 w o u l d carry the index number o f 1. As just discussed, the infon w o u l d be further identified by its position (first dimension), directness (second dimension), and source (third dimension). I n summary, although several infons w i t h i n a message can favor the same position, each individual infon can o n l y support one position. Whenever a message supports more than one position, that message must be d i v i d e d into infons, at least one for each o f the positions favored. Every infon is identified by indices reflecting the four dimensions o f position favored, directness, source, and the index number of the message containing the i n f o n . A n example w o u l d be an i n f o n supporting more defense spending (first dimension) resulting f r o m a direct statement (second dimension) by the President o f the U n i t e d States ( t h i r d dimension) i n a message indexed by the investigator as message n u m b e r 1 ( f o u r t h d i m e n s i o n ) . The purpose o f specifying the f o u r dimensions is to permit infons to be grouped for further analysis. A l l groupings are based o n these dimensions. After a message is subdivided into its infons for the topic under study, each infon is then assigned three properties independent o f each other. These properties are assumed to be sufficient to explain the infon's persuasive effects i n the same way that

Copyrighte<

14

Predictions

of Public

Opinion from

the Mass

Media

trajectory and megatonnage w i l l yield a description o f the damage f r o m nuclear warheads: 1. The content o f an i n f o n is a numerical score describing the a b i l i t y o f the contents o f the i n f o n to persuade appropriate subpopulations. I n this book, the measured infons all came f r o m A P dispatches, so i n f o n content was measured as the number o f t y p i c a l A P paragraphs f a v o r i n g the infon's viewpoint. 2. The validity o f an infon is attributed to the reputation o f the m e d i u m o f the message. L i k e i n f o n content, i n f o n v a l i d i t y is scored n u m e r i c a l l y . T y p i c a l l y , a receiver is unable or does not have the time to check the reliability o f the m e d i u m , so the receiver depends on its general reputation. For instance, the validity w i l l be much higher for a trusted friend than for a total stranger. Since this book's infons came f r o m AP dispatches, all validities were assigned a value characteristic o f the A P , w h i c h was the medium. 3. The audience size o f an i n f o n is a mathematical function describing the numbers o f people exposed to the i n f o n as t i m e proceeds. Since i t is a mathematical function, the audience size is unlike the infon's content and validity scores, w hich are f i x e d numbers. Obviously, the audience size is zero before the infon is emitted. Also, the audience size w i l l be much larger for an infon from the mass media than one f r o m a personal experience where there is o n l y one receiver. The audience size w i l l have a very short duration for a one-on-one conversation, returning to zero as soon as the discussion terminates. The equivalent duration for a book can be quite long since the book may continue to be sold and read for months. W i t h this brief overview, the individual properties o f infons can be considered i n greater detail. The i n f o n content score combines aspects o f both salience and directionality as used by previous authors for coding persuasive messages. T o compare i n f o n content scores w i t h schemes used by others, consider, for instance, t w o messages, A and B, both p r o v i d i n g 100 percent support for the position of more defense spending. Suppose that message A is more persuasive because more is said, because what is said is more effective, or because the quoted source is more credible. The effectiveness can be due to cognitive and'or affective appeals (Abelson et al., 1982; Conover and Feldman, 1986; Marcus, 1986; Rosenberg, M c C a f f e r t y , and Harris, 1986). One type o f score used for persuasive messages has been directionality. For instance, Page and Shapiro (1983a) coded a number o f news stories i n the New York Times and o n television evening news on a five-point scale f r o m "clearly pro," to "probably pro," "uncertain or neutral," "probably con," and "clearly c o n . " W i t h this method, both stories A and B favoring more defense spending w o u l d be given a score o f clearly pro. Story A, being more persuasive, w o u l d either have a higher salience or a greater "quality," w h i c h has been defined by Page and Shapiro (1983a) to include the "logic, factuality, and degree o f truth or falsehood." Therefore, messages A and B w o u l d be characterized by t w o different quality and salience scores and a common directionality score. The scoring i n ideodynamics is somewhat different. The first step is to define the positions relevant to the polled question. A n y number o f positions is permitted. Then, for each persuasive message, ideodynamics assigns one or more infons to each position, w i t h different infons having different sources and directness o f appeal. Each o f these infons w i l l then have a characteristic content, v a l i d i t y , and audience size. I f the message has no component supporting a particular position, then the content scores o f the infons favoring that position are zero. Therefore, besides incorporating the trait o f directionality, the content score also incorporates the salience and quality values because the higher the salience and/or quality, the higher w i l l be the content score. Obviously, salience is not o n l y included i n the content o f the message but is also related to the message's audience size, as w i l l be discussed i n the next section.

Copyrighted material

Formulation

of Ideodynamics

15

The i n f o n content score, therefore, incorporates portions o f the concepts o f directionality, salience, and quality. Returning to the example o f defense spending, i t is possible t o specify a very simple structure where infons are defined to favor o n l y the t w o positions o f more spending o r less spending. Also, i t is possible t o consider o n l y direct infons and to make n o distinctions based o n i n f o n source. I n this case, infons w i l l o n l y be distinguished by the index number o f the corresponding message and the positions favored. W h e n this simple structure is applied to the t w o hypothetical messages A and B , both f a v o r i n g o n l y more defense spending, both messages w o u l d have a content score o f zero f o r the infons favoring less spending. Since both messages support more spending, the content scores o f their infons f a v o r i n g this position w o u l d both be positive, w i t h the more persuasive message (message A ) having the higher score. A n important advantage o f the ideodynamic coding o f messages is that infons can easily code quite complicated issues where there are many positions. This flexibility was demonstrated i n the example o f the Democratic primary o f 1983-1984. For this issue, persuasive messages were divided into six different infons favoring the positions o f advantageous f o r John G l e n n , advantageous f o r W a l t e r M o n d a l e , advantageous for Others, disadvantageous for Glenn, disadvantageous for Mondale, and disadvantageous for Others. I t w o u l d obviously have been more difficult t o use a single pro-con scale to distinguish six positions. Since the content score describes the a b i l i t y o f the i n f o n to persuade the audience, this score depends on the interpretation by the message receivers rather than by the sender. Since social changes result f r o m changes among the receivers, i t is their perceptions w h i c h are o f greatest importance. The decision to code directionality o f mass media messages f r o m the receivers' point o f v i e w was also taken by Page and Shapiro ( 1 9 8 4 ) , a l t h o u g h these investigators have subsequently coded their directionality in terms o f the intent o f the message source (Page, Shapiro, and Dempsey, 1987). T h e change was not made for theoretical reasons but rather for ease o f scoring. However, these authors noted that there was generally good agreement between the t w o approaches for information scoring i n die mass media. The coded material was either text from New York Times articles or summaries f r o m the Television News Index and A r c h i v e s f r o m the Vanderbilt Television News Archives. I t is conceivable that conflicts between a sender's intention and a receiver's perception m i g h t have been more pronounced i f there had been inclusion o f non-verbal messages such as those transmitted v i a a television screen. Validity was the second property assigned to infons and refers to the reputation o f the m e d i u m carrying the message containing the infon. Introduction o f the v a l i d i t y score recognizes that the audience makes t w o c r e d i b i l i t y decisions about information attributed t o a quoted source. First, the audience must decide that the quoted source actually said what was reported. This decision is given in the v a l i d i t y score reflecting the reputation o f the m e d i u m However, the audience must also evaluate whether the quoted source is trustworthy. This credibility is included i n the infon content score. Infon validity as used in ideodynamics has not always been included explicitly by other investigators i n assessing the effects o f the mass media. For example, Page and Shapiro (1984) and Page, Shapiro, and Dempsey (1985, 1987) assumed that information i n the New York Times and over network television has a very high validity i n that the public w i l l assume that the president actually made a statement i f he is quoted as having done so. Their analyses were performed entirely i n terms o f the c r e d i b i l i t y o f the sources quoted by the m e d i u m . This high reputation o f the m e d i u m is clearly reasonable for much o f the mass media i n the major Western democracies. However, i n the more general case, the m e d i u m can contribute importantly to the believability o f information since sources l i k e trusted friends, respected news agencies, and personal experiences w i l l have high validities w h i l e sources l i k e suspected pathological liars and untrustworthy scandal sheets w i l l have low validities.

Copyrighte

16

Predictions

of Public

Opinion from

the Mass

Media

The importance o f the m e d i u m has been recognized f o r many years. F o r instance, H o v l a n d (1959) proposed that the impact o f a message was greater i n an experimental setting, when the subject thought the medium o f the message had the sanction o f the investigator (see also Eagly and Himmelfarb, 1978). The third infon characteristic is audience size. This property is simply the curve describing the number o f people exposed to the message containing the infon as time proceeds. Messages like A P dispatches have large audience sizes just after emission w i t h the audience decreasing gradually thereafter. F o r the mass media, i t is convenient t o introduce four more terms: the time at which the message is broadcast; a persistence constant describing the rate at w h i c h the message becomes inaccessible to the population; a memory constant describing the rate at w h i c h people forget about the infons i n the message; and the audience size at the broadcast time. These four terms can describe the audience size at all times for A P messages (see Appendix L i k e the content score, the audience size also incorporates aspects o f the salience o f persuasive messages. The ideodynamic audience size increases when more people are exposed due to messages having higher salience.

1.5 R E L A T I O N S H I P S B E T W E E N I D E O D Y N A M I C S T R U C T U R E S Infons are the last o f the important structures w i t h i n w h i c h ideodynamics organizes data. A t this point, i t is useful t o consider the relationships between the basic ideodynamic structures describing issues, messages, and the population. A t the center is the issue, w h i c h is d i v i d e d i n t o any number o f m u t u a l l y exclusive positions. T h e generality o f the model is reflected i n the lack o f limitation on position numbers. Both the population and messages are organized according to the positions o f the issue. The population is divided into subpopulations f o l l o w i n g members' responses to opinion polls. Each subpopulation favors a unique position o f the issue. A s noted already and as will be discussed below, this definition of subpopulations permits an

explicit mathematical modeling o f the resolution o f cognitive dissonance i n favor o f opinion reinforcement. Section 1.4 described how all messages are divided into infons, w i t h each infon able to favor o n l y one position. A l t h o u g h there is frequently great overlap, the positions o f infons need not coincide absolutely w i t h the positions corresponding to the subpopulations. For example, i n the example o f the Democratic primary, there were infons disadvantageous to Mondale. There was no corresponding p o l l o r subpopulation position corresponding to opposition to Mondale. O n the other hand, it is possible to perform an analysis w i t h infons scored as favoring only more or less defense spending, w i t h no infons supporting same spending, even though there is a subpopulation favoring same spending.

1.6 O V E R V I E W O F O P I N I O N C A L C U L A T I O N S Ideodynamics examines opinion formation using issues w i t h mutually exclusive positions, and using subpopulations and infons each favoring a single position. The strategy is to analyze opinion structure through an examination o f o p i n i o n change. The argument is that o p i n i o n can be calculated at any later time i f o p i n i o n is available at an earlier time and i f all intervening opinion changes are k n o w n . In the same way, the location o f an automobile is defined i f an earlier position is specified and i f the entire subsequent pathway is given. This emphasis on analysis o f change is also at the heart o f e p i d e m i o l o g i c a l ^ based models predicting logistic increases for the adoption o f innovations like new technologies (Bartholomew, 1976; H a m b l i n , Jacobsen, and M i l l e r , 1973). M o d e l i n g through change permits public opinion to be dependent on opinion at an earlier time. T h e dependence on past o p i n i o n is one important means f o r a

C

Formulation

of Ideodynamics

17

population to reflect its history. I f there was much successful persuasion i n the past so that o p i n i o n f a v o r i n g a p o s i t i o n became h i g h , then o p i n i o n w o u l d s t i l l be favorable i f there very l i t t l e i n the In i f this favorable persuasion had not occurred, then there might be few p e o p l e h o l d i n g the favorable v i e w p o i n t at the earlier t i m e . As a result, there w o u l d s t i l l be very few people in favor i f there was not much intervening opinion change. I n focusing on opinion alterations, ideodynamics assumes that infons favoring a p o s i t i o n w i l l cause members o f appropriate subpopulations h o l d i n g different v i e w p o i n t s to change their minds. The power o f infons to affect o p i n i o n are described mathematically using persuasive force functions" (see Appendix A ) . One i n f o n can alter the persuasive power o f another due t o phenomena l i k e o p i n i o n reinforcement and message saturation. Reinforcement means that an opposing infon w o u l d be w e a k e n e d and less able t o cause o p i n i o n conversions. Therefore, ideodynamics models opinion reinforcement by permitting infons favoring a position to decrease the persuasive force functions for opposing infons. Infons repeated too shrilly and frequently can lose their effectiveness. Ideodynamics also permits this excessive propaganda to have diminishing returns. Once the persuasive force functions have been formulated, i t is necessary to examine the expected effects o f all the different infons acting on the population. Considering again the analogy w i t h ballistic missiles, the infons o f a message are l i k e the component warheads o f a single M I R V e d rocket i n that a l l these infons are launched together. T h e n the i n d i v i d u a l infons h i t d i f f e r e n t target subpopulations w i t h different effects. For example, infons favoring more defense spending are assumed i n Chapter 5 to convert members supporting the same spending position to favor more spending However, i t is not necessary for an i n f o n favoring a particular position to recruit persons o n l y to that position. A g a i n , i n the defense spending case, infons favoring more spending were also assumed to be able to cause those favoring less spending to move halfway and support same spending. The specifications o f the appropriate target subpopulations and the resulting conversions for all possible sets o f infons are given i n "population conversion models," w h i c h w i l l vary from issue to issue. Ideodynamics assumes that the larger a susceptible target subpopulation is, the greater w i l l be the conversion. This statement is equivalent to saying that more deaths w i l l result f r o m a nuclear missile landing i n a densely populated area. I n addition, opinion conversion is increased i f the persuasive force function is higher, in the same way that more deaths w i l l also occur i f the megatonnage o f a warhead increases. These arguments are both incorporated i n t o the basic ideodynamic equations for o p i n i o n change. Since change is proportional to both the size o f the target subpopulation and the strength o f the persuasive messages, the terms i n the ideodynamic equations are non-linear and t h e r e f o r e differ significantly f r o m linear models such as many i n econometrics.

was

change

meantime.

contrast,

would

1.7 D E T A I L S O F O P I N I O N C A L C U L A T I O N S FOR T H E A W A R E S F o l l o w i n g the principles just enunciated, the actual projection o f public opinion proceeds i n three main steps: 1. The first is to construct mathematical functions describing persuasive forces due to infons. 2. The second is to develop a "population conversion" model for each issue studied. As mentioned above, infons favoring a position are presumed to act on appropriate target subpopulations to convert a portion o f the members to j o i n other subgroups. By studying ensuing changes, there is no need to be concerned w i t h the reasons for the initial state reflecting the previous history o f the population. Therefore, any opinion p o l l can be taken as the starting time for the opinion projections. F r o m that time forward, an analysis o f intermediate attitudinal changes w i l l be sufficient for calculating opinion at any later time.

Copyrightec

78

Predictions

3.

of Public Opinion from the Mass

Media

The third step is to calculate public opinion using the infon persuasive force functions and the ideodynamic equations for opinion change corresponding to the population conversion models. Each o f these three steps is discussed i n greater detail below, focusing o n opinion derived f r o m A P stories since those are the examples i n this book. This discussion begins by e x a m i n i n g a population made u p entirely o f awares h o l d i n g different positions. The modifications for the unawares are considered i n the f o l l o w i n g section. As noted i n the preceding section, the awares are able to digest i n d i r e c t information as w e l l as direct information, so persuasive force functions for the awares include both classes o f infons. Each infon's persuasive p o w e r is encoded as a persuasive force function w h i c h changes w i t h time (Appendix A , Equation A.7). As justified i n Appendix A, the persuasive force function for an individual infon can be approximated by the product o f the infon's content score, validity score, and audience size function. That is, the persuasiveness o f an i n f o n increases whenever there is a larger content score i n favor o f a position, the medium has a higher reputation, or the audience size is increased. W h i l e each infon has its o w n characteristic content score, most A P infons for any one issue w i l l have approximately the same reputation d u r i n g the entire t i m e periods studied, so the same constant validity score can be assigned to all A P infons. Besides the content and validity scores, the persuasive force function is also dependent o n the audience size. Since this function varies w i t h t i m e , i t is the audience size function w h i c h w i l l govern the time-dependent shape o f the persuasive force function. As far as audience size is concerned, i t is assumed that A P infons have their maximal effect on the day o f their transmission w i t h that effect d w i n d l i n g e x p o n e n t i a l l y w i t h t i m e . The exponential drop seems plausible since p u b l i c exposure to i n f o r m a t i o n i n p r i n t media is l i k e l y to decrease c o n t i n u o u s l y , but rapidly, w i t h no sharp cut-off. The rate o f decrease is characterized by a "persistence" constant w h i c h was optimized (Chapter 5) to have a one day half-life suitable for all issues studied. W i t h a one day half-life, the persuasive effect o f a message drops to one-half after one day, one-quarter after two days, one-eighth after three days, and so o n . A l t h o u g h newspapers published A P stories a day o r so after the dates o f the dispatches, the dispatch dates themselves were used for the calculations i n this book since it seemed l i k e l y that radio and television messages w i t h the same approximate content w o u l d have appeared on the dispatch dates. As described i n Chapter 4, A P messages were scored for the numbers o f paragraphs supporting each o f the positions w i t h i n the issue being analyzed Since the scores were i n numbers o f paragraphs, the heights o f the persuasive force plots for A P infons are given i n paragraphs. Figure 1.1 shows the persuasive forces for t w o separate infons (top t w o frames) favoring the same position. These functions describe the strengths o f the infons w i t h respect to their a b i l i t y to cause o p i n i o n change. The simplest model for the combined effect o f t w o infons w o u l d involve adding the persuasive force functions o f the i n d i v i d u a l infons (Figure 1.1, bottom frame). This addition requires the assumption that the t w o infons behave as independent units. The j u s t i f i c a t i o n for this a p p r o x i m a t i o n is based o n the second law o f thermodynamics, which can be restated as the " l a w o f the 24-hour day." T i m e does not go backwards, so every i n d i v i d u a l , regardless o f intellect, capability, or interest, only has time to consider t h o u g h t f u l l y a very small number o f issues. Due to this time constraint, knowledge must be superficial f o r the vast majority o f issues about w h i c h a person has opinions. Therefore, o n l y a very small percentage o f the public w i l l be experts able to f o r m carefully considered opinions for any g i v e n issue; each issue w i l l have its o w n collection o f experts, w i t h those experts changing f r o m issue to issue. Downs (1957) has also noted that the bulk o f the population does not devote much time or effort to careful analysis o f most issues. For instance, most people have neither the t i m e nor resources to consider seriously all the factors relevant to the question o f whether defense spending should

Formulation

of Ideodynamics

19

be increased o r decreased. Nevertheless, 90 percent or more o f the population (Appendix B ) usually had opinions on this topic. W i t h o u t the time or background to study the problem i n detail and hence to relate different of with each other, the majority o f the public is likely to treat all new infons as i n d e p e n d e n t pieces o f i n f o r m a t i o n , and may even treat i n d i v i d u a l phrases w i t h i n persuasive messages as independent infons (Chapter 4). The independence o f persuasive force functions also bears on the modeling o f opinion reinforcement. That is because the impact o f infons is described by these functions. Therefore, i f reinforcement occurs, the persuasive effects o f opposing infons w i l l be decreased, resulting i n a persuasive force function w i t h a smaller value. There are t w o basic m e t h o d s for including reinforcement i n the model. The most direct is simply to reinterpret a message as having a conversion infon w i t h a lower persuasiveness i f reinforcing information is present. This decreased persuasiveness means a lower persuasive force value at the measurement time and could arise from the content o f the conversion i n f o n having a lower content score or a smaller audience size as people avoid unfavorable information. The infon content scores i n this book were obtained by computer to assure consistency, but this strategy w o u l d have been difficult to apply i f the interpretation o f all infons depended o n the context of other persuasive information w i t h i n both the same and other messages received by the public. However, i t is also possible to model o p i n i o n reinforcement mathematically based on infons scored as i f there were no other infons around. In such a procedure, small message units such as A P paragraphs w o u l d be scored for (he positions they supported w i t h o u t regard to paragraphs i n either the same A P story or i n other stories. Then, the a b i l i t y o f a reinforcing i n f o n to d i m i n i s h a conversion i n f o n w o u l d be m o d e l e d by mathematically decreasing the persuasive force function for the conversion infon i f a reinforcing infon had occurred earlier (Appendix A, Equations A. 12 and A . 13). F o r these equations, an i n f o n w h i c h is more effective at making opinion conversions also has a greater ability to reinforce. This mathematical approach has the adv antage that the corresponding equations contain constant parameters d e s c r i b i n g the importance o f reinforcement Fortunately, satisfactory o p i n i o n calculations were obtained when these parameters were set to zero, w h i c h is consistent w i t h the law o f the 24-hour day and the idea that most people f o r most issues are sufficiently distracted that they do not spend the time to associate one piece o f i n f o r m a t i o n w i t h another. As a result, i t was possible to ignore reinforcement i n computations o f expected public opinion. The phenomenon o f o p i n i o n reinforcement is but one example o f infons interacting w i t h each other. Although a reinforcing infon can theoretically decrease the activity o f a conversion i n f o n , i t is also possible that continued rapid repetition o f conversion infons can also lead to information saturation, w i t h diminishing effects for additional infons. This is included i n A.13 (Appendix A ) . As w i t h reinforcement, message saturation c o u l d also be ignored, w i t h their corresponding constant parameters being set to zero i n empirical tests o f the model. Since a d d i t i v i t y o f i n f o n persuasive force functions gives acceptable opinion calculations, i t is possible t o ignore all infon interactions, even those not associated w i t h o p i n i o n reinforcement and information saturation. The result o f additivity is that the combined persuasive power o f the t w o infons (Figure 1.1, bottom frame) is the sum o f the individual effects. I n other words, the residual o f the first infon are added to the effects o f the second. T h i s same strategy o f i g n o r i n g message interaction was also adopted, for instance, by Page, Shapiro, and Dempsey (1987), w ho scored the directionality o f television news f r o m the viewpoint o f an "intelligent, attentive audience w i t h average A m e r i c a n beliefs and values." These investigators made no effort to relate the contents o f one message to another, either i n the scoring o r i n the subsequent calculations. Once the persuasive force curves are d r a w n , the next step is to construct a population conversion model giving the likely effects o f the various persuasive forces o n each o f the subpopulations. For c l a r i t y , i t m i g h t be useful t o consider the

pieces

possibility

information

Equation

effects

Copyrighte

20

Predictions

of Public Opinion from the Mass

Media

example o f defense spending w i t h the population conversion model represented i n Figure 1.2. I n this Figure, the names for all the subpopulations begin w i t h " B " (for believers" i n a viewpoint) and the names for all persuasive force curves begin w i t h " G " i n reference to the G " functions used for the calculations (Appendix A, l i q u a t i o n A.29). Therefore, the subgroups were B M o r e , BSame, and BLess and the persuasive force functions were G M o r e , GSame, and GLess favoring, respectively, more, same, and less spending. Each action leading to an o p i n i o n change is represented by an arrow. The name(s) besides the arrows refer to individual persuasive forces. The tail o f the arrow leads f r o m the target population whose size is decreased b y the persuasion. The arrow head points to the destination population whose number is increased i n the same process. I n Figure 1.2, infons favoring more spending are assumed t o be able to persuade those favoring less spending to alter their o p i n i o n to support same spending. The same infons can also persuade those already favoring same spending to support more spending. I n this m o d e l , there is "sequential c o n v e r s i o n " i n that those f a v o r i n g less spending w i l l first favor same spending before favoring more spending. This model also takes no account o f the length o f time a person stays i n the sume spending g r o u p en route f r o m f a v o r i n g less t o more spending. The t i m e can be almost instantaneous for some people and much longer for others. The basic point is that more information is needed to move f r o m less to more spending than f r o m less to same. The same infon persuasive force function can act on different subpopulations. For instance, i n Figure 1.2, infons favoring more and same defense spending are both assumed to persuade those preferring same spending to favor more spending and to convince those holding the less-spending position to prefer same spending. A t the same time, different types o f infons can also act on the same target subgroups to cause the same conversions. I n general, the population conversion model provides the f u l l description o f the target populations w h i c h lose members, the persuasive forces functions causing the conversions, and the destination populations whose numbers are increased by the opinion changes. The details o f the population conversion models w i l l depend on the total number o f p o l l e d positions, the types o f infons, and the changes w h i c h are l i k e l y to occur. In the case of the Democratic p r i m a r y , the most complex example i n this book (Chapter 5), there were four polled positions (pro-Mondale, pro-Glenn, pro-Others, and N o O p i n i o n ) and six types o f infons (pro-Mondale, pro-Glenn, pro-Others, conMondale, con-Glenn, and con-Others). Once the infons have been scored, the results used to construct persuasive force functions, and the p o p u l a t i o n conversion models have been f o r m u l a t e d , p u b l i c opinion is calculated using ideodynamic equations. These equations, applied to mass media infons, are based on the f o l l o w i n g principles: 1. The number o f people converted is proportional to the persuasive force functions. 2. The number o f people converted is also proportional t o the number o f people i n the target population. I n the extreme case o f everyone already holding the o p i n i o n o f the infon, there w i l l be no nonbelievers and hence a nonexistent target population, so there w i l l no be converts regardless o f the strength o f the persuasive infons. 3. The constant o f proportionality is the "persuasibility" constant ( A p p e n d i x A, Equation A . 14). This constant reflects the fact that some attitudes are closer than others to the core beliefs o f an i n d i v i d u a l . V e r y f i r m l y held attitudes w i l l be m u c h more d i f f i c u l t to alter. T h u s , the persuasibility constant for defense spending w i l l be larger than that for religious beliefs, w h i c h are probably m u c h more refractory to change. The d i f f i c u l t y o f changing religious beliefs is seen i n the tight correlation between religious beliefs across generations (Cavalli-Sforza et al., 1982). The persuasibility constant is related to the v o l a t i l i t y o f opinion and the malleability o f the population under the influence o f new information. Both

Copyrighte<

Formulation

of Ideodynamics

21

opinion volatility and population malleability are large for issues w i t h high persuasibility constants. Based on the studies i n this book, there can be 4.

wide differences in the persuasibility constant depending on the issue.

I n a d d i t i o n , ideodynamics also includes " r e f i n i n g w e i g h t " constants to all o p i n i o n conversions f r o m one target subpopulation to another proceed b y exactly the same p e r s u a s i b i l i t y constant. I n effect, the persuasibility constant can be considered to be the coarse adjustment and the refining weights to be the fine adjustments, since the persuasibility constants in Chapter S c o u l d differ by a factor o f fifty from issue to issue w h i l e the refining weights varied less than threefold for any one o f the six issues tested (Table 5.2). As discussed i n Appendix A, the refining weights also include adjustments to account for variations i n message scoring. These principles lead to Equation A.26 (Appendix A ) , w h i c h is adapted to the individual population conversion models for each polled issue. The resulting set o f projection equations is then used to simulate o p i n i o n behavior using the o p i n i o n percentages o f the first poll i n a time series as the starting point. For these computations the persuasibility constant is replaced by a " m o d i f i e d persuasibility" constant (Appendix A, Equation A.25), w h i c h also includes other constants applic able to the entire calculation, such as the v a l i d i t y scores, w h i c h are assumed to be the same for all infons since the reputation o f the A P is l i k e l y to have been the same for all stories. T o illustrate the opinion calculation, consider an example where the population is equally d i v i d e d among the three positions for defense spending (Figure 1.3). Assume that no relevant infons arrive at the population f r o m time -1 to 0 weeks (top frame). The result w i l l be that no persuasive force function w i l l have values greater than zero during this period. Ideodynamics asserts that no opinion change w i l l occur i n the absence o f information to cause a change. Therefore, opinion stays unchanged d u r i n g this time period (three bottom frames). Assume, further, that a relevant A P message arrives at time 0 weeks. Suppose that this message only contains an infon w i t h a content o f one average A P paragraph favoring more defense spending and none favoring same or less spending. I n this case, the infon persuasive force function (Figure 1.3, top frame) w o u l d be the same as i n Figure 1.1 (top frame). F o l l o w i n g the population conversion model o f Figure 1.2, some people favoring same spending w i l l be convinced by this persuasive force to favor more spending. Therefore, there is a drop i n o p i n i o n favoring same spending and an increase i n opinion favoring more (Figure 1.3). However, this infon also persuades a fraction o f the people favoring less spending to favor same spending. I n consequence, there is a d r o p i n opinion favoring less spending, and the influx o f these people into the samespending p o o l partly compensates for those leaving to enter the group favoring more spending. The net result is a change, but a relatively small one for those favoring same spending. The shifts i n the g r o u p s at the extremes are larger because there is no cross-movement o f people both into and out o f these groups. Since an infon's power is dissipated w i t h i n approximately one week (Figure 1.3, top frame), all o p i n i o n alterations occur w i t h i n that time, after w h i c h there is n o further change since no new infons arrive during the time plotted. T h e magnitudes o f the o p i n i o n shifts are p r o p o r t i o n a l to the m o d i f i e d persuasibility constant, set here at the unrealistically large value o f 2000 per A P paragraph per day to demonstrate the shapes o f the curves. A l l refining weights are assumed to be the same and are set to 1.0. Also, the time between calculations is the very short interval o f 2.4 hours to illustrate the fine structure o f the curve. I n summary, the general rule i n ideodynamic calculations is that all o p i n i o n conversions increase w i t h the sizes o f the target subpopulations and the magnitudes o f the persuasive force functions, w h i c h depend i n turn on the infons' contents, validities, and audience sizes. The durations o f the audience size functions for A P infons are governed by the persistence constant

account for the fact that not

Copyrighte U

22

Predictions

of Public

Opinion from the Mass

Media

As a result, opinion calculations only depend on a persistence constant w h i c h is c o m m o n for a l l issues, a modified persuasibility constant different f o r every issue, and refining weights w h i c h usually have the value o f 1.0. I n the simplest cases, like that f o r defense spending, the m o d i f i e d persuasibility constant was the o n l y independent constant w h i c h needed to be fixed since the initial default assignment o f 1.0 ror all refining weights was satisfactory. It is the empirical testability o f ideodynamics and its success i n six out o f six cases w h i c h permits the suggestion that this parsimonious model is both robust and highly predictive. I n ideodynamics, persuasive messages can both reinforce opinion and lose their effect due to over-repetition as discussed above. However, i n initial computations, i t was assumed that neither reinforcement nor saturation was i m p o r t a n t Since these m i n i m a l approximations gave good opinion projections, they are the ones used for the results in the f o l l o w i n g chapters.

1.8 D E T A I L S O F O P I N I O N C A L C U L A T I O N S FOR T H E U N A W A R E S

case

The previous section considered the special where the entire population consisted o f awares. I f unawares are also added, then the equations w i l l change somewhat because the unawares cannot remember information w h i c h d i d not lead to awareness o f the issue. I n addition, o n l y direct information can act on the unawares as noted i n Section 1.4, so indirect infons are not included i n persuasive force functions for the unawares. The details o f the mathematics for the unawares are presented i n Appendix A . U p o n acquiring awareness, the unawares can adopt one o f the positions o f the awares o r become aware and undecided, w i t h the choice of these t w o possibilities depending on the details o f the population conversion model for the unawares. Rogers (1983), for example, prefers the movement o f the unawares first i n t o the group o f aware but undecided before subsequendy adopting a position. However, the direct adoption o f a position might also occur. A s w i t h o p i n i o n conversion, the acquisition o f awareness is assumed to be proportional to the relevant persuasive force functions and the target population size. However, the constant o f proportionality w i l l be different. This constant is referred to as the "attentiveness" constant (Appendix A ) and recognizes that the population may be more or less attentive to the issue being considered. Besides acquiring awareness, the population can also forget about an issue, so the ideodynamic equations also include a loss of awareness due t o forgetting (Appendix A ) . For the cases i n this book, the entire population could be considered to be aware o f the issue since the number o f Don't K n o w s i n the poll results was usually less than 10 percent. Therefore, there was n o need to model the conversion o f the unawares to awareness.

1.9 T I M E S C A L E O F I D E O D Y N A M I C

ANALYSES

T h e equations o f ideodynamics discussed so far have a l l assumed that membership i n the population does not change during the entire time span o f the analysis. Thus there is assumed to be no birth or death and no migration either into or out o f the population. These approximations were satisfactory for time intervals up to the nine years studied for defense spending. As times increase to generations and centuries, modifications w i l l be needed i n the model. A t a m i n i m u m i t w i l l be necessary to account for birth and death. Death can easily be modeled by the loss o f awares using terms similar to those describing forgetting. B i r t h can be modeled by the introduction o f new unawares into the population as described i n the previous section. I n accounting for birth and death.

Copyrights U

Formulation

of I de adynamics

23

corrections w i l l also be needed i n the ideodynamic equations i f the p o p u l a t i o n changes size.

Besides birth and death, long-term opinion calculations over generations will

also need to account for the fact that the constants i n the ideodynamic equations might change s l o w l y w i t h time. For instance, public belief i n press trustworthiness might diminish i f evidence of unreliability is presented. I n practice, however, the public is concerned w i t h a large number o f issues only w i t h i n time spans substantially shorter than a generation. For example, the issue o f the Democratic primary o f 1984 was not important for more than a few months. Therefore, this book has not considered opinion changes over very long time spans. Although the emphasis in this chapter has been on public o p i n i o n , ideodynamics can also be extended to other social traits such as habits and addiction (Fan, 1985b).

Copyrighte<

The p o p u l a t i o n was assumed to be exposed to t w o A P dispatches at the beginning (top frame) and t w o days after the beginning (center frame) of week 0. The information in each dispatch is separated into infons favoring different polled positions. T h i s figure plots the infons from the t w o dispatches favoring one o f the several possible positions. The top t w o frames g i v e the effects o f the t w o infons separately w i t h a one day persistence half-life. The combined effect (bottom frame) is the sum o f the individual forces. The units l o r the i n f o n force curves are average A P paragraphs. I n this example, both the first and second infons had values o f one paragraph o n their emission date. Figure

1.1. Example

F i r a t

Infon

of persuasive

(AP

forces

of infons.

Paras)

1.2 1.0 0.8 0.6 0.4 0.2 0.0 Second

Both

Infon

Infons

(AP

(AP

Para

sj

Paras)

1.2 1.0 0.8 0.6 0.4 0.2 0.0

-

1

0

1

2

3

Weeks 24

CopyriohlGti mstcris

J)

1 sc •s

CT

on

U

* o £> Sr

«i

o

C X>

ex

il

w

•H

s. O >

> 1>

Ci)

C

g-i "Si C «

"H-

Ë •S -

•W-

iy

-

C

«

3J2

S ¿ ° c c 2 Oí •5 |g«2-fi o *s O. Cl a

M•W-

•gg.

"

a

S* o

3

2 S «

y.

C

»

o g o o

3

C ^ I

4)

f •H-

+*

•H-H•W-

•*

•H

c

ISI

•S « M *t (f -M•H-M-

C

3

2.- S a a 3

Si

U p

•M-

"

*•

I I IQ A

25

c .2 « s; E "S

(A

ï Cl

** *•

J¡

U . •Os

8ı_

i>

v:

E*¿ È c H

Figure 13. Illustration of the impact of a single persuasive infon favoring more defense spending. A single infon w i t h a m a x i m u m value o f one A P paragraph was

assumed to arrive at the population at week 0. The infon's persuasive force (top frame) has the same shape as those i n Figure 1.1. The population at - 1 weeks is presumed to be evenly distributed among those favoring more, same, and less defense spending. The effects o f the infon i n the top frame on public o p i n i o n are drawn i n the three lower frames using the population conversion model o f Figure 1.2 and assuming a modified persuasibility constant o f 2000 per A P paragraph per day. 1nfon

Favor

-

More

1

Weeks

0

(AP

Paraa)

1

2

3

2

Ideodynamics and Previous Models

The previous chapter examined the formulation o f ideodynamics. The present chapter considers the principal features o f ideodynamics i n the context o f other models for social change. As noted i n the introduction, and as w i l l be discussed again at the end o f this chapter, an important g o a l o f ideodynamics is to provide a framework encompassing seemingly disparate conclusions. Throughout (his discussion, special attention w i l l be paid to three features o f ideodynamics w h i c h are unique i n being present simultaneously: empirical testability, parsimony, and equations grounded in real time.

2.1 S I G N I F I C A N T F E A T U R E S O F I D E O D Y N A M I C S One o f the unusual aspects o f ideodynamics is its capacity to predict o p i n i o n trends for w h i c h the time intervals o f the computation are arbitrarily small. A s a result, the computed trends can reflect rapid opinion changes. The intervals o f six or twenty-four hours used for the examples i n this book were chosen because i t seemed unreasonable t o calculate at intervals much shorter than six hours since the message and opinion p o l l data were not k n o w n t o any greater accuracies. Twenty-four hours was used for time trends lasting longer than one year in order t o decrease the time needed for the computations. The logistic equation is the best k n o w n other example o f a calculation for social change where the t i m e i n t e r v a l o f c a l c u l a t i o n can be o f i n f i n i t e s i m a l size (Bartholomew, 1976; Fan, 1985a; H a m b l i n , Jacobsen, and M i l l e r , 1973). Other investigators w h o have e x p l i c i t l y included time i n their models have usually used t i m e intervals ranging f r o m weeks to months. O b v i o u s l y , the longer the time interval, the less precise w i l l be the calculations o f public opinion or any other social response. As the time interval diminishes, messages appearing i n one time interval w i l l continue t o exert their influence i n the next t i m e interval. T o account f o r this phenomenon, time lags have been invoked f o r the continued persuasive force o n messages. A s i n this book, lagged information has typically been assumed by other authors to decrease geometricail v or exponentially over weeks or months (e.g. Hibbs, 1979; Ostrom and Simon, 1985). Besides permitting o p i n i o n calculations over time intervals as short as hours, ideodynamics is also testable empirically. For such tests, i t is essential that the number o f parameters i n the model be small w i t h respect to the number o f predicted values w h i c h can be compared w i t h e m p i r i c a l data. G i v e n enough variable parameters, a general model might f i t any set o f data. In ideodynamics, i t is possible to obtain a very large number o f computed values by calculating o p i n i o n time series.

Copyrighted material

28

Predictions

of Public Opinion from the Mass

Media

In fact, there is no theoretical l i m i t to the number o f points w h i c h can be tested because a new set o f values is predicted as soon as another time interval is added. A new interval can be added either by extending the test time or b y s u b d i v i d i n g the original test interval into smaller subintervals. B y creating a new time interval, the constants i n the ideodynamic equations do not increase. Therefore, after fitting the parameters to a few time points, i t becomes impossible to adjust the constants to f i t later data points. A d d i t i o n a l p o l l values w i l l then test the model critically since the constants w i l l already have been set and can no longer be adjusted. I f the model gives good fits for a large number o f cases, despite the paucity o f parameters, then die model w i l l have been shown to be both general and r o b u s t I t was to explore generality and robustness that ideodynamics was tested w i t h six examples using the simplest formulation w i t h the m i n i m u m number o f parameters. The c r i t i c a l empirical testing o f ideodynamics then takes advantage o f time series w i t h hundreds or thousands o f t i m e steps and a s i m i l a r number o f predictions (Chapter 6). Besides being able to study the sufficiency o f very few parameters, the testability o f ideodynamics also permits b o l d s i m p l i f y i n g approximations i n the choice o f relevant persuasive messages. Therefore, A P messages alone were assigned to represent all mass media messages. This justification is based o n the structure o f American news diffusion. In the U n i t e d States, the w r i t t e n press read by the majority o f the population is locally based. A relatively small percentage o f the population w i l l get their news f r o m either news magazines l i k e Time or Newsweek or newspapers w i t h national circulation like the New York Times, the Walt Street Journal, or USA Today. Most local newspapers do not have the resources to have their o w n reporters on the national or international scene. Therefore, these papers rely on the wire services for their coverage o f non-local n e w s - t h e news for all o f the topics i n this book. For these topics, then, most readers read material c o m i n g directly from the wire services. A m o n g these, the A P is clearly d o m i n a n t Given its prominence and its very wide distribution, the A P also tries to take neutral positions so that its stories w i l l be acceptable to publishers w i t h different political preferences. The other c o m m o n source o f news is the electronic media, w h i c h were not included i n the analyses i n this book. This omission was due to the d i f f i c u l t y i n assessing T V and radio news. I t m i g h t have been possible to use the news summaries i n the Vanderbilt Television News Archives (Chapter 3). However, these summaries were extremely brief and d i d not give a complete idea o f broadcast content As a r e s u l t the approximation was made that A P stories c o u l d also represent news i n the electronic media even though those stories were not quoted verbatim i n news shows. C o m m o n observation o f the s i m i l a r i t y i n news f r o m the A P and electronic broadcasts suggests that this approximation is also plausible. Indeed, Paletz and Entman (1981) have reported that there are frequently great similarities i n reports f r o m various segments o f the mass media. W i t h these justifications, the A P alone was used to represent all national and international news f r o m both the w r i t t e n and electronic press. Given the testability o f the model, this approximation c o u l d at least be t r i e d . I f it was i n v a l i d , then inaccurate opinion time trends w o u l d be calculated and it w o u l d be known that one or more aspects o f the model, including the choice o f news source, was f a u l t y . The p r o b l e m c o u l d be traced to the choice o f the A P i f other choices for persuasive messages gave better calculations. O n the other hand, accurate computations verified empirically for a large number o f issues w o u l d suggest that the model is predictive and that A P news is sufficient for opinion calculations despite the peculiarities o f each issue. One significant aspect o f ideodynamics is its disaggregation o f the population. The result is a nonlinear model i n w h i c h o p i n i o n change is due to the product o f persuasive force functions and target population sizes. In contrast alternative linear models have been proposed where relationships are d r a w n between o p i n i o n and information w i t h o u t subdividing the population.

Ideodynamics

and Previous

Models

29

A m o n g the influential studies i n the persuasion literature, for instance, are those by Funkhouser (1973a, 1973b) reporting that the important items on the p u b l i c agenda correlated w e l l w i t h mass media coverage over a seven-year period. I n the recent study by O s t r o m and S i m o n (198S), presidential popularity was calculated f r o m a number o f factors, including legislative success and activity i n the domestic f i e l d , foreign p o l i c y , e c o n o m i c prosperity, war, sympathy, and unanticipated variables. A t monthly intervals, the factors were computed and correlated directly w i t h presidential popularity. Some information was permitted to act in a lagged fashion extending from month to month in a decreasing fashion. S t i l l on the topic o f presidential politics, Markus (1982) examined the effects o f party identification, perceptions o f candidate traits, and incumbent performance dissatisfaction on political candidate evaluations during a period o f about one calendar year. Since one o f the cases i n this book concerns inflation versus unemployment as a priority problem, i t is useful to note that Hibbs (1979) has also studied the relationship between public perception o f the relative importance o f these t w o problems i n r e l a t i o n to contemporaneous and past behavior o f the economy i n these areas. This is but a partial review o f studies in w h i c h opinion was calculated f r o m the i n f o r m a t i o n structure alone, w i t h o u t subdividing the responding population and modeling the differential action o f different information on different subpopulations. One of the logical consequences o f calculating opinion from the information structure alone is that the computation yields the same results regardless o f the population structure before the calculation time. A n analysis o f the ideodynamic equations ( A p p e n d i x A ) shows that such undifferentiated calculations are equivalent to evaluating the ideodynamic equations under conditions where the population is at e q u i l i b r i u m w i t h the news structure. I n other words, a l l changes caused by persuasive messages are assumed to have occurred. For defense spending, for example, that w o u l d mean that the same o p i n i o n w o u l d have been found in the time interval o f calculation regardless o f whether the population in the previous time interval consisted o f only those favoring more spending or only those supporting less spending. This approximation becomes progressively more inappropriate as the time interval o f computation decreases, since at very small time intervals such as the six or twenty-four hours used in this book, it is rather u n l i k e l y that opinion w o u l d totally reflect the messages in that restricted t i m e period. Ideodynamics is able to overcome the restriction o f population e q u i l i b r i u m by s u b d i v i d i n g the population and m o d e l i n g o p i n i o n f o r m a t i o n via change in opinion f r o m a previous time. One o f the important considerations underlying the formulation o f ideodynamics was that it should be possible to separate message creation f r o m message impact. This separation recognizes that the essential link between message generation and message impact is the message itself. The purpose o f this book is to demonstrate successful tests o f the pan o f the model dealing w i t h message impact. Success here w o u l d mean that the structures used for coding messages could incorporate all the crucial features o f persuasive information needed for explaining message effect. As a result, a full understanding o f persuasion w o u l d be obtained f r o m an analysis o f how messages were generated once message effect could be computed. Message structure in ideodynamics is based on infons and their properties. The a p p r o x i m a t i o n is that messages can be separated i n t o infons using the four dimensions o f position favored, directness, message source, and message index number. The only relevant features o f infons postulated to be essential to persuasion are the content scores, v a l i d i t y scores, and audience size functions. Successful empirical tests w o u l d suggest that models for message generation could stop w i t h the coding o f messages as infons. In fact, models for message creation have already been examined where the output messages are structured as infons. One example is the ideodynamic derivation o f the logistic equation for the diffusion o f innovations. For this equation, it is assumed that infons favoring the adoption o f social innovations are generated i n proportion to the number o f people who have already adopted that innovation. This w o u l d be the case i f all adopters o f an innovation w o u l d have shown and told others

CoDvriahtod material

30

Predictions

of Public

Opinion from the Mass

Media

about the i n n o v a t i o n w i t h a p p r o x i m a t e l y the same effectiveness. This approximation, together w i t h the approximation that the non-adopters d i d not send messages opposing the i n n o v a t i o n , d i r e c t l y leads t o the logistic equation (Fan, 1985a). I n another instance o f the addition o f a message generation step to ideodynamics, it was f o u n d that there w o u l d be no change i n the ratios o f people for and against a position i f partisans o f both sides broadcast their views w i t h the same efficiency (Fan, 1985a). A l t h o u g h this result may n o t seem obvious at first glance, i t is reasonable w h e n the analogy w i t h genetic systems is examined. The ratio o f colorblind to non-colorblind people w i l l stay constant i f both groups can reproduce as efficiently and have the same life expectancy. These t w o examples demonstrate that infons can provide the needed structure to model message generation as w e l l as message impact. Part o f the usefulness o f infons for m o d e l i n g both message synthesis and message impact lies i n their f l e x i b i l i t y . For instance, infons and i n f o n persuasive force functions also permit the modeling o f the reinforcing effects o f the media and information saturation due to message overload. Ideodynamics is unusual i n being able t o include these phenomena e x p l i c i t l y i n its mathematics (Chapter 1 and A p p e n d i x A ) . T h i s is done b y p e r m i t t i n g r e i n f o r c i n g infons to attenuate the persuasive force functions o f infons acting to convert the reinforced individuals. A t the same t i m e , saturation by conversion infons c o u l d also lead to decreased effectiveness o f the conversion infons themselves. However, based o n e m p i r i c a l tests, i t appears that there is n o significant time-dependent o p i n i o n reinforcement (Chapter 7). The f l e x i b i l i t y o f infons is also seen i n their capacity to provide a u n i f o r m structure w h i c h is readily adaptable for analyses o f i n f o r m a t i o n w i t h persuasive components i n many directions, such as favorable and/or unfavorable to one o r more of three groups o f presidential candidates. This ability to consider several different positions simultaneously is d i f f i c u l t to achieve w i t h schemes such as those o f Page and Shapiro (1983a, 1983b, 1984), and Page, Shapiro, and Dempscy (1985, 1987) i n w h i c h communications are scored on a directionality scale f r o m pro to con. Important advantages also derive f r o m the nonlinear aspect o f ideodynamics resulting f r o m the m u l t i p l i c a t i o n o f persuasive force functions by the sizes o f appropriate target subpopulations. One example is the ability to overcome i m p l i c i t restrictions on population structure. T o illustrate, consider m i x e d messages i n the simple case o f o n l y t w o positions, p r o and con. A concrete example w o u l d be warning notices in cigarette advertising. The message here is clearly m i x e d . The manufacturer's component is favorable to smoking w h i l e the warning notice is n o t T o s i m p l i f y the analysis, assume the h y p o t h e t i c a l case where the p r o - and antismoking messages have e q u a l persuasive force. I n this case, the directionality w o u l d be neutral using the previously cited methodology o f Page, Shapiro, and Dempsey. I n their calculations, such a neutral message should not affect public opinion. T o examine the situation more closely, consider a population consisting entirely o f smokers. For such a group, the antismoking component o f the message m i g h t make some o f the members want to q u i t . Whether they w o u l d actually q u i t is explored in the extension o f ideodynamics to habits (Fan, 1985b). Therefore, a neutral message w i t h an antismoking component is l i k e l y to w i n converts t o the nonsmoking cause when the population consists only o f smokers. Conversely, i f the population consisted o n l y o f nonsmokers, a certain number o f these persons w o u l d want to start when faced w i t h a message w i t h a p r o - s m o k i n g component. T h i s analysis demonstrates that n e u t r a l messages are not necessarily neutral since population shifts can occur. Instead, neutral messages can cause an increase i n the number o f either smokers or nonsmokers—depending on the starting population. Therefore, the persuasive effects o f messages cannot be calculated in the absence o f know ledge o f the starting population. I f the latter does not consist o f o n l y smokers or nonsmokers, but is rather a m i x e d group, then the arguments just presented w i l l apply to the i n d i v i d u a l subpopulations. I f the pro- and antismoking

CopyriohlGti mstoris

Ideodynamics

and Previous

Models

31

messages were o f equal force and i f each group o f messages is able to persuade one percent o f their target subpopulation, then one percent o f the smokers should want to quit w h i l e one percent o f the nonsmokers should want to start. O n l y w i t h equal numbers i n the t w o groups w o u l d there be equal traffic in both directions, resulting i n a net change o f zero in the desire to smoke. Therefore, i f directionality is the only measure for persuasive messages, and i f population shift is correlated w i t h directionality alone, the i m p l i c i t approximation is that the population is sufficiently close to equal division in favor o f pro and con that there is an approximate net balance between the movement f r o m pro to con and con to pro. Ideodynamics does not make this approximation. Instead, the sizes o f the target subpopulations are entered directly into computations o f o p i n i o n m o v e m e n t f r o m both pro to con and con to p r o - u s i n g products o f these population sizes and their corresponding persuasive force functions. The adaptability o f ideodynamics is further demonstrated by its a b i l i t y to examine the frequently mentioned concept o f opinion leadership originally proposed by Lazarsfeld, Berelson, and Gaudet (1944) (see also Campbell, 1979; Katz, 1957; Katz and Lazarsfeld, 1965; Weiss, 1969). I n this concept, information in the mass media is postulated to act first on opinion leaders, w h o then transfer their o w n ideas to the rest o f the population. Ideodynamics can treat opinion leadership i n t w o different ways. I n the most complete way, the total population w o u l d first be d i v i d e d into subpopulations o f o p i n i o n leaders and followers. The analysis w o u l d first be made for the impact o f media messages on the leaders. Then each infon generated by the o p i n i o n leaders c o u l d be modeled for its content, source, and audience size. F i n a l l y , ideodynamics w o u l d consider infons f r o m the opinion leaders acting on the population as a whole. Such an analysis c o u l d be quite complex, requiring the inclusion o f significant imponderables such as the identification o f the opinion leaders for particular issues and their infon generation patterns. The alternative method for including opinion leadership is based on an analysis o f the functions o f the o p i n i o n leader. I f media messages are not substantially distorted during the two-step transfer, then the opinion leaders could be considered to be amplifiers o f the message. A significant amount o f time might be needed for the opinion leaders to retransmit messages. This w o u l d only enter into the ideodynamic calculations by shifting the audience size curve to later times, assuming that the original source o f the infons was the mass media. Therefore, i f most media messages were retransmitted accurately by the opinion leaders and i f the retransmission had a characteristic time delay, then the mass media c o u l d still be considered to be the original message sources w i t h the effects o f the opinion leaders being absorbed into a time delay in the audience size curve for mass media infons. F r o m this argument, the decrease in audience size after the transmission o f a mass media message w i l l g i v e an idea about the l i k e l i h o o d o f the importance o f opinion leaders responsible for a second step transfer o f mass media information. In the studies in this book, the effects o f a l l mass media infons disappeared exponentially w i t h a one day h a l f - l i f e . Therefore, any delays due to o p i n i o n leadership must occur w i t h i n a very few hours. So far, the stress has been o n the ability o f ideodynamics to permit different information to have different effects on different subpopulations. Another type o f population heterogeneity tolerated by the model involves n o n u n i f o r m information transfer. For example, no d i f f i c u l t y is presented by people preferentially exposing themselves to i n f o r m a t i o n compatible w i t h their o w n system o f beliefs (Klapper, 1960; Sears and Freedman, 1967; Campbell, 1979). Since such information should reinforce beliefs, preferential exposure is reflected in the constant m u l t i p l i e r describing the persuasive power o f the reinforcing infons (Appendix A, denominator o f Equations A. 12 and A . 13). I f a l l subpopulations are as a v i d in a v o i d i n g i n f o r m a t i o n favoring other positions, then the same constant can be used for all reinforcing infons. It is also unnecessary that every member i n the population have the same chance o f receiving all infons. The model only requires that an infon s chances o f reaching

32

Predictions

of Public

Opinion from

the Mass

Media

more people increase i n proportion to the number o f copies o f the infon released, regardless o f an infon's geographic location or other special broadcast circumstances. For instance, one class o f infons, those f r o m personal experience, w i l l o n l y affect the 'person w h o had the experience. I t is sufficient that more individuals w o u l d receive personal experience infons as more are generated-which is certainly the case. This class o f infons is obviously just one example o f the geographic localization o f persuasive messages. Here, the sender and receiver occupy exactly the same site. In a less extreme case, persuasive messages might generally be localized to the region o f the sender, as was f o u n d b y Haegerstrand (1967) for the d i f f u s i o n o f such agricultural innovations as the antitubercular vaccination o f cattle i n Sweden. Again, there is no d i f f i c u l t y so long as more i n f o r m a t i o n means more people having a chance o f becoming exposed. It is only important that there be no large population isolates where the information available to the general population penetrates w i t h either m u c h greater ease or difficulty. I f the isolated populations are small, then their effects can be ignored. I n ideodynamics, there is n o statement o f the amount o f time that a person needs to stay i n a p a r t i c u l a r s u b p o p u l a t i o n before u n d e r g o i n g an o p i n i o n change. Therefore, a person might move f r o m favoring less defense spending to the position o f supporting more spending almost instantaneously. The model o n l y states that at least t w o infons are required for this conversion w h i l e o n l y one is required for the movement f r o m less spending to same spending.

2.2 M O D E L C O M P A R I S O N S H a v i n g discussed some o f the i m p o r t a n t features o f ideodynamics, i t is appropriate to compare this model w i t h others. A reasonable starting point m i g h t i n v o l v e the topic j u s t discussed, namely heterogeneities w i t h i n the population. A variety o f other mathematically based models do not similarly discuss the question of population inhomogeneities ( A l l e n , 1982; Bartholomew, 1976, 1981, 1982; Bender, 1980; Bender and Speckart, 1979; Brams and Riker, 1972; Cavalli-Sforza and Feldman, 1 9 8 1 ; C a v a l l i Sforza et a l . , 1982; Coleman, 1964; C o o k et ah, 1983; Daley and Kendall, 1965; Goldberg, 1%6; Gray and v o n Broembsen, 1974; Huba and Bentler, 1982; Huba, W i n g a r d , and Bentler, 1 9 8 1 ; Karmeshu and Pathria, 1980a, 1980b; M c l v e r and Carmines, 1981; Sharma, Pathria, and Karmeshu, 1983). T u r n i n g to less mathematical models, ideodynamics is consistent w i t h the idea that at any one t i m e there w i l l be "innovators" and "laggards" (using Rogers and Shoemaker's, 1971, terminology) among the remaining individuals w h o have not been converted The requirement i n ideodynamics is only that the same proportion o f a subgroup s h o u l d be recruited, at a l l times, upon contact w i t h an i n f o n o f a particular strength. Thus there w o u l d be n o c o n f l i c t w i t h laggards b e c o m i n g gradually more persuasible as additional infons favoring change are sent to the target subpopulation and as the innovators are convinced to change their minds. The separation o f a subgroup into laggards and innovators is but one example of d i v i d i n g a subpopulation i n t o members ranging f r o m one extreme to another. Obviously, different results are expected whenever measurements are made on persons at one extreme or another, regardless o f the d i m e n s i o n used for scoring the individuals. For instance, Coleman, Katz, and Menzel (1966) measured the time of adoption o f a new antibiotic among physicians w i t h different social traits. The percentage making the adoption was charted o v e r a period o f about a year and a half for t w o sub populations at the low and high extremes for characteristics such as attendance at meetings, subscriptions to j o u r n a l s , and social participation i n the medical c o m m u n i t y . The most sensitive time for comparing the t w o populations was that at w h i c h 50-percent adoption occurred, since the steepest rate or increase typically occurs at this time. The t i m e for 50-percent adoption usually differed by t w o to five months for extreme subgroups depending on the social characteristic used for partitioning the population. However, the general shapes o f the adoption curves always had the typical S-shape characteristic o f the logistic equation. The curves

Copyrighted material

Ideodynamics

and Previous

Models

33

began j u s t after the introduction o f the antibiotic, hence zero-percent adoption, and Finished at approximately 100 percent, w i t h the period required for the total change being i n the range o f a year and a h a l f to t w o years. I n general, the early adopters were physicians w i t h the greater social interactions. G i v e n the s i m i l a r i t i e s i n the shapes o f the curves for subpopulations w i t h extreme characteristics, the curve for the total population w o u l d also have had the same logistic shape somewhere between the curves for the population extremes. I n other w o r d s , the p o p u l a t i o n treated i n these studies was p r o b a b l y s u f f i c i e n t l y homogeneous that the logistic plot c o u l d apply to the population average w i t h o u t denying the existence o f population heterogeneities. I f desired, ideodynamics can also be applied to any subpopulation o f the total. I n this book, for example, the unawares are ignored. T h i s is one o f the advantages o f the introduction o f infons. The analysis involves defining the infons acting on each o f the subpopulations remaining in the analysis and then describing the action o f these infons on a single subpopulation or group o f subpopulations, a step w h i c h can be accomplished w i t h no changes i n the model. It should be noted that any population, however s m a l l , can always be further subdivided to make even finer distinctions among the i n d i v i d u a l members u n t i l the subpopulations are so s m a l l as to have o n l y one member. A s l o n g as a subpopulation is larger than one person, almost all social science models assume that the subpopulation is sufficiently homogeneous that some generalizations can be made. Thus when Coleman and his colleagues drew time trends for doctors w h o attended four or more meetings, the approximation was that these physicians were sufficiently homogeneous that their data could be pooled to give a c o m m o n curve. I n brief, Rogers and Shoemaker's separation o f laggards f r o m innovators makes generalizations about these subgroups. S i m i l a r l y , Coleman and his colleagues draw generalizations f r o m their physicians pooled by social traits. The broad generality i n ideodynamics about population behavior is to assume that the percentage recruitment f r o m a target p o p u l a t i o n is p r o p o r t i o n a l to the size o f that s u b p o p u l a t i o n . Fortunately, this approximation permits certain population heterogeneities and means that ideodynamics is compatible w i t h the results o f other investigators, as j u s t demonstrated and as discussed in the preceding section. In addition to being compatible w i t h other models f r o m the standpoint o f the treatment o f p o p u l a t i o n heterogeneities, ideodynamics can also encompass other models. One group that can be encompassed includes models for social change i n w h i c h time does not appear e x p l i c i t l y i n the analyses and in w h i c h the information passing f r o m information sender to receiver is not measured, but deduced. Some o f these models have emphasized the pathways by which social decisions occur w i t h i n a single i n d i v i d u a l . For instance, Rogers (1983) has proposed that knowledge o f an innovation is f o l l o w e d by persuasion before a final decision to adopt. H e has also noted that innovators adopt an i n n o v a t i o n earlier than laggards. The steps i n decisionmaking have also been studied by Bentler and colleagues (Bentler, 1980; Bentler and Speckart, 1979; Huba and Bentler, 1982; Huba, W i n g a r d , and Bentler, 1981) based on modifications o f the model o f Fishbein and Ajzen (1975). These authors have proposed for the case o f chemical dependency that an individual typically moves f r o m usage o f alcohol to cannabis to hard drugs like heroin. As already discussed, the system o f social networks through w h i c h persuasive i n f o r m a t i o n passes has been studied extensively by Coleman, K a t z , and Menzel (1966) and Granovetter (1973, 1978, 1980), among others. Here, the emphasis was on w h o was l i k e l y to interact w i t h w h o m and suggested that innovations first entered social networks through weak interactions w i t h other groups. Then, the innovation spread rapidly upon penetration into a group o f tightly knit homophiles comprised o f closely interacting individuals sharing common traits. Since studies o n the pathways o f social interactions and d e c i s i o n m a k i n g frequently use data f r o m one-time surveys, the messages and their associated infons are not measured d i r e c t l y . Instead, inferences are made about the transmitted messages. For instance, awareness o f innovations was assumed to be due to messages arriving at the population at the time that awareness occurred; messages

Copyrighted material

34

Predictions

of Public

Opinion from

the Mass

Media

favoring hard drug consumption were assumed to be correlated w i t h cannabis usage; and additional messages about an innovation were presumed t o be transmitted as interpersonal communications occurred in a social network. W i t h its emphasis on real t i m e , ideodynamics is generally consistent w i t h models describing the steps j u s t described for social changes. A n ideodynamic analysis w o u l d formulate the implied messages i n terms o f infons and w o u l d add the crucial element o f real time. F o r instance, i f the time f r o m awareness through persuasion and decision is relatively rapid, then the population conversion model and its associated equations w o u l d treat these t w o steps as occurring essentially simultaneously, w i t h persons moving directly f r o m unawareness to adopting a position. O n the other hand, some decisions might indeed take a long time. For instance, the time between awareness and adoption o f the innovation o f 2-4D weed spray among I o w a farmers was i n the range o f years (Beal and Rogers, 1960). I n this case, the population conversion m o d e l w o u l d reflect movement o f the unawares first into the pool o f aware but undecided, and then into adoption o f the innovation. For cases where economic decisions are involved, the ideodynamic equations for habits (Fan, 1985b) m i g h t be more useful. Other time-independent studies have been performed where subjects are exposed to actual persuasive communications i n a laboratory setting, w i t h the characteristics o f the population being measured both before and after the exposure. Using this protocol, for instance, Iyengar et a l . (1984) have suggested that i t is more d i f f i c u l t t o influence experts than novices through television news o n the subject o f energy policy. Ideodynamics suggests that these studies might benefit f r o m a consideration o f the time needed for persuasion. For instance, i t may be informative to compare the results o f survey questionnaires at various times after the exposure to television news rather than o n l y once, such as immediately after information exposure. I f there is any lag i n the decisionmaking process, say from awareness through persuasion to decision, then the time o f the postinformational measurement m i g h t be o f great importance. Aside f r o m social science descriptions i n w h i c h real time is n o t e x p l i c i t l y included, there are other models i n w h i c h time is explicitly incorporated. I n one class o f these models, persuasive messages are not e x p l i c i t l y described as is done i n ideodynamics. Instead, the messages are inferred f r o m the characteristics o f the message senders. O f these, perhaps the m o s t successful has been the e p i d e m i o l o g i c a l l y based m o d e l y i e l d i n g the logistic p l o t for the d i f f u s i o n o f innovations w h i c h was discussed i n the preceding section. This equation derives f r o m assuming that a l l people adopting the innovation generate favorable infons equally. I n a more recent variant, Sharma, Pathria, and Karmeshu (1983) postulate that some people receiving a message about an innovation are "stifiers" w h o do not generate favorable infons. Another model in which infons are deduced from the structure o f message senders is based o n genetics (Cavalli-Sforza and Feldman, 1931). Here the messages f r o m parents (vertical transmission) are assumed to have forces different from those f r o m peers (horizontal transmission) and teachers (oblique transmission). This model was formulated i n terms o f generations and is therefore most appropriate for slow-moving transmissions o f social traits such as culture or religion, w h i c h typically change very little w i t h i n a generation (Cavalli-Sforza et al., 1982). F r o m a historical perspective, i t is understandable that models based o n epidemiology and genetics do not e x p l i c i t l y include measured messages because the factors responsible for change i n these areas are not easily determined directly. For example, the logistic curve is based o n epidemiological models o f infectious diseases where the infectious virus, bacteria, or other microorganism usually cannot be traced d u r i n g movement f r o m an infected i n d i v i d u a l to a new v i c t i m . S i m i l a r l y , the transmissible agents i n genetics are the genes i n the sperm and eggs w h i c h , again, are not easy to measure directly i n the natural population. Ideodynamics can also encompass other models w i t h i m p l i e d messages by e x p l i c i t l y coding the i m p l i e d messages as infons (Fan, 1985a, and Chapter 7 ) .

Copyrighted material

¡deodynamics and Previous

Models

35

There are also other models that are like ideodynamics in including real time and measured messages. The key advantage o f measured messages is that they reflect the uncertainty i n the w o r l d . However, new can never be predicted w i t h certainty, so calculations o f social change based on actual messages can never extend beyond the time when the messages are available. Since the crucial aspect o f measured messages is that their composition cannot be predicted ahead o f time, measured messages for this discussion w i l l include not o n l y direct messages but also those presumed to be correlated w i t h historical events. The main constraint is that the measured messages should not be deduced solely f r o m the structure o f the population. Therefore, some measured messages can be quite direct, such as newspaper articles when scored, for example, by Page and Shapiro (1983a, 1983b). Other measured messages m i g h t be more indirect, such as those correlated w i t h historical facts, e.g. roll-call votes i n Congress (Ostrom and Simon, 1985). The p u b l i c is not l i k e l y to be aware o f such votes. H o w e v e r , indirect messages arising f r o m such votes might indeed be disseminated to the p u b l i c . Ideodynamics has the advantage over more restrictive models o f s u b d i v i d i n g the population before computing opinion. Time-dependent models i n w h i c h messages are e x p l i c i t l y measured include the agenda-setting and persuasion models by E r b n n g , Goldenberg, and M i l l e r (1980) and M a c K u e n ( 1 9 8 1 , 1983, 1984). It has already been shown that these models can all be considered to be special cases o f ideodynamics (Fan, 1984). L i k e ideodynamics, these models also divide the population into different subpopulations. Throughout the previous discussion o f other models, an effort was made to see i f previously reported phenomena were inconsistent w i t h ideodynamics. I t is reassuring that this m o d e l is c o m p a t i b l e w i t h these other social science models, some superficially disparate. Therefore, as noted i n the introduction, ideodynamics is not so m u c h a competing as an umbrella m o d e l - t h e equivalent to the elephant i n the elephant analogy- with a structure suitable for incorporating the details o f many o f the other models discussed above. Important novelties o f ideodynamics include the incorporation o f real time i n the analyses, the existence o f the logistic equation as a special case, and the e x p l i c i t mathematical modeling of o p i n i o n reinforcement and other message interactions. A n y alternative models s h o u l d also be able to account for a l l these features simultaneously. The capacity to derive the logistic equation is especially important given its ubiquitous usefulness i n e x p l a i n i n g social science t i m e courses. The importance o f concurrent studies i n real time o f a n u m b e r o f measurable social variables-media messages and public opinion i n this b o o k - h a s also been stressed by Neuman (1987) i n his extensive review of the persuasion literature.

real

messages

Cc

terial

Copyrighte U

3

Data For Calculating Public Opinion

The theory i n Chapter 1 provided a method for calculating the results o f o p i n i o n polls i n a t i m e - d e p e n d e n t fashion given the information available to the public. It was also demonstrated that this model could be tested using t w o types o f data, a time series o f p u b l i c o p i n i o n p o l l s , and a representative sample o f the i n f o r m a t i o n available to the population for each p o l l e d issue. T h i s chapter concerns the actual data used for the e m p i r i c a l testing o f ideodynamics. Further details are given i n Appendix D. 3.1 T I M E SERIES O F O P I N I O N P O L L S As discussed i n the introduction, one o f the key goals o f this book is to assess the generality o f ideodynamics. Therefore, i t was important to study issues w h i c h were varied and disparate i n nature. As a result, case studies were performed for t w o foreign policy issues, a domestic policy issue, t w o economic issues, and an electoral campaign. One important criterion was that the information influencing the p u b l i c should be r e a d i l y available for study. I t w o u l d o b v i o u s l y have been very d i f f i c u l t to calculate public opinion f r o m persuasive messages i f they could not be captured for analysis. Since the p u b l i c record was most complete for the news p o r t i o n o f the mass media, the decision was made to concentrate o n issues where this was the source o f the majority o f the relevant messages. The most convenient method for obtaining mass media messages is to retrieve them f r o m an electronic data base such as the Nexis data base sold by M e a d Data Central o f D a y t o n , O h i o . Since the data base extended back to 1977 for the A P dispatches used for these studies, the decision was made to study o n l y issues w i t h poll time series since 1977. As argued in Chapter 2, the Associated Press was l i k e l y to be representative o f both the written and electronic press. Another consideration was that the p o l l points f r o m most t i m e series should have changed s i g n i f i c a n t l y d u r i n g the p o l l i n g period. Since very simple models could predict no change at a l l i n p o l l results, the most dramatic tests w o u l d involve polls w i t h marked opinion changes. Parenthetically, the c o n d i t i o n o f o p i n i o n change meant that the p u b l i c was reasonably persuasible for those issues where change was f o u n d . B y d e f i n i t i o n , public o p i n i o n w o u l d have stayed constant for issues for w h i c h the public had very f i r m convictions. For instance, the abortion p o l l series since 1977 w i t h the largest number o f time points was one f r o m N B C news f r o m 1977 to 1982. I n this series, the widest opinion swings were only five to ten percent (data f r o m B . I . Page and R. Y. Shapiro, c o m p i l e d at the National O p i n i o n Research Center ( N O R C ) i n Chicago).

Copyrighted material

38

Predictions

of Put>lic Opinion from the Mass

Media

Five issues were chosen for careful analysis on the basis o f (1) representing issues i n many domains, (2) being marked by many time points since 1977, (3) reflecting significant o p i n i o n changes, and (4) h a v i n g most relevant persuasive messages c o m i n g f r o m the mass media. The polls (see Appendix B for details) were m a i n l y f r o m the extensive compilations o f time series made by Page and Shapiro at N O R C (see Page and Shapiro, 1982). For comparison, a sixth test issue involved opinion on Contra aid for w h i c h o p i n i o n stayed reasonably constant. F o r this issue, the poll series were obtained f r o m the Roper Center at the University o f Connecticut. One o f the most interesting p o l l series was that o f public o p i n i o n toward the advisability o f more, same, o r less defense spending. The important persuasive messages for this topic were most l i k e l y to be localized i n the mass media since the relevant considerations were constantly changing. A l l other media, such as books, were too slow to reflect these fluctuations. The public certainly c o u l d not calculate the amount o f money w h i c h should be spent on defense by making calculations f r o m a set o f general principles. Very few people w i t h an o p i n i o n on defense spending even knew what the defense budget actually was. The Page and Shapiro compilations contained four poll series f r o m the period 1977-84. These time series were all pooled into one because a c o m m o n curve could be drawn through all the points (Figure 3.1). However, the reader can concentrate on any o n e data set since each time series is d e n o t e d by a different s y m b o l . I n i t i a l l y , opinion projections were made for 1977-84. Later, a test was made for the ability to extend the text analysis and o p i n i o n calculations to 1986. F o r these studies, additional published polls were obtained f r o m the Roper Center at the University o f Connecticut, y i e l d i n g a total time series w i t h 62 polls. The defense spending issue was interesting because changes i n p u b l i c opinion u n d e r w e n t a large increase and then a marked decrease. However, the changes were slow. D u r i n g 1979, those favoring more defense spending rose dramatically f r o m 20¬ 30 percent to approximately 70 percent Opinion then remained high for another year before the one-year drop back to 20-30 percent i n 1981. Another issue w i t h striking opinion shifts was whether more A m e r i c a n troops should have been sent to Lebanon i n 1983-84. The troops had been sent as part o f the Multinational Peacekeeping Force after Israel withdrew f r o m Lebanon in 1982. In contrast to the defense spending example, o n l y t w o to three months were needed for people i n favor o f sending more troops t o change from 7 percent up to 31 percent and back down to 9 percent Together, troops i n Lebanon and defense spending permitted the testing o f the computational methods for dramatic opinion changes spanning days or months. These t w o topics concerned p o l i c y issues. The next topic i n v o l v e d p o l i t i c a l popularity i n the 1983-84 presidential campaign. The choice was made to study the Democratic candidates because none had the extra advantage o f incumbency. A l s o , the analysis ran f r o m 1983 to 1984, stopping j u s t short o f the first real test o f strength, the Iowa caucuses i n 1984. A f t e r that t i m e , relevant i n f o r m a t i o n became less and less restricted to the journalistic press because candidate advertising and other campaign messages became progressively more s i g n i f i c a n t Unfortunately, these non-news media messages were much more difficult t o obtain. They were also much less u n i f o r m l y broadcast to the nationwide p o p u l a t i o n . The d i f f i c u l t y o f studying the non-news messages m i g h t have made further o p i n i o n calculations less precise, since accuracy depends o n analysis o f all the information available to the public. W i t h o u t doubt, there was significant local advertising and other campaigning i n I o w a i n preparation for the caucuses and i n N e w Hampshire in anticipation o f the primary w h i c h occurred shortly thereafter. However, the p o l l series for this study was taken for the country as whole. For the nationwide public, the national news media was probably reasonably representative of the information available. The three previous issues share the feature that most members o f the population were probably not personally affected by them. A l t h o u g h a small fraction o f the population was l i k e l y to have been directly involved for any one o f these issues, the majority probably felt no special individual concern about these topics. Even for

Copyrighted material

Data for Calculating Public Opinion

39

troops i n Lebanon, the number sent was only several hundred, so most people w o u l d not personally have k n o w n anyone in the Marine c o n t i n g e n t It seems l i k e l y that the population should have been more malleable for such distant issues than others where most people have a personal stake (Everson, 1982). A n example o f the latter type o f issue is the economic climate o f the country; there was a steady increase i n those feeling that the economy was i m p r o v i n g and a parallel decrease i n those w i t h a downbeat mood during the period f r o m M a r c h 1981 to M a r c h 1984. There was not m u c h change i n the percentage f e e l i n g that economic conditions were staying the same. The economic issue o f the relative importance o f unemployment and inflation is s i m i l a r t o the p r o b l e m o f the economic climate i n that the public had personal experiences o r observations w i t h both problems. B o t h issues m i g h t have been influenced b y factors outside the mass media, i n w h i c h case the model might not have been expected to w o r k i f o n l y A P messages were used to represent the d r i v i n g forces for change. Polls for unemployment versus inflation showed that the public seemed to be very aware o f the problem, as was true for the economic climate, w i t h less than 3 percent i n the N o t Sure g r o u p . T h i s topic was also chosen because there were significant movements i n public o p i n i o n , w i t h o p i n i o n focusing on the importance o f unemployment having a high o f around 50 percent and a l o w o f around 20 percent i n the time period f r o m 1977 to 1980. As a reference point for the other computations, i t was useful to have an issue for w h i c h there was little o p i n i o n change. I f the model is v a l i d then i t should also predict o p i n i o n constancy when required. This test was performed for the issue o f whether aid should have been sent to the Contra guerrillas fighting the government o f Nicaragua f r o m 1983 to 1986. This issue was similar to that o f troops i n Lebanon in being concerned w i t h military involvement i n a small foreign country. F r o m the o p i n i o n standpoint the major difference was the large opinion fluctuations for troops i n Lebanon and the very small variations for Contra a i d . Parenthetically, the polls for Contra aid were o n l y obtained f r o m the Roper Center after scoring the relevant news messages to insure that the scoring w o u l d not be biased by knowledge o f p o l l results.

3.2 R E L E V A N T P E R S U A S I V E M E S S A G E S I N T H E A S S O C I A T E D PRESS W i t h the polls i n hand, the next step was to collect the pertinent persuasive messages. A l t h o u g h i t w o u l d have been desirable to sample both the electronic and w r i t t e n press, i t was d i f f i c u l t and tedious to o b t a i n the unaltered contents o f television and radio news. For television, there were the Television News Index and Abstracts o f the Vanderbilt Television News Archives. However, the summaries were very brief, w i t h abstracters condensing news segments as long as a few minutes i n t o a single phrase. Clearly, analyses based on these summaries w o u l d have depended heavily on the abstracters. The case was much better for the print media. Here, i t was possible to retrieve an entire news item i n v i r g i n f o r m f r o m electronic data bases. These sources contain not o n l y the f u l l texts o f stories f r o m w i r e services such as the A P and U n i t e d Press International, but also newspapers such as the New York Times and the Washington Post, and news magazines such as Newsweek and Time. Since A P stories are l i k e l y to be representative o f the mass media in general (Chapter 2), this book focuses on A P dispatches. I n a d d i t i o n , messages were restricted to this w i r e service because the audience size was more u n i f o r m for news f r o m a single source. It w o u l d have been d i f f i c u l t to w e i g h t the relative audience sizes o f items i n the A P w i t h those, for example, f r o m news magazines w i t h more l i m i t e d circulations. E v e n for A P stories, the impact o f a dispatch depended o n whether i t was actually p r i n t e d and, i f so, whether the placement was p r o m i n e n t . I m p o r t a n t

Copyrighted material

40

Predictions

of Public

Opinion from the Mass

Media

information in this regard could have been obtained f r o m the "news budget" the A P r o u t i n e l y sends to its subscribers twice a day. These budgets are summaries o f important dispatches i n preparation. Newspaper editors save space i n u p c o m i n g editions based on these budgets. Later, a dispatch corresponding to a budget item is sent carrying the designator B J T i n its heading region. These budget stories typically stand a very good chance of appearing in a featured place. Unfortunately, the budget designator was stripped from A P dispatches before entry into the Nexis data base used for these studies. Therefore, there was n o convenient way t o k n o w the probable disposition o f an A P dispatch so all A P dispatches were assumed to be equally important. I n future studies, i t may be more useful to use the V U / T E X T data base f r o m w h i c h the BJT designator is not removed. Fortunately, i t is o n l y necessary that all A P dispatches on a given polled topic have the same prominence since comparisons were only made between stories o n the same issue. There w o u l d have been no problem with all articles on one topic being b u r i e d i n small items i n obscure places i n the mass media w h i l e a l l stories o n another were given featured treatment

3.3 R E T R I E V A L S F R O M T H E N E X I S D A T A B A S E One attractive feature o f using a full-text data base l i k e Nexis is the retrieval method, since the search is not at the mercy o f abstracters. Instead, the investigator chooses combinations o f key words, for w h i c h every w o r d i n a l l articles i n the data base is scanned. A l t h o u g h the basic elements o f the search commands are simple, complex thoughts can be expressed by combining groups o f words (see Appendix B for details o f the retrievals). A l l stories w i t h the d e s i r e d w o r d combinations were identified even i f their main topic was not o n the polled issue. For instance, an article d w e l l i n g on poverty might have included a single statement that too much was b e i n g spent on defense, given the magnitude o f the economic problem. Such a dispatch w o u l d have been identified i n the Nexis search for articles on defense spending. The same article m i g h t w e l l have been missed b y human readers. Therefore, the number o f pertinent dispatches identified by the Nexis search for this book was probably more complete than i n previous reports using human coders. The decision c o u l d have been made to study only those articles concentrating o n the polled issue. However, a deliberate decision was made to include the stories i n w h i c h the issue was o n l y mentioned p e r i p h e r a l l y . T h i s choice seemed more appropriate since the public at large was exposed to the mass media i n a different way f r o m trained readers specifically scoring particular issues. The bulk o f the population was probably not trying to draw critical judgments for most issues and m i g h t not have noticed when one article began and another stopped. Therefore, for the general population, an isolated statement favoring less defense spending might have had as much o f an impact in an article on poverty as in a story on defense spending. W i t h the possibility that the public m i g h t assimilate i n f o r m a t i o n for an issue f r o m a w i d e number o f contexts, the decision was made to be inclusive rather than exclusive i n i d e n t i f y i n g potentially relevant A P dispatches. The c u l l i n g o f t r u l y pertinent stories occurred at later steps i n the text analyses. The result was that over a thousand relevant dispatches were found for each o f the issues i n this book. I t seemed both i m p r a c t i c a l and unnecessary to analyze all dispatches for a l l issues. For four o f the six issues, i t was decided t o retrieve o n l y a few hundred dispatches at random for detailed study. Since the dispatches were all numbered in reverse chronological order f r o m the most recent to the one furthest back i n time, retrieval by random dispatch number should have yielded a representative sample. For u n e m p l o y m e n t versus i n f l a t i o n and Contra a i d , almost a l l o f the identified articles were retrieved. These more complete retrievals were made to test the possibility that a few hundred dispatches were not enough and that a larger sample was needed

Copyrighted material

Data for Calculating Public Opinion

41

Although it w as possible to retrieve the full texts of each of the chosen articles,

the decision was made to restrict all retrievals to text w i t h i n fifty words both before and after one o f the key words used i n the original search. Careful reading o f typical stories indicated that this condition led to the retention o f essentially all relevant text w i t h large portions o f irrelevant text being discarded (see sample text i n Appendix C ) . One of the most frequent examples i n v o l v e d extracts and descriptions of news conferences in which comments were made on a variety o f totally unrelated topics. For defense spending, the search was for AP dispatches w i t h "defense" or a s y n o n y m being close to "spending" or a s y n o n y m . Since the a i m was to f i n d all possibly relevant dispatches, no commands restricted the search to American defense spending. Dispatches on non-American defense spending were eliminated at a later step. A m o n g those dispatches retrieved using the fifty-word rule, the amount o f collected text was equivalent to about 199 words o f retrieved text per dispatch. Since the typical A P dispatch contained 400-900 words, o n l y small extracts were obtained f r o m most dispatches, consistent w i t h the idea that comments on defense spending were frequently i n articles discussing other topics. Text on other topics outside the fifty-word limit was discarded. The i n i t i a l set o f retrievals was made for calculations f r o m 1977 to 1984. A second set o f retrievals was made f r o m 1981 to 1986 to study the ease w i t h w h i c h the computations c o u l d be extended beyond 1984. For this second study, a set o f retrievals was also made for stories about defense waste and fraud. These stories were also analyzed to see i f they contributed significantly to opinion favoring less defense spending. For the topic o f American troops i n Lebanon, the search was for articles w i t h words referring to Lebanon, America, and troops. In contrast to the case for defense spending, many dispatches on troops i n Lebanon were entirely about this topic and were retrieved i n toto since about 420 words were retrieved per dispatch. I n fact, the scanning o f a random collection o f stories showed that many were devoted entirely to Lebanon. In the Nexis data base search for the Democratic primary, there were seven candidates. Therefore, the search was for all dispatches w i t h the name o f at least one o f the candidates. Because the typical retrieval had 310 words per dispatch, under the 4 0 0 - 9 0 0 words per t y p i c a l article, many dispatches were not retrieved i n their entirety. For the economic issues, i t was conceivable that personal experience w o u l d c o n t r i b u t e substantially to o p i n i o n o n the e c o n o m i c c l i m a t e . In principle, i d e o d y n a m i c c a l c u l a t i o n s s h o u l d have i n c l u d e d infons f r o m those sources. Obviously, those infons are very d i f f i c u l t to study. Therefore, the assumption was made that the mass media included most o f the important persuasive messages. The v a l i d i t y o f this a p p r o x i m a t i o n w o u l d be tested by the accuracy o f the o p i n i o n projections. As usual, A P dispatches were considered to be representative o f mass media messages so the Nexis data base was searched for w o r d pairs referring to the economic climate. The retained dispatches typically had 197 words per average story. This number was close to the 199 found for defense spending and corresponded to less than half the text o f a typical A P dispatch. Therefore, most o f the retrieved text came from dispatches in w h i c h the economic climate was o n l y one among several topics discussed. For unemployment versus inflation, A P dispatches were obtained from the Nexis data base by searching for dispatches i n w h i c h unemployment and inflation were close enough that there was l i k e l y to be nearby text comparing the t w o . This gave approximately 177 words per dispatch, a figure comparable to the topics o f defense spending and the economic climate. Therefore, for unemployment versus inflation as w e l l , most dispatches covered several topics, o f w h i c h none were discussed throughout. The A P stories i n the Nexis data base relevant to Contra aid were identified by l o o k i n g for the w o r d Contras i n close p r o x i m i t y to a w o r d referring to spending

Copyrighted material

42

Predictions

of Public Opinion from the Mass

Media

T h i s topic was also l i k e all others except for troops i n Lebanon i n having fewer retrieved words (258 words) per average dispatch than were present i n the typical A P story.

CoDvriahted material

Figure 3.L Poll data for defense spending. A f t e r subtraction o f the Don't K n o w s and the N o t Sures, the data i n Table B . l (Appendix B ) were normalized to 100 percent and plotted using the same symbols, one for each o f the four independent time series. %

Favor

MORE

(4

P o l l

S e r i e s )

?LFavor

SAME

(4

P o l l

S e r i e s )

%

LESS

(4

P o l l

S e r i e s )

Favor

77 78 79 80 Y e a r s 1977-1984

81

82

83

84

Copyrighted material

Copyrighted material

4

Computer Text Analysis by Method of Successive Filtrations

Since the methodology i n this book was designed to project p u b l i c o p i n i o n f r o m persuasive messages i n Associated Press dispatches, t w o steps were i n v o l v e d . The First was t o score the A P dispatches i n the format o f ideodynamics (Chapter 1 and Appendix A ) . The second was to use the scores to project p o l l results. T h e messages were scored using the I n f o T r e n d c o m p u t e r m e t h o d . T h e advantages were t w o f o l d . First, consistency was guaranteed, so human judges c o u l d not be accused o f scoring, either consciously or subconsciously, w i t h one eye o n the measured o p i n i o n t i m e trends i n order t o i m p r o v e the projections. Second, the dictionaries and rules for computer analyses w o u l d shed light on those features o f the text w h i c h were especially important for influencing public o p i n i o n .

4.1 G E N E R A L T E X T A N A L Y S I S P R O G R A M S E a r l y i n the h i s t o r y o f computer-assisted text analysis, the strategy was t o c u s t o m i z e the analyses t o i n d i v i d u a l p r o b l e m s (see, e.g., M c C l e l l a n d ' s N Achievement i n Stone et al., 1966). In more recent times, the emphasis has been on the development o f general text analysis programs and dictionaries applicable to a wide variety o f t e x t The first w i d e l y used program o f this latter sort was the General Inquirer (Stone et which been updated ( K e l l y Stone, The General Inquirer and a number o f relatives assign words i n the text to a

al., 1966), has and 197S). number of predefined categories (typically in the range of a hundred) using

dictionaries and disambiguation rules w i t h various degrees o f complexity (see Weber, 198S for a recent review). The disambiguation is necessary because many words in natural language c a n take on different m e a n i n g s in different c o n t e x t s . A frequently

cited example is the w o r d "spring," w h i c h can refer to a c o i l o f metal, a time o f year, a source o f water, or a j u m p i n g action. A m o n g recent dictionaries and complex disambiguation routines for the General Inquirer are the H a r v a r d V I Psychosocial D i c t i o n a r y ( D u n p h y , B u l l a r d , and Crossing, 1974; K e l l y and Stone, 197S) and the L a s s w e l l V a l u e D i c t i o n a r y (Lasswell and N a m e n w i r t h , 1968; N a m e n w i r t h and Weber, 1984). Text parsing programs have also come f r o m the natural language area o f artificial intelligence. M a n y o f these are also general i n nature, adaptable to diverse texts. A m o n g these are A C T O R ( H e w i t t 1976), B O R I S (Dyer, 1982), H E A R S A Y I I (Fennel and Lesser, 1977), M A R G I E (Schank et al., 1973), R E A D E R (Thibadeau, J u s t and Carpenter, 1982), D E R E D E C (Lecomte, Léon, and M a r a n d i n , 1984), and R E L A T U S (Duffy and M a l l e r y , 1986).

CoDvriahted material

46

Predictions

of Public Opinion from

the Mass

Media

Another general text analysis program based on the assignment o f words to predefined categories is the M C C A program o f M c T a v i s h and Pirro (1985). l i k e the programs mentioned above, M C C A is meant to be useful for a wide variety o f t e x t I n i n i t i a l studies, the A P dispatches retrieved for the studies i n this book were analyzed using the M C C A program, but the output d i d not correlate w e l l w i t h the sense o f the dispatches i n terms o f supporting individual polled positions. A n examination o f the o u t p u t suggested several problems. One o f the most immediate was due to assignment o f individual words to predefined categories. A s a result, relationships between words were lost. A prime example was the use o f negation words l i k e "not." After the program r u n , there was no way to determine what was modified by this w o r d . This was a serious problem for w i r e service stories w h i c h c o u l d cover many different topics i n the same dispatch, as happened i n reports o f press conferences. I n such stories, o n l y a small p o r t i o n o f the dispatch was relevant to the issue under study. Therefore, the mere abundance o f negation words was not h i g h l y indicative o f whether the story had a negative connotation for the topic being examined. A f u r t h e r p r o b l e m was the predefined categories i n the M C C A p r o g r a m , including those for Good, Sin, and Faith-Belief. These categories could give a good idea o f the psychological state o f the message generator. However, the categories were not easily adaptable to policy questions such as whether American troops should be sent to Lebanon, especially since some o f the most crucial words were h i g h l y unusual i n c l u d i n g " I s r a e l i , " " S y r i a n , " " B r i t i s h , " " F r e n c h , " " I t a l i a n , " " C h r i s t i a n , " S h i i t e , " " D r u s e , " and others. T o be successful, a general text analysis program w o u l d have required an astronomical n u m b e r o f w o r d categories to d i s t i n g u i s h between these troop sources w h i l e simultaneously separating A m e r i c a n p o l i c y i n Lebanon f r o m that i n Grenada, where troops were sent simultaneously. T h e unsatisfactory nature o f preassigned categories was not restricted to the M C C A program but was characteristic of all automated text analyses assigning words to a l l purpose categories. Another important drawback was the multipurpose function o f M C C A and other general-purpose programs. T o be useful for a w i d e variety o f text, as mentioned above, disambiguation procedures were needed to account for words having different meanings i n a w i d e variety o f contexts. Therefore, the dictionaries tended t o be relatively s m a l l , w i t h the disambiguation subprograms concentrating on assigning correct meanings to c o m m o n words. As dictionaries g r o w , the d i s a m b i g u a t i o n problems increase even more rapidly. The l i m i t e d dictionaries frequendy were inappropriate for text i m p o r t a n t t o public o p i n i o n . K e y words i n such text were often h i g h l y specialized, i n c l u d i n g many proper nouns as just discussed. Furthermore, each i n d i v i d u a l issue was also l i k e l y to have words requiring customized and unusual implications. For instance, "neglect" i n the context o f defense spending i m p l i e d that the m i l i t a r y budget should be increased.

The refined disambiguation subprograms i n general text analysis programs also meant that the dictionaries were very d i f f i c u l t to change because a single addition o r deletion meant that all disambiguation steps had to be checked to see i f any should be m o d i f i e d by the w o r d change. The absence o f disambiguation procedures w o u l d certainly have facilitated dictionary changes-at the cost, however, o f increased confusion in w o r d usage.

4.2

S T R A T E G Y FOR FILTRATIONS

CONTENT

ANALYSIS

USING

SUCCESSIVE

The previous considerations suggest that an i m p r o v e d method o f computer content analysis should (1) be able to assure a high degree of precision by examining key words i n proper relationships w i t h each other, (2) be able to resolve ambiguities in natural language, and (3) permit the use o f flexible dictionaries i n c l u d i n g very specialized words.

Copyrighted material

Computer

Text Analysis

47

The strategy o f the InfoTrend procedure was to abandon the idea o f making a single pass through the text using a general program and a fixed dictionary. Instead, there was a return to the philosophy o f tailoring the analysis to individual issues. As noted by Weber (1985), single-concept coding schemes often have high validity and reliability." One o f the important departures f r o m single-pass programs l i k e those discussed i n the previous section was the decision to process the text through a series o f " f i l t r a t i o n " computer runs to remove different irrelevant material at each step. Each filtration r u n was guided by a small number o f rules and a rather l i m i t e d dictionary w i t h o n l y a few key words, words w i t h few ambiguity problems i n the context o f the remaining text. There were some unexpected advantages to this strategy o f m a k i n g m u l t i p l e passes through the text. B y doing so, i t was possible for the dictionary and rules at each pass to be very simple, thereby m i n i m i z i n g the errors i n their construction. As soon as one or t w o criteria were identified by reading a random sample o f the text (typically 150,000 to 200,000 characters), a small dictionary and set o f rules c o u l d be devised for eliminating irrelevant passages. A t the end o f the r u n , the text was more homogeneous than that i n the input file. This meant that it was much easier to decide on the dictionary and rules for the next filtration step. Greater text homogeneity meant that the reader's m i n d was not cluttered w i t h material w h i c h was not pertinent. However, as important, i f not more so, was the fact that w o r d connotators and relationships useful for the analysis were typically established by each filtration r u n . For example, the first filtration step for dispatches on defense spending restricted this spending to A m e r i c a n p o l i c y (see section b e l o w ) . Therefore, this step already imposed the relationship that defense spending be l i n k e d to the United States. As a consequence, there was no need to w o r r y about this linkage i n further steps i n the analysis. Another advantage o f the filtration steps was the disambiguation accomplished. As mentioned above, the w o r d "neglect" t y p i c a l l y i m p l i e d the support o f more military spending. W i t h o u t the filtration steps to focus on paragraphs specifically discussing defense spending, it w o u l d not have been possible to give "neglect" this very special meaning. Superficially, i t m i g h t be supposed that multiple computer runs w o u l d greatly slow the analysis. Interestingly, this was not the case. A l t h o u g h there was time lost i n analyzing the same text several times, there was a compensating gain. First o f all, as noted above, the steps for developing the dictionaries and rules were simple and hence rapid. I n addition, the machine time needed for each run was shortened because the dictionaries and rules were shorter than w o u l d have been needed for the more complex processing needed for a single-pass analysis. Therefore, a great deal o f computational t i m e was saved. Furthermore, w i t h each succeeding pass the amount o f text diminished due to the elimination o f irrelevant material, so the runs became even faster. Given the slowness o f checking long dictionaries and evaluating complicated assignment rules, the total machine run time c o u l d be shorter w i t h a strategy o f multiple runs using simple dictionaries and rules. A f t e r successive filiations, the text became sufficiently homogeneous that a simple dictionary and a small number o f rules c o u l d be constructed for assigning numerical scores for the extent to w h i c h each A P dispatch favored the positions being considered. These scores were for the infon contents discussed in Chapter 1.

4.3 S K E T C H O F F I L T R A T I O N RUNS

AND

SCORING COMPUTER

PROGRAM

A l t h o u g h the text analytic steps --both for filtration and scoring-are described i n detail i n Appendix C for each o f the six cases studied, i t is useful to o u t l i n e the procedure here. The analysis involved (1) checking the text for dictionary words, (2)

Cc

48

Predictions

of Public

Opinion from

the Mass

Media

identifying clusters o f key words, and (3) making filtration or scoring decisions based o n the w o r d clusters. T o begin w i t h the first step, a new dictionary was constructed for each analysis w i t h words grouped under a l i m i t e d number o f concepts. For defense spending, for instance, one o f the essential concepts was " A m e r i c a n " since the topic was spending by the U n i t e d States and not some other country. The words under this thought d i d not actually have to be a s y n o n y m for America. I t was sufficient that there be a strong i m p l i c a t i o n o f the U n i t e d States. Therefore, acceptable words included not o n l y " A m e r i c a , " " U n i t e d States," and "U.S.," but also "House," "Senate," " F o r d , " "Carter," and "Reagan." After the dictionary check, the concepts corresponding t o the dictionary words were evaluated. I n scoring defense spending, for instance, one filtration step was to select A P dispatches discussing the idea o f ' A m e r i c a n defense spending." This idea r e q u i r e d the simultaneous presence i n the article o f the three concepts of: "American," "defense," and "spending." This example is the simplest case, where the o n l y requirement was that certain concepts all be together i n the same t e x t I n more complex cases, it was necessary to insist that t w o words be close together to belong t o the same t h o u g h t For instance, a "defense" w o r d had to be close to a "spending" w o r d i n the filtration step to select paragraphs actually devoted to defense spending. Otherwise, the spending was probably for some other purpose. I n yet other cases, the sequence was also important. I n the presidential primary example, the w o r d "endorsed" i m p l i e d the concept o f "favorable to candidates whose names are further into the text." I n other words, "endorsed" was interpreted to favor a candidate o n l y i f i t appeared before the candidate name, as i n "endorsed Mondale." There was no such connotation when the sequence was reversed, as i n " M o n d a l e endorsed/ However, "endorsed" was sometimes also found i n the context o f "endorsed by." I n this w o r d pair, the w o r d "endorsed" was combined w i t h the concept o f "change the direction o f action o f the w o r d " due to the presence o f the w o r d " b y . " The result was the idea o f "favorable to candidates whose names are earlier i n the text." I n this way, the phrase "Mondale endorsed b y " was scored as supporting Mondale while "endorsed by M o n d a l e " was not. Similarly, i t was not difficult to add an additional w o r d l i k e " n o t " o f the negation class so that the phrase "Mondale was not endorsed by" c o u l d be considered to be unfavorable to Mondale. Therefore, both the order and distances between words could be used for the filtration or scoring decisions. After the words i n the text were represented by the concept categories and after groups o f concept categories were evaluated to describe more complex meanings, decisions were made about the text being considered. These decisions depended on whether the text was to be filtered or scored. For filtration steps, certain complex meanings led to the text being discarded w h i l e others meant that the text was retained. For some runs, the decisions pertained t o the entire retrieved text from a dispatch w h i l e , for others, decisions were made for individual paragraphs w i t h i n a story. The program output f r o m a filtration run was a shortened file containing o n l y the nondiscarded text, w h i c h was then either refiltered or scored using another dictionary and set o f rules. For scoring runs, i n d i v i d u a l paragraphs were evaluated as favoring a polled position i f there was a w o r d cluster referring to support for that position. For a paragraph w i t h more than one scored cluster, the score was d i v i d e d among the corresponding positions. The dispatch scores for the scored positions were the sums o f the individual paragraph scores. Because every scored paragraph added to the infon content score, a dispatch had a higher salience i f i t transmitted more relevant material (Chapter 1). The text analyses for the six cases in this book are sketched below. 1

Copyrighted material

Computer

4.4

T E X T A N A L Y S E S FOR

Text A nalysis

49

DEFENSE S P E N D I N G

Stories o n this topic were processed through t w o successive f i l t r a t i o n steps before the scoring r u n . I n Appendix C, one o f the retrieved dispatches for defense spending is traced through all the steps to illustrate the method. That dispatch also demonstrated that text further than fifty words f r o m one o f the search words (Chapter 3 ) was usually irrelevant to the polled topic. The text analysis for defense spending began w i t h a filtration step using the c r i t e r i o n that the dispatch concern A m e r i c a n defense spending. Therefore, the requirement was that the dispatch have at least one w o r d each referring to America, defense, and spending. I n addition, entire dispatches were discarded i f they had the words " f u n d " or " a i d . " " F u n d " ( w i t h no trailing characters like funds, funding, etc.) was f o u n d in articles on topics like the Environmental Defense F u n d . " A i d " (not aiding or aided) tended to refer to American aid for defense spending in another country. A f t e r this dispatch filtration step, the 692 dispatches were reduced to 377, w i t h 199 words per average dispatch. After the previous step removed entire dispatches irrelevant to American defense spending, a single c r i t e r i o n was used for the next filtration step. O n l y paragraphs

directly concerned with defense spending were retained. This was accomplished by

requiring that a w o r d referring to defense be close to a w o r d connoting spending. Since the previous filtration step had already discarded dispatches o n non-American defense spending, there was no need to require a reference to the United States i n this second filtration. Some spending words d i d not refer directly to money. For instance, " c u t " and "side" almost always i m p l i e d funding and were therefore considered to be spending w o r d s . The w o r d "side" is perhaps a surprise, but i t was f o u n d i n phrases l i k e "defense side" in budgetary discussions. The ability to use a multipurpose w o r d l i k e "side" i n this very specific fashion was due t o the disambiguation performed d u r i n g

both the initial Nexis search (Chapter 3 and Appendix B) and the previous filtration step.

A f t e r the second filtration, all but 37 o f the 377 stories retained f r o m the previous filtration were f o u n d to contain pertinent paragraphs. However, f o r each dispatch, many paragraphs concerned other subjects since the text w i t h i n each article was reduced to 27 percent o f the original. A f t e r the t w o filtration steps, the r e m a i n i n g paragraphs were s u f f i c i e n t l y homogeneous that the computer could easily score for infons i n favor o f more, same, and less defense spending. I n this scoring, the first condition was the removal o f all phrases referring to "non -defense." T h e n , t w o alternative sets o f dictionaries and rules were used. I n one, the dispatches were scored for support for all three o f the positions (more, same, and less defense spending). I n the alternative scheme, the scoring was o n l y for support o f more and less spending. The t w o scoring schemes differed i n both their dictionaries and their rules. I n the scoring f o r three positions, o n l y one c r i t e r i o n was used to score a paragraph as favoring more, same, or less defense spending. T h a t c r i t e r i o n was chat a defense w o r d be i n close p r o x i m i t y to m o d i f i e r words w h i c h favored one o f the positions. Naturally, the orders and distances between modifier words were used to fine tune the sense o f m o d i f i c a t i o n . Since the previous filtration step had already required that defense words be close to spending words, there was no need to examine both spending and defense i n the final scoring step. This step was performed on 5-10 percent o f the words i n the o r i g i n a l retrieved text. A m o n g the 377 stories t r u l y relevant to defense spending, 7 2 percent had at least one scored paragraph. Therefore, the scoring was not based on o n l y a small number o f dispatches. The texts scored for three positions as just described were rescored for the support o f only the t w o positions o f favoring more or less spending. I n the two-position scoring, the rules were s i m p l i f i e d by removing the category o f infons favoring same defense spending. Thus, i n f o r m a t i o n was forced into t w o

Copyrighte

SO

Predictions

of Public

Opinion from the Mass

Media

categories, that favoring more and that favoring less spending. T h e result was that words l i k e "keep" and " m a i n t a i n , " w h i c h referred clearly to same spending, were deleted f r o m the dictionary. A l s o , some words l i k e "freeze" were reassigned f r o m supporting same spending to favoring less spending. Besides the changes dictated by decreasing the scoring categories, the dictionary was further altered. Some o f the words f a v o r i n g m o r e ( " b o l s t e r " ) and less ("alternate," "weaken," " w i t h o u t ") were deleted. A l s o , nuclear arms reduction was interpreted to favor less defense spending. I n the three-position analysis described earlier, nuclear arms reduction had no implications for defense spending. T h u s the words "nuclear" and " a r m s " were only present i n the dictionary for the two-position analysis. This inclusion o f arms reduction meant that eight more dispatches were scored as favoring more or less defense spending. Therefore, the final number o f dispatches w i t h at least one paragraph w i t h a positive score was 280, or 74 percent o f the relevant 377. B o t h the i n i t i a l N e x i s retrievals and the subsequent text analyses focused o n information directly relevant to defense spending. However, i t was also possible that indirect messages c o u l d influence o p i n i o n on this topic. G o o d candidates for such indirect persuasive information were stories o n defense waste and fraud. Therefore, a separate set of A P dispatches was retrieved w i t h text that had a w o r d connoting defense close to a w o r d referring to waste, fraud, or corruption (Appendix B ) . There was no requirement that the story be about defense spending. O f the 878 identified from January 1, 1977 to A p r i l 12, 1986, 512 were retrieved at random for text analysis. The harvested text was passed through t w o filtration steps, the first discarding entire dispatches i f they were about waste and fraud in countries other than the U n i t e d States. T h e second filtration retained those paragraphs w i t h w o r d combinations implying both defense and waste, fraud, or corruption. At this point, every remaining paragraph was scored as favoring less defense spending i f i t contained a w o r d cluster u n i t i n g the ideas o f defense and waste, corruption, or fraud. The implication was that such stories w o u l d suggest that funds for the military were unwisely spent and should therefore be reduced. O f the original 512 retrieved dispatches, 147, o r 29 percent, were scored as having an average o f 1.3 paragraphs each on defense waste and fraud.

4.5 T E X T A N A L Y S I S FOR TROOPS I N

LEBANON

The text for troops i n Lebanon was the most complex o f the six examples. Nevertheless, o n l y t w o filtration runs were needed before the final scoring step. Since essentially none o f the retrieved dispatches were t o t a l l y irrelevant t o the topic, the first filtration focused on selecting relevant paragraphs rather than e l i m i n a t i n g entire dispatches. The m a i n criterion was that paragraphs s h o u l d be d i r e c t l y concerned w i t h A m e r i c a n troops i n Lebanon. Therefore, the retained paragraphs had to have words mentioning America, troops or policy, and Lebanon. I n a d d i t i o n , backup conditions e l i m i n a t e d paragraphs o n other topics l i k e the domestic American economy or troops f r o m other countries. The rules for the Lebanon analysis permitted ideas to be carried forward from one paragraph to the next, so that pronouns l i k e " t h e y " and " t h e m " referred to troops i f troops were mentioned i n an earlier paragraph. S i m i l a r l y , the understanding was that a story c o n t i n u e d t o be about Lebanon i f that area o f the w o r l d was mentioned p r e v i o u s l y unless there was a change o f locale to places l i k e Grenada or the Caribbean, since the U n i t e d States also had troops i n Grenada during part o f the same time period. I n addition, a new Lebanon reference was needed when words l i k e " A r a b , " "Christian," and "Druse" were found i n the previous paragraph. America was also carried forward as a descriptor until mention o f another country such as B r i t a i n , Italy, and France, w h i c h were also part o f the M u l t i n a t i o n a l Force stationed i n Lebanon in 1983-84.

Copyright

Computer

Text Analysis

51

While references to America and Lebanon continued to be carried forward until a paragraph had a w o r d to cancel the thought, the troop idea was o n l y carried forward i f the next paragraph actually had an appropriate pronoun. I n this first filtration step, the text on Lebanon was reduced to 31 percent o f the o r i g i n a l retrieval. However, the number o f dispatches w i t h at least one relevant paragraph only dropped by 11 percent. As a result, the average dispatch now had 156 instead o f 420 words. T h e second filtration step was influenced by the requirement o f retaining words like "keep" and "support" i n the final scoring dictionary. These words were needed because the scoring was for the positions o f favoring more, same, or less troops i n Lebanon. H o w e v e r , such words were also p l e n t i f u l i n text describing m i l i t a r y c o m b a t For instance, sentences sometimes spoke o f artillery support for g r o u n d forces. T o avoid confusion, i t was decided to remove a l l text o n actual combat and all paragraphs on entertainer Bob Hope's Christmas v i s i t to Lebanon. T h e remaining paragraphs directly advocated positions on the deployment o f troops i n Lebanon. In this step, the i m p l i c i t decision was that descriptions o f combat c o u l d be ignored. Such descriptions were not logically associated w i t h the need to send either more or less troops. I f the combat situation was unfavorable, the public c o u l d either feel that more troops were needed to correct the problem or that troops should be w i t h d r a w n because the situation was hopeless. T h i s second filtration step reduced the text to 15 percent o f the o r i g i n a l . The dispatches w i t h at least one relevant paragraph o n l y decreased to 80 percent o f the original. The remaining text o f each story averaged eighty-five words, quite similar to the eighty-one found after the t w o filtration steps for defense spending. After the t w o filtration runs for troops i n Lebanon, the text was scored for favoring more, same, and less troops. The major criterion was to score for a w o r d referring to troops o r p o l i c y near a m o d i f i e r w o r d connoting more, same, or less. Some w o r d s - l i k e "stay and " w i t h d r a w " - w e r e able by themselves to favor keeping or removing troops since the concepts o f America and Lebanon were already present i n all dispatches. The previous filtration step had assured that the remaining paragraphs advocated policy positions o n American troops in Lebanon. Therefore, as for defense spending, the previous filtration steps simplified the final scoring. Even after the filtrations, some words retained important ambiguities. One o f the most notable was "peacekeeping." T h i s w o r d c o u l d occur i n "peacekeeping force," a neutral statement regarding troops* whereas elsewhere i n the text there w o u l d be phrases l i k e "peacekeeping role," w h i c h tended to support a continued troop presence. This problem was overcome by first asking i f "peacekeeping" preceded "force" w ithin a short distance. I f so, then "peacekeeping" and "force" gave the concept o f "troops." Otherwise, "force" was ignored because i t was used as a verb connoting coercion a significant amount o f the time. I f "peacekeeping" was not before "force," then i t c o u l d act on a policy w o r d to g i v e the sense that troops should be kept i n Lebanon. A different strategy was used to assign the proper meaning to "state" (as i n "he stated that"), a w o r d w i t h policy implications. Since "state" was also often found i n the phrase "secretary o f state," a dictionary search was first made for the pair o f words " o f state." Once this w o r d pair was found and removed, "state" c o u l d refer to policy. Policy o n troop deployment was frequently qualified by conditional words such as " i f , " "question," "reappraise," and " w h y . " Since the majority o f the debate was between troop retention and troop removal, such a qualifier led to paragraphs having equal scores favoring the t w o positions. U s i n g these main conditions, the text after the t w o filtration steps was scored for favoring the three positions o f more, same, and less troops. O f the o r i g i n a l 467 dispatches, 54 percent had at least one paragraph w i t h a non-zero score, again indicating that the scores were not based on a small minority o f the total dispatches. 5,

Copyrighted material

52

Predictions

of Public

4.6 D E M O C R A T I C

Opinion from

the Mass

Media

PRIMARY

F r o m the p o l l data for the Democratic p r i m a r y ( A p p e n d i x B), people w i t h opinions were d i v i d e d into three groups, those favoring Móndale, Glenn, and Others. Therefore, the text was scored for these three groups o f candidates. T w o different text analyses were used. One was based on what m i g h t be called " b a n d w a g o n " words describing a candidate's successes and failures. As noted by Brams and Riker (1972) and Straffin (1977), i t is possible to model the bandwagon effect mathematically assuming that uncommitted persons recognize the w i n n i n g side and adopt the traits o f that side w h i l e shunning the losers. However, the public must somehow assess w h i c h sides are w i n n i n g and losing. For political candidates, the analysis i n this book assumes that these perceptions are formed f r o m descriptions i n the press o f the candidates' successes and failures. The bandwagon effect i m p l i c i t l y assumes that the decision to support a candidate is made p r i n c i p a l l y f r o m these perceptions rather than on a thoughtful analysis o f other pertinent factors, such as a candidate's stand on issues. Therefore, the bandwagon words used in this book refer only to candidates' successes and failures. The alternative text analysis used candidate name counts and was predicated o n the idea that name recognition, regardless o f context, was the crucial consideration for a candidate's p o p u l a r i t y ( K i n d e r and Sears, 198S; M u e l l e r , 1970a; Stang, 1974; Zajonc, 1968, 1980). The more complicated analysis was obviously the one dependent on bandwagon w o r d s . Here, u n l i k e the cases o f defense spending and troops i n Lebanon, the positions d i d not lie on a continuous scale ranging f r o m one extreme to another. Instead, i n f o r m a t i o n for each candidate c o u l d be placed o n some scale f r o m very favorable to very unfavorable. Rather than choosing a very Finely graded scale, the text was simply j u d g e d to be favorable or unfavorable to a candidate. In the bandwagon analysis, the first step was a filtration to select for paragraphs c o n t a i n i n g the last n a m e o f at least one o f the Democratic candidates. T h i s step reduced the text to 55 percent o f the o r i g i n a l . A l l the dispatches still had at least one paragraph w i t h a candidate name. After this reduction, 199 words were left i n the average dispatch. After this filtration, paragraphs were scored as favorable and unfavorable to all three groups o f candidates. The m a i n c r i t e r i o n was that a candidate's name should appear near an advantageous or disadvantageous cluster o f modifier words. These were bandwagon w o r d clusters i n that no attention was paid to the reasons for an o p i n i o n on a candidate. The i m p l i c a t i o n was that the public favored a candidate i f it looked as i f that candidate was getting support f r o m some source, whatever the reason. The most c o m m o n o f these bandwagon words i n the retained text were "elected" (found 112 times), "backed" (found 94 times), and "favored" (found 73 times). There were also negative words, o f w h i c h " f a i l e d " ( f o u n d 24 times), " w e a k " ( f o u n d 16 times), and "vulnerable" (found 11 times) were among the most frequent. T h i s bandwagon scoring approach ignored the positions and activities o f the candidates. For instance, Jesse Jackson was frequenüy i n the news during the period studied because he was negotiating the release or a navy pilot shot d o w n during a raid over Lebanon. Also, there were reports o f A l a n Cranston supporting the idea o f a nuclear freeze. These and other reports o n the candidates' activities and stands on issues were consciously omitted during the bandwagon analysis. Since candidate activities and positions were not scored, a majority o f the articles had zero scores favorable or unfavorable to the candidates. Nevertheless, 37 percent o f the o r i g i n a l stories d i d have positive scores for at least one o f the six possible positions. As an alternative to the bandwagon text analysis, it was possible to imagine that candidate name use was sufficient to determine popularity, so the same dispatches were analyzed for their mentions o f candidate names. Paragraphs w e r e scored as m e n t i o n i n g Móndale, G l e n n , o r others since distinctions were not made for the contexts i n w h i c h names were mentioned. W i t h

Copyrighte U

Computer

Text Analysis

S3

all names being counted and no context being scored, all 425 dispatches had at least one paragraph i n support o f a candidate.

4.7

T E X T A N A L Y S I S FOR T H E E C O N O M I C C L I M A T E

The text analysis for this topic was performed i n three steps. The first was to keep o n l y those stories o n the U n i t e d States, the second was to discard paragraphs w h i c h were not about the economy, and the third was to assign the numerical scores. Sixty-six percent o f the o r i g i n a l dispatches had at least one paragraph scored as supporting one or more o f the three positions, w i t h many dispatches having m i x e d messages.

4.8

T E X T A N A L Y S I S FOR U N E M P L O Y M E N T V E R S U S I N F L A T I O N

For the text analysis for this topic there was o n l y one filtration step i n w h i c h dispatches on non-American economies were eliminated. Then the paragraphs were scored for the relative importance o f unemployment and inflation using the criterion that words for these concepts should be close to modifier words suggesting their importance. Forty-four percent o f the original dispatches had at least one paragraph taking a position o n this issue.

4.9 T E X T A N A L Y S I S FOR C O N T R A A I D The text analysis for this topic was developed w i t h o u t k n o w i n g the results of public o p i n i o n polls to assure that the analysis c o u l d not be influenced by p o l l results. A t the beginning o f the analysis, the Roper Center was contacted at the University o f Connecticut to determine that a sufficient number o f polls existed to obtain a reasonable time series. The p o l l data were received after the end o f the analysis. This topic also tested the ability o f persons besides the author (Fan) to construct the dictionaries and rules for the text analysis. W o r k i n g as a team, three graduate research assistants ( S w i m et a l . , see A p p e n d i x C ) f r o m the D e p a r t m e n t o f Psychology at the University o f Minnesota constructed the dictionaries and rules for both the filtration and scoring steps. S w i m et al. first devised a filtration step to keep o n l y those paragraphs w i t h words connoting the United States, aid, and Nicaragua. The reference to the United States was required because the polls were for American o p i n i o n o n American aid to the Contras. A t this step, no attempt was made to separate a i d to the Contras opposing the Nicaraguan government from aid to the government itself. After this filtration, S w i m et al. decided that the text was ready for final scoring. Since the scoring is t y p i c a l l y the most subtle step, rules and dictionaries were developed independently both by Fan and S w i m et a l . The text s u r v i v i n g the filtration was then analyzed twice, once using the procedures o f Fan and once using the method o f S w i m et al. Fan used the same general strategy employed for the other analyses o f this book, insisting that w o r d combinations favoring or opposing aid be i n close juxtaposition w i t h words referring to the Contras or a synonym. The result was paragraph scores either favoring o r opposing aid, or both. The result o f supporting both positions c o u l d either be due to t w o w o r d clusters, one opposing and one favoring aid, or to conditional words l i k e " i f appearing i n paragraphs w i t h only one w o r d combination favoring either idea. U n l i k e Fan, S w i m et al. counted many w o r d clusters only indirectly favoring or opposing aid. For example, "talks"..."failed" was interpreted to i m p l y that aid should be given. I n another instance, "Schultz"..."over eager" was interpreted to mean that no aid should be given. Furthermore, there was no requirement that any o f these

Copyrighted material

54

Predictions

of Public

Opinion from

the Mass

Media

words be close to a m e n t i o n o f the Contras, taking advantage o f the fact that the filtration had already restricted the paragraphs to those discussing America, aid, and Nicaragua. B y counting many indirect statements w h i l e Fan d i d not, S w i m et al. f o u n d that 906 o f the 969 retrieved dispatches had paragraphs relevant to Contra aid. In contrast, the number for Fan was 7 7 0 . Therefore, Fan d i d not score 15 percent o f the stories w h i c h S w i m et a l . f o u n d to be relevant These stories presumably had only indirect statements o n Contra a i d . I n addition, S w i m et a l . found 3.4 paragraphs per average scored dispatch to be relevant t o Contra aid w h i l e Fan o n l y f o u n d 2,0. r e l e v a n t Therefore, by using a different dictionary and a set o f rules i n c l u d i n g a significant number o f indirect statements, S w i m et a l . gave scores to almost twice as many paragraphs as Fan. Fortunately, both text analyses gave essentially the same o p i n i o n projections (Chapter 5). This result w i l l be discussed further in Chapter 6.

4.10

S U M M A R Y FEATURES FILTRATIONS

OF

TEXT

ANALYSIS

BY

SUCCESSIVE

One o f the obvious strengths o f the InfoTrend text analysis was that each issue was probed w i t h a customized dictionary and set o f rules. Therefore, i t was possible to have m u c h greater specificity and flexibility than was possible w i t h general programs using fixed dictionaries and rules. T h e necessity o f different c r i t e r i a was clear f r o m the poor overlap i n the dictionaries and rules for the six examples i n this book. For troops i n Lebanon, proper nouns l i k e Lebanon, Beirut, C h r i s t i a n , and Shiite were c r i t i c a l . F o r the Democratic p r i m a r y , another set o f proper nouns, the names o f the presidential candidates, were crucial instead. A m o n g the few words appearing consistently i n all analyses were those referring t o the U n i t e d States and words l i k e " n o " and " n o t " connoting negation. The specificity o f the analysis also meant that a single story could be relevant to a number o f different issues. Thus the same dispatch c o u l d be analyzed w i t h t w o different sets o f dictionaries and rules to provide t w o different types o f infon scores. For example, some o f the articles identified as relevant to the economic climate were probably also pertinent to defense spending and/or unemployment versus inflation. Those dispatches w o u l d have yielded different types o f scores depending on the issue being studied. The same article c o u l d have had paragraphs both f a v o r i n g more defense spending and supporting an improving economic climate. Another advantage o f the method o f successive filiations is its generality. I t may seem contradictory t o speak o f the generality o f the method when the advantage discussed i n the previous section was specificity. However, the method was general due t o the o v e r a l l strategy o f progressively c u l l i n g irrelevant text i n a series o f filtration steps. I n a d d i t i o n , the I n f o T r e n d text analysis benefited f r o m the use o f s i m p l e dictionaries and rules. T h i s was made possible by using t w o sequential f i l t r a t i o n steps i n w h i c h t w o distinct criteria could be separately dev eloped to remove irrelevant text. The resulting simplifications i n the rules m i n i m i z e d l o g i c a l errors i n their construction. The simplifications were possible due to the ability o f filtration steps to relate thoughts w i t h i n the text to each other. For defense spending, for instance, the first f i l t r a t i o n imposed the c o n d i t i o n that all dispatches be about the U n i t e d States. Therefore, i n subsequent steps, the concept o f A m e r i c a was already i m p l i c i t i n the remaining t e x t There was no need to include a reference to the U n i t e d States in the second filtration step selecting for paragraphs directly speaking o f defense spending. In consequence, words referring t o the United States could have been anywhere in the text i n the first filtration and totally absent i n the second. A single set o f rules for a one-pass program w o u l d have been m u c h more complex than the t w o individual sets o f rules combined. The rules for a one-pass run w o u l d have had to have been able to

Copyrighted material

Computer

Text Analysis

55

retain paragraphs on defense spending when there was also reference to America either earlier or later i n the dispatch. The filtrations permitted the use words w i t h unusual meanings. Very general words like " c u t " and "side" c o u l d safely be used i n special senses (like spending) i f previous filtration steps put such code i n t o the proper contexts. Also, the specific, unusual, and j a r g o n - l i k e nature o f many words was revealed o n l y as the text became more homogeneous. Therefore, the text filtration was very useful for w o r d disambiguation. The keys to handling code words were (1) the strategy o f text filtration before final scoring and (2) the tailoring o f the rules and dictionaries to i n d i v i d u a l situations. The use o f code words is much more problematical for general text analysis programs where the same program and dictionary is used for a wide variety o f t e x t Contrary to what might be expected, the t i m e required to analyze the text was decreased rather than increased by the filtration steps, even though some portions o f text were reanalyzed several times. This economy occurred at three steps. The first was i n the construction o f the dictionaries and rules. T y p i c a l l y , after examination o f 150,000-200,000 characters o f text f r o m random dispatches, s u f f i c i e n t repetition was seen that most o f the relevant rules and dictionary words could be extracted. The time needed for obtaining a workable dictionary and rules was shortened by using only one or t w o criteria for each filtration r u n . After the f i l t r a t i o n , the text was more homogeneous and more dispatches were represented for a given number o f characters o f text since significant amounts o f irrelevant text were discarded. Therefore, the text for constructing the dictionaries and rules was derived f r o m more and more dispatches as the files were filtered. This had the advantage that the most subtle decisions, those used for the scoring, were made from large samples o f relevant text, more than was practical w i t h the u n f i l t e r e d stories, where the density o f pertinent material was low. I t is obviously more time-efficient to make rules f r o m text h i g h l y p u r i f i e d for relevance. Also, w i t h fewer rules and smaller dictionaries, less time was needed to debug the logical errors i n the rules. The strategy of using simple dictionaries and rules also resulted in shortening the computer runs themselves. The i n i t i a l filtrations usually e m p l o y e d the simplest rules and the smallest dictionaries, so the runs were quite r a p i d . W i t h successive filtrations, the amount o f text d i m i n i s h e d so that more complex rules and larger dictionaries d i d not lead to significant increases i n time. U s i n g a one-pass computer program, the rules w o u l d have been more complex than those used at any one step and the dictionaries w o u l d have been larger, thereby s l o w i n g the runs. I n other words, a great deal o f computer time w o u l d have been spent carefully examining irrelevant text. The InfoTrend strategy is to tailor each analysis, at a l l steps, to the specific question being asked. Consequendy, no additional time was required to interpret the outputs o f the computer runs. A s described in this chapter, the final scoring runs gave the i n f o n content scores i n terms o f the number o f paragraphs supporting i n d i v i d u a l positions. These scores were then used directly for calculating public opinion w i t h o u t further interpretation. The situation is quite different for analyses using the predefined dictionaries and w o r d categories o f general text analysis programs. For those programs, no t i m e is spent i n the construction o f the dictionaries and rules. Instead, t i m e is used i n interpretation o f the data generated, and that time can be considerable. Sometimes, as for the examples i n this book, there is even no obvious way to interpret usefully the output o f a general text analysis program like M C C A . The methods i n this book were also designed so that the logic o f every step was direcdy related to the initial goal o f each analysis. The customized nature o f each step meant that i t was possible to make the positions much more specific than was possible using the predefined categories i n general text analysis programs. However, even w i t h such general programs, it is sometimes possible to extract themes not present i n individual predefined dictionaries by performing factor analyses based on separate runs using different dictionaries. For example, Weber (1985) used

successive text

also

of

words

Copyrighte U

56

Predictions

of Public

Opinion from

tfie Mass

Media

this technique to show that editorials w i t h frequent reference to American leaders and topics w i l l also have a high content o f words relating to h i g h status and occupations. A l t h o u g h factor analysis is a powerful procedure, i t is difficult to consider its greater indirectness to be an advantage. Another important goal i n developing the InfoTrend methods was to o b t a i n a system where i t was very d i f f i c u l t to fit the text analysis to p o l l data, either consciously or unconsciously. There is no easy way to adjust the I n f o T r e n d text analysis w i t h the purpose o f matching o p i n i o n percentages. Adjustments i n the content analysis w o u l d have i n v o l v e d changing the rules o r the d i c t i o n a r i e s . However, any such changes w o u l d have affected the analysis for all AP dispatches. Therefore, there was no g o o d w a y o f foreseeing the effects o n the o p i n i o n projections. A n y changes i n the dictionaries and rules for the text analysis also had to be justifiable as logical consequences o f the meanings w i t h i n the text. There was no place in the programs to add arbitrary correction factors. Furthermore, the dispatches were examined by design in random order during the constructions o f the dictionaries and rules. Therefore, i t was never obvious w h i c h dispatch scores needed to be changed i n order to get a better o p i n i o n projection. I n fact, for no analyses were o p i n i o n p o l l data examined during the text analysis steps. Therefore, the devising o f the dictionaries and rules proceeded without regard to poll data for all analyses even though i t was o n l y for the Contra aid example that those data were not available until the analysis was finished. Furthermore, the dispatch scores went through additional mathematical manipulations before the final o p i n i o n projections were obtained (Chapter 5). I t was very d i f f i c u l t to guess the final shapes for the opinion projection curves f r o m the raw infon content scores. The quantitative predictions required calculations using the projection computer program. A l l these considerations meant that the text analyses had to be performed independently o f the opinion projections based upon them.

4.11 E X T E N S I O N S O F T H E T E X T A N A L Y S I S P R O C E D U R E The computer content analysis i n this book should be applicable to texts other than those used for calculating public o p i n i o n . For instance, responses to open-ended questionnaires can be examined for their comments o n different topics. Employee letters of recommendation might be scored quantitatively for support of different traits desired for a j o b . Different analyses w o u l d require different dictionaries and rules, but the overall strategy is broadly applicable using the InfoTrend text analysis procedures. The strategy described in this chapter required that all the dictionaries and rules be formulated by the investigator. There are n o provisions for machine refinement. as more experience is gained w i t h the sort rules w h i c h are admissible and the types o f words w h i c h f i t i n t o various classes, software might be w r i t t e n w h i c h w ill permit the computer to a i d i n the development o f the dictionaries and rules. The machine m i g h t i n i t i a l l y be assigned tasks such as l o o k i n g for similar words to add to a dictionary. Another computer function might be to check for consistency whenever new words and rules are added to a preexisting dictionary and accompanying set o f rules. Later, as the guidelines for dictionary and rule constructions become clearer, these guidelines m i g h t be included i n computer programs to suggest both new rules and dictionary words. As more and more such guidelines are included, the procedure might gradually be converted from a f u l l y manual system to one w i t h ever greater degrees o f automation in dictionary and rule development.

However,

of

Copyright

5

Projections of Public Opinion

The first chapter i n this book o u t l i n e d ideodynamics and s h o w e d h o w this mathematical model c o u l d be used t o calculate p u b l i c o p i n i o n f r o m persuasive messages available to the population. The third chapter proceeded t o describe six sets of data used for the computations. Those data were time series o f public o p i n i o n polls and Associated Press dispatches relevant t o the p o l l e d issues. T h e fourth chapter then presented a new computer method for scoring A P dispatches f o r their support o f various p o l l e d positions. The present chapter details the use o f these scores to compute percentage values for public o p i n i o n throughout the time interval for w h i c h p o l l data were available. O p i n i o n is calculated for all six o f the cases examined. Since many o f the examples point to the same general conclusions, the ramifications o f the individual studies are not discussed here, but i n Chapters 6 and 7. As o u t l i n e d i n Chapter 1 a n d discussed further i n A p p e n d i x D , the o p i n i o n calculations i n v o l v e d three steps: ( 1 ) conversion o f the A P dispatch scores i n t o persuasive force functions appropriate f o r the computations, ( 2 ) construction o f detailed population conversion models for the effect o f persuasive forces o n various subpopulations, and (3) calculations o f expected p o l l results. These steps are n o w considered in detail for the six issues analyzed.

5.1 O P I N I O N P R E D I C T I O N S FOR D E F E N S E S P E N D I N G A P stories were collected for defense spending and processed to obtain numerical scores for each o f the stories retained. T w o sets o f scores were obtained f o r each story. For one set, the dispatches were scored for their support o f the three positions o f more, same, and less spending. For the other set, the s t o r i e s w e r e scored for their support o f only the t w o positions o f more and less spending. B o t h sets o f scores were then used t o c o m p u t e persuasive force functions describing the time trends o f the persuasive forces favoring each o f the scored positions. For every position, a separate persuasive force function was computed. When there were three sets o f scores, there were three functions favoring the three positions o f more, same, and less spending (Figure 5.1). S i m i l a r l y , t w o sets o f scores yielded t w o functions (Figure 5.2). T o compute these functions, a curve like that described i n Chapter 1 (Figure 1.1, t o p frame) was computed f o r each i n f o n . Then a l l the functions f o r individual infons favoring the same position were added together as shown i n Figure 1.1 ( b o t t o m frame). T h e additivity meant that opinion reinforcement a n d i n f o r m a t i o n saturation were ignored (Chapter 1). W h e n the resulting functions were plotted (Figures 5.1 and 5.2), the steep rises followed by the

i

^> 1

58

Predictions

of Public

Opinion from

tlic Mass

Media

gradual declines i n Figure 1.1 were compressed into spikes due to the condensed time scale o f seven years. The separate sets o f persuasive force functions belonging t o the t w o sets o f infon scores were used independently for opinion calculations. For both sets o f functions, i t was assumed that A P dispatches c o u l d represent all the relevant information available to the public, that all scores had the same weights, that the score for a position only contributed to the persuasive force for that position, and that the persistence constant had a one day h a l f - l i f e . The major difference between the t w o sets o f plots was the deletion o f infons favoring same spending. M o s t o f those infons were partitioned between the upper and lower curves o f Figure 5.1. T o use the persuasive force functions, i t was necessary t o devise a population conversion model as was already done for the defense spending analysis (Figure 1.2). When used for infons scored for more, same, and less spending, all three persuasive forces were used i n the calculations. W h e n the same model was applied to content scores for infons favoring o n l y more and less spending, the persuasive force function favoring same spending was always zero. The ideodynamic equations corresponding to the population conversion model were then w r i t t e n (Chapter 1 and Appendix A ) . T o use these equations, i t was first necessary to set their parameters: the persistence half-life, the modified persuasibility constant, and any required r e f i n i n g weights. I n i n i t i a l trials, the r e f i n i n g weights were all set to 1.0 corresponding to the approximation that all i n f o n content scores had the same weight. This was a safe strategy because the refining weights usually differed very little f r o m 1.0. T o set the other constants, o p i n i o n time trends were calculated using arbitrary values for the persistence half-life and the modified persuasibility constant. A t each t i m e corresponding to that o f an actual p o l l , deviations w e r e computed between calculated opinions and the actual values starting w i t h the measured opinions at the time o f the first p o l l (Appendix A , Equation A . 2 6 ) . The squares o f these deviations were calculated for all p o l l values, and averaged to g i v e the Mean Squared D e v i a t i o n ( M S D ) . The chosen persistence and m o d i f i e d persuasibility constants were those g i v i n g the m i n i m u m M S D . R e f i n i n g weights different f r o m 1.0 were o n l y tested i f the predictions were systematically h i g h or l o w for one or more o f the o p i n i o n positions. I f a r e f i n i n g w e i g h t gave a s i g n i f i c a n t i m p r o v e m e n t i n the M S D , then the persistence and modified persuasibility constants were reoptimized for the new r e f i n i n g w e i g h t s ) . The f i n a l constants were those g i v i n g the least M S D for a l l constants, unless otherwise stated. Rather than s i m p l y c o m p u t i n g the M S D f o r every set o f trial constants, t i m e trends o f o p i n i o n projections resulting f r o m a number o f arbitrary values for the constants were t y p i c a l l y plotted to examine q u a l i t a t i v e l y the consequences o f decreasing particular constants. Based on these plots, i t was clear that persistence half-lives m u c h longer than a day usually meant that the p o p u l a t i o n w o u l d have responded to media information more s l o w l y than was actually f o u n d . A l s o , as the m o d i f i e d persuasibility constant increased, the fluctuation i n o p i n i o n calculations became larger and larger around a general time trend. T h i s result was expected since a larger persuasibility constant corresponds t o more v o l a t i l e issues, w i t h more people being persuaded for the same amount o f information. G i v e n these q u a l i t a t i v e observations, systematic trials for the persistence constant t y p i c a l l y started w i t h a one day h a l f - l i f e . T h e n , additional values were tested, increasing iteratively by factors o f t w o until M S D values five to ten times the M S D at the one day half-life were reached. Test values for the persuasibility constant usually began w i t h values very close t o zero (e.g. 0.001 per A P paragraph per day), corresponding to a population being impervious t o persuasion, and then increased i n t w o f o l d steps beyond the value for w h i c h the m i n i m u m M S D was reached. I n the region where different parameter values gave approximately the same M S D , values were tested on a finer linear scale between the t w o f o l d j u m p s . For all p a r a m e t e r s , the M S D was plotted against trial values o f the p a r a m e t e r s (e.g., Figures 5.4 and 5.11). For these plots, the values for the other parameters

increasing or

Copyrightec

Projections

of Public

Opinion

59

those g i v i n g the m i n i m u m M S D . I n this way, the reader can assess the sensitivity o f the M S D to changes i n the t r i a l parameters. Where the optimization curves are steep, in the range o f a m i n i m u m , the M S D is very sensitive to parameter changes. Where the curves are shallow, w i d e variations i n the parameters w i l l have relatively s m a l l effects o n the M S D . I t is possible to read f r o m the o p t i m i z a t i o n curves the amount o f permissible variation i n a parameter before the M S D increased beyond the tolerance limits set by the analyst. The values discussed i n the remainder o f this chapter w i l l refer t o the values o f the parameters at the m i n i m a o f the optimization curves. T u r n i n g away f r o m general strategy and t o w a r d the specifics o f the projections using infons scored for more, same, and less spending, i t was found that the optimal modified persuasibility constant was 0.6 per A P paragraph per day. Expected opinion throughout the time period o f message collection was calculated using this optimized constant, the i n i t i a l p o l l values, and the persuasive force curves i n F i g u r e S.l computed using the best value for the persistence half-life, one day ( A p p e n d i x A , Equations A.2o and A.29). The expected o p i n i o n is plotted together w i t h actual poll data i n Figure 5.3. The one day value for the o p t i m a l persistence half-life was f o u n d by p l o t t i n g t r i a l values for the h a l f - l i f e versus the M S D (Figure 5.4, l o w e r frame). This o p t i m i z a t i o n p l o t shows that the M S D was s t i l l decreasing as the h a l f - l i f e was shortened to one day. I t was conceivable that an even shorter half-life w o u l d have been appropriate. However, i t seemed unreasonable to set the half-life much shorter than one day since there was at least that much ambiguity i n the t i m i n g o f the p o l l points and i n the t i m i n g o f the A P dispatches. The optimization curve f o r the persistence constant d i d , however, show a second m i n i m u m over 100 days before a rapid increase above that t i m e . The precise explanations for the steep rise at long half-lives are not clear. The upper frame o f Figure 5.4 shows the o p t i m i z a t i o n for the m o d i f i e d persuasibility constant. This constant clearly gives a single, well-behaved m i n i m u m M S D at 0.6 per A P paragraph per day. Comparison o f the projections i n Figure 5.3 w i t h actual p o l l points did not indicate that scores favoring any position were either systematically too high or too l o w . Therefore, all refining weights were left at the value o f 1.0 used for o p t i m i z i n g the persistence and m o d i f i e d persuasibility constants. These values meant that all infons were given the same weight. For the projections o f Figure 5.3, o n l y 692 o f the 9,314 identified dispatches were studied. It was conceivable that smaller samples c o u l d g i v e estimates w h i c h were j u s t as good. Therefore, the 692 stories were d i v i d e d into t w o approximately equal, random subgroups o f 325 and 383 dispatches each. O n l y sixteen dispatches i n one group were also present i n the other. U s i n g the m o d i f i e d persuasibility constant optimized above, opinions were recalculated using the t w o dispatch subsets (Figures 5.5 and 5.6). N o t surprisingly, there were greater deviations between the predictions and the p o l l results w i t h the smaller sample sizes. These differences c o u l d be seen quantitatively by the increase f r o m 7.2 p o l l percent for the total dispatch set to 9.4 percent and 10.3 percent for the t w o subsets (Table 5.1). Yet another projection was made to explore what w o u l d have happened to a subpopulation comprised only o f those favonng more defense spending. Therefore, the full set o f infons and the optimized modified persuasibility constant and u n i f o r m weights were used to remake the projections assuming that the i n i t i a l population only had people supporting more spending (Figure 5.7). The calculation snowed that after three years, the subpopulation should have behaved much as the population as a whole. T h i s result was significant technically because it meant that there was no need to account for the statistical errors inherent to the first p o l l point. The calculations u l t i m a t e l y h o m e d to the values dictated b y the i n f o r m a t i o n structure. The calculations for later times were not adversely affected even by extremely inaccurate i n i t i a l conditions. I f the errors i n the i n i t i a l p o l l p o i n t were not large, then the achievement o f the proper values w o u l d have occurred much more rapidly. were

Copyrighted material

60

Predictions

of Public

Opinion from

the Mass

Media

So fax, the discussion has concerned o p i n i o n calculations made f r o m infons scored for the three positions of more, same, and less defense spending. In addition, projections were also made for infons scored only for more and less spending (Figure 5.8). F o r these projections the data were f r o m all 692 dispatches. T h e same modified persuasibility constant o f 0.6 p o l l percent per A P paragraph per day was used for a l l infons so that the projections c o u l d be compared directly. A g a i n , all scores were given the same weight. The square root o f the M S D ( R M S D ) was 8.3 poll percent for the t w o - p o s i t i o n model i n contrast to 7.2 percent for the threeposition model. Therefore, the two-position scores gave slightly less accurate results than the three-position scores. I n the calculations f r o m both the t w o - and three-position scores, the projected time trends o f public o p i n i o n appeared to move i n steps since the time between infons was usually large relative to the week d u r i n g w h i c h an infon had its e f f e c t This was reasonable since o n l y 7 percent o f the total identified dispatches were studied. On an expanded scale, each step w o u l d have had the shapes in Figure 1.3. The steps were also much less tall because the modified persuasibility constant had a value over 3,000 times smaller i n Figures 5.3 and 5.5-5.8, than i n Figure 1.3. U s i n g infons scored for either two or three positions, the time courses for all three opinions favoring more, same, and less spending followed quite w e l l the main features o f the actual poll data. The change was most dramatic for people favoring more spending. There was a dramatic rise in o p i n i o n from 1979 to 1980 and an equally impressive drop f r o m 1981 to 1982. B o t h the timing and the magnitudes o f the actual changes were mirrored i n the calculated opinion. Comparison o f the opinion projections w i t h the i n f o n force curves (Figures 5.1 and 5.2) showed that the rise i n opinion favoring more defense spending i n 1979 was due to the great increase i n i n f o r m a t i o n favoring this position. D u r i n g this t i m e there was no d i m i n u t i o n i n messages arguing for less spending. T h e subsequent drop in support for more spending was not due to the disappearance o f messages favoring this idea. There was instead a significant augmentation in opposing messages. Besides permitting the calculation o f the best m o d i f i e d persuasibility constant, the optimization curve for this u n k n o w n (Figure 5.4, upper frame) also shows that the projected values are much better than w o u l d have been predicted by the model that o p i n i o n had stayed constant throughout the p o l l i n g period. The c o n d i t i o n o f n o o p i n i o n change is equivalent to a very small value for the modified persuasibility c o n s t a n t W h e n this constant is zero, the population is completely resistant to i n f o r m a t i o n and w i l l never undergo o p i n i o n change regardless o f the presence o f persuasive messages. Therefore, a very l o w modified persuasibility constant such as 0.001 per A P paragraph per day gave three almost unchanging o p i n i o n curves throughout the seven year p e r i o d - f l a t plots corresponding to the curves i n Figure 5.3. F r o m the optimization curve o f Figure 5.4, the M S D was over 250 poll percent squared for k'2 - 0 . 0 0 1 . The corresponding value o f around 50 p o l l percent squared for the projection using the best persuasibility constant was less than 1/5 as large. This decrease i n the M S D meant that the ideodynamic f i t was m u c h better than the model of no opinion change. For comparison, 1,000 simulations were made for predicted p o l l values drawn at random. For each simulation, the M S D was calculated for the differences between actual p o l l values and r a n d o m , predicted p o l l points. T h e p r o b a b i l i t y that the ideodynamic predictions were no better than chance was ascertained by counting the number o f simulations among the 1,000 where the M S D f r o m random p o l l results was smaller than that f r o m the ideodynamic predictions (right-hand c o l u m n o f Table 5.1). F o r completeness, R M S D values were computed b o t h for the r a n d o m and ideodynamic estimates. The average value among the 1,000 simulations is also given i n Table 5.1 (second column). Because there were 1,000 independent draws, it was also possible to compute a standard deviation for the M S D , and f r o m this standard deviation, the number o f standard deviations f r o m the random M S D to the ideodynamic estimate (Table 5 . 1 , third c o l u m n ) .

Copyrighted material

Projections

of Public

Opinion

61

The chances that the calculated fit was no better than that o b t a i n e d f r o m random p o l l points were less than 0.001 (Table 5.1). Therefore, the ideodynamic projection is statistically m u c h better than those obtained by either no o p i n i o n change or by a random choice o f p o l l values. I t s h o u l d be noted that both p o l l points and projected o p i n i o n values were correlated w i t h each other as time proceeded, since o p i n i o n at a later t i m e was dependent to some extent on opinion at earlier times. Therefore, i n the absence o f s i m p l i f y i n g a p p r o x i m a t i o n s w h i c h are d i f f i c u l t t o j u s t i f y r i g o r o u s l y , i t was inappropriate to calculate r2 regressions requiring time independence for the p o l l values. Ideodynatnics postulates that the parameters i n the opinion projection equations should be constant, changing very little over the time period o f the calculations. I f this is true and i f language usage remained the same, then i t should have been possible to extend the opinion projections to a later date simply by retrieving more A P dispatches and running the same programs under the conditions used for the studies f r o m 1977 to 1984. T o test this hypothesis, the Nexis data base was searched f r o m January 1, 1981 t o A p r i l 12, 1986. using the commands first e m p l o y e d for defense spending. F r o m the 10,451 dispatches identified, 1,067 were retrieved at random and analyzed using the same text analysis described earlier for the three positions favoring more, same, and less spending. The o n l y change was a single alteration i n the dictionary for the first filtration step. After 1984, the disease o f acquired immune deficiency syndrome became much more prominent. Therefore, its acronym A I D S was found i n a significant number o f dispatches describing spending for defense against A I D S . Consequently, i t was necessary to eliminate dispatches i f they contained the w o r d A I D S , and this w o r d was added t o the dictionary i n the first filtration together w i t h the words " f u n d " and " a i d . " When any o f these words appeared i n a dispatch, the story was eliminated as being irrelevant to American military spending. A f t e r the text analysis, 507 stories had non-zero scores f o r defense spending. This was about half the initial number o f dispatches and was not very different f r o m the 39 percent f o u n d i n the analysis for stories f r o m 1977 to 1984. Perhaps the 39 percent was a little l o w , since text ceased to be collected whenever the reader felt that a story was u n l i k e l y t o be about defense spending. Therefore, late mentions o f defense spending w o u l d have often resulted i n discarded articles i n the first set o f retrievals. T h i s manual interference i n the collection d i d not occur for the retrievals from 1981 to 1986. Since neither set o f retrievals included all possible stories, and since there was overlap between the t w o data sets, all paragraph scores were corrected to the expected value corresponding to ail dispatches being collected. For instance, i f o n l y 1/10 o f all dispatches were collected at random i n a time period, all paragraph scores i n that period w o u l d have been m u l t i p l i e d by ten. The persuasive force curves i n c l u d i n g these corrections are shown i n Figure 5.9. These data showed that there was very little change i n the i n f o r m a t i o n structure f r o m 1982 to 1986, w i t h the ratios o f information favoring the three positions staying relatively constant. Based on these results, i t was expected that o p i n i o n w o u l d also be quite stable. This was indeed the case both for projected and measured o p i n i o n (Figure 5.10). Stability i n the p o l l results c o u l d be seen even though there were fluctuations in the data f r o m different polling organizations after 1982. Despite this scatter, i t seemed plausible that o p i n i o n f a v o r i n g more spending might have been systematically overestimated after 1983. Therefore, a least squares optimization was performed over the entire time period f r o m 1977 to 1986 to see i f a better projection c o u l d be obtained by g i v i n g information favoring less spending a greater weight. F r o m the optimization curve i n Figure 5 . 1 1 , a weight o f about 1.2 gave a marginally better fit to the p o l l points. Given the small effect o f the weight increase (decrease i n M S D o f less than 1/20), i t was not used in the computations for Figure 5.10. One interesting result o f this optimization is that the small improvement i n the fit required a greater w e i g h t for data f a v o r i n g less spending. That meant that

Copyrighted material

62

Predictions

of Public

Opinion from the Mass

Media

information opposing more defense spending was more effective than i n f o r m a t i o n supporting more spending. Since many pro-spending infons came f r o m the Reagan administration, the suggestion is that a popular president and his administration were i n fact less effective than the opposition i n swaying public o p i n i o n , a result different f r o m that obtained by Page and Shapiro (1984) and Page, Shapiro, and Dempsey (1987). One possible explanation for the underestimate i n o p i n i o n favoring less defense spending was the presence o f additional indirect i n f o r m a t i o n not included i n the analyses described above. A good candidate seemed to be stories on waste, fraud, and corruption by defense contractors. Therefore, additional dispatches were retrieved focusing on these topics and all paragraphs discussing these issues were scored as favoring less defense spending. A l l these paragraphs were given the same weight as paragraphs direcUy supporting less defense spending and a persuasive force curve based on these waste and fraud infons alone was constructed using the usual one day half-life (Figure 5.12, top frame). I n c l u s i o n o f these extra infons i n die o p i n i o n computations meant that their persuasive force curve was added to the persuasive force curve for infons directly advocating less spending (Figure 5,9, b o t t o m frame). Q u i c k inspection shows that the waste and fraud infons (Figure 5.12, top frame) were negligible w h e n compared to the direct infons favoring less spending. Therefore, there was very litde difference between the persuasive force curves w i t h (Figure 5.12, center frame) and w ithout (Figure 5.9, bottom frame) the added infons. This sameness i n persuasive force meant that there was very little difference between the o p i n i o n projections f o r more spending w i t h and w i t h o u t considering waste and fraud ( t w o lines i n bottom frame, Figure 5.12). Therefore, b y taking into account all relevant i n f o r m a t i o n , an ideodynamic analysis was able to suggest that opinion on defense spending was not greatly influenced by information o n waste and fraud. The o n l y caveat to this interpretation is that the public might have weighted this information much more heavily than information directly speaking to the issue. Unfortunately, there was no direct method to test this possibility.

5.2 O P I N I O N P R E D I C T I O N S FOR T R O O P S I N L E B A N O N The issue o f troops in Lebanon was like defense spending in that o p i n i o n both increased and decreased significantly. As usual, opinion calculations began w i t h the construction o f infon persuasive force curves using A P dispatches scored for favoring more, same, or less troops. Then computations o f poll percentages were made w i t h the m o d i f i e d p e r s u a s i b i l i t y c o n s t a n t b e i n g the o n l y v a r i a b l e p a r a m e t e r . U n f o r t u n a t e l y , a good f i t was not o b t a i n e d . E x a m i n a t i o n o f the projections suggested that t w o modifications could improve the calculations. F i r s t o f a l l , o p i n i o n f a v o r i n g less troops seemed t o be s y s t e m a t i c a l l y underestimated so scores favoring this position were given a r e f i n i n g weight o f 1.6 by least squares o p t i m i z a t i o n . Persuasive force curves i n c l u d i n g this weight for scores opposing more troops and a weight o f 1.0 for other infons are plotted i n Figure 5.13. Furthermore, o n October 23, 1983, there was the unexpected explosion by a terrorist o f a truck laden w i t h explosives i n the headquarters o f the U n i t e d States Marines i n Beirut, k i l l i n g over 200 soldiers. I t seemed reasonable to suppose that the population reacted viscerally to this report, feeling that some action was required, either p u t t i n g more troops i n or p u l l i n g the ones there o u t . Therefore, a new persuasive force curve was computed i n c l u d i n g eighty paragraph equivalents f o r i n f o r m a t i o n f a v o r i n g more troops (Figure 5.14, lower frame) and no paragraphs f a v o r i n g t r o o p r e m o v a l . For reference, the persuasive force w i t h o u t this truck bombing infon is replotted f r o m Figure 5.13 (Figure 5.14, top frame). One difference between the plots o f Figures 5.13 and 5.14 is the beginning time for the curves. Figure 5.13 illustrates the fact that data retrievals always began at least six months before the date o f the first p o l l d a t e unless that w c a s b e f o r e J a n u a r y 1, 1977, the beginning date o f the Nexis data base for the AP. This was to assure

Copyright

Projections

of Public

Opinion

63

that the residual effects o f prior messages were included i n the o p i n i o n calculations. For the persuasive force curves described above, the universal persistence half-life o f one day was used. The o p i n i o n computations themselves then r e q u i r e d the f o r m u l a t i o n o f a population conversion model. T h e best one f o r troops i n Lebanon was a " d i r e c t conversion model" (Figure 5.15) where individuals could move directly f r o m any one subpopulation t o another w i t h o u t passing through any intermediate subgroups, however transient I y. As was standard practice, the persuasive force curves i n Figures 5.13 and 5.14 were calculated using constants o p t i m i z e d b y m i n i m u m M S D . These constants included the modified persuasibility constant and the weight f o r scores favoring less troops (Figure 5.16), and the persistence constant (Figure 5.17). The effect o f news o f the truck bombing o n October 2 3 , 1 9 8 3 , was modeled b y injecting a r t i f i c i a l l y , o n that date, t w o infons o f u n k n o w n m a g n i t u d e , a truck b o m b i n g i n f o n supporting more troops and a truck b o m b i n g i n f o n f a v o r i n g less troops. T h e persuasive force curves f o r these t w o infons were assumed t o be characterized by the same one-day persistence. For comparability, the content scores for these truck bombing infons were given i n A P paragraph equivalents. W h i l e the o p t i m i z a t i o n for the truck b o m b i n g i n f o n f a v o r i n g more troops showed a marked improvement at a value o f eighty A P paragraph equivalents (Figure 5.16, b o t t o m frame), there was n o need even t o i n v o k e a truck b o m b i n g i n f o n favoring troop w i t h d r a w a l since the optimization showed that the fit d i d not improve s i g n i f i c a n t l y as more paragraphs were added up to about f o r t y A P paragraph equivalents (Figure 5.17, lower frame). A s the number o f paragraphs increased above this number, the fit got appreciably worse. As f o r defense spending, the optimization for the persistence constant showed t w o m i n i m a , one w i t h a one day half-life and one w i t h a half-life o f fourteen days (Figure 5.17, upper frame). The criterion o f m i n i m u m M S D meant that the lower m i n i m u m corresponding to a one day half-life was chosen. A s argued for defense spending, i t seemed unreasonable to have an even shorter half-life. T h e p o l l data f r o m October 2 3 , 1983, were o m i t t e d f r o m the o p t i m i z a t i o n calculations since that was the date o f the truck b o m b i n g i n f o n . G i v e n the r a p i d changes i n the i n f o n and p o l l data, accurate projections o n this date w o u l d have required that the infons and p o l l values be assigned t o specific hours, an impossible task given the uncertainties i n the timing o f both the infons and p o l l itself. E x a m i n a t i o n o f the o p i n i o n projection patterns shows that one o f the largest effects was due to the introduction o f the October 23 truck b o m b i n g i n f o n . W i t h o u t this i n f o n , there w o u l d have been little o p i n i o n change d u r i n g the entire p o l l i n g p e r i o d , i n disagreement w i t h the p o l l data (Figure 5.18). Once the i n f o n was introduced, o p i n i o n fit quite w e l l (Figure 5.19). Comparison o f projections w i t h and w i t h o u t the truck b o m b i n g i n f o n favoring more troops (Figure 5.20) showed that this i n f o n d i d have a very large effect immediately after October 23. However, w i t h i n a couple o f months the effect was effectively dissipated, since n e w o p i n i o n reflected A P i n f o r m a t i o n at later times. This result is l i k e that i n Figure 5.7 f o r defense spending, where o p i n i o n m o v e d more gradually to conform w i t h the information structure. As discussed above, there was n o need to postulate any truck bombing infons f a v o r i n g fewer troops. T h a t was because there was already a large amount o f i n f o r m a t i o n favoring fewer troops (Figure 5 . 2 1 , t o p frame). T h e i n t r o d u c t i o n o f forty more paragraphs o n October 23 d i d not significantly change the shape o f the persuasive force curve favoring less troops (Figure 5.21, lower frame). Therefore, there was also litde effect o n the o p i n i o n calculations (Figure 5.22). This result is quite like that for stories o n waste and fraud for defense spending, where the additional news was insignificant compared w i t h other infons supporting less spending. In contrast, the truck b o m b i n g i n f o n f a v o r i n g more troops had a very large impact (Figure 5.20) because there was very l i t t l e other news i n favor o f that position (Figure 5.14).

CoDvriahted material

64

Predictions

of Public Opinion from the Muss

Media

The i m p r o v e d f i t by the addition o f the truck b o m b i n g i n f o n was manifested both by visual inspection o f the o p i n i o n projections (Figure 5.19) and by the R M S D (3.5 percent w i t h the i n f o n and 9.1 percent w i t h o u t Table 5.1). Table 5.1 (righthand c o l u m n ) also shows that the best projection had a probability less than 0.001 o f being found by chance.

5.3 O P I N I O N P R E D I C T I O N S FOR T H E D E M O C R A T I C

PRIMARY

Besides being able to m o d e l policy i s s u e s , ideodynamics i s a l s o applicable to electoral situations such as the Democratic primary o f 1984. The dispatches for this topic were scored both by the p r o x i m i t y o f bandwagon words to candidate names and by a count o f the candidate names (Chapter 4). Therefore, persuasive force curves were computed for both analyses. The bandwagon analysis yielded curves both favorable and unfavorable to M o n d a l e , G l e n n , or Others (Figures 5.23 and 5.24) while o n l y infons mentioning the three candidates were computed i n the name-count analysis ( F i g u r e 5.25). A l l scores were g i v e n the same w e i g h t o f 1.0 and the persistence constant had a one day half-life. The first study used the bandwagon content analysis. Since the persuasive force curves for this analysis were for positions both favorable and unfavorable to all three groups o f candidates, neither the sequential population conversion model (Figure 1.2) nor the direct conversion model (Figure 5.15) was appropriate. Instead, a m i x t u r e o f both models was used (Figure 5.26). T h i s model was unique among the models i n this book i n i n c l u d i n g persons w i t h N o O p i n i o n . I n other models, these persons were ignored, equivalent to the approximation that the majority o f o p i n i o n changes involved those w h o already had an o p i n i o n . However, for the Democratic primary, it seemed unreasonable that information unfavorable to a candidate w o u l d cause a supporter to favor any particular one o f the other candidates. Rather, i t s e e m e d more plausible that t h e conversion w o u l d be to Undecided or N o O p i n i o n . Therefore, this category was i n c l u d e d f o r the b a n d w a g o n analysis. Information actually favoring a candidate was presumed t o be able t o draw recruits f r o m any other subgroup. This model had features o f both the direct and sequential conversion models. A person favoring a particular candidate c o u l d be convinced to favor another, either direcUy or by first becoming disenchanted and moving temporarily into the undecided pool. W h e n content scores f r o m the name-count analysis were used, there was n o information unfavorable t o a candidate and hence favoring N o O p i n i o n . Therefore, the same approximation was made as for the other studies i n this b o o k . That is, the N o O p i n i o n subpopulation was assumed to consist o f people w h o were unconcerned about the and were not responsive t o i n f o r m a t i o n about the c a m p a i g n . Consequently, a l l persons w i t h an opinion were normalized to 100 percent. As noted earlier, the inclusion or exclusion o f the N o Opinions was not crucial t o the final curves since they only comprised a r o u n d 10 percent o f the total population at m o s t W i t h the exclusion o f the N o Opinions, the direct conversion model used f o r troops i n Lebanon was the most reasonable (Figure 5.27) since i t c o u l d not be argued that support for any candidate should be preceded by support for another. For the o p i n i o n projections for the bandwagon analysis, the persuasive force curves (Figures 5.23 and 5.24) were constructed using die persistence half-life o f one day and the same weight for all scores. The remaining u n k n o w n was the modified persuasibility constant. Both the persistence and modified persuasibility constants were fixed by least squares o p t i m i z a t i o n (Figure 5.28, top t w o frames) using the population conversion m o d e l o f F i g u r e 5.26. T h e best m o d i f i e d persuasibility constant was 1.5 per A P paragraph. The actual poll projections using the optimized constants showed a reasonable fit (Figure 5.29), w i t h the most dramatic change being the almost t w o f o l d decrease i n support for G lenn. F o r c o m p a r a b i l i t y , the same persuasibility constant was used to compute the poll projections using the name-count analysis. These calculations were based on the

issue

Copyrighted material

Projections

of Public

Opinion

65

persuasive force curves o f Figure 5.25 and the model of Figure 5.27. The projections (Figure 5.30) were o b v i o u s l y unsatisfactory. T h i s is seen in the R M S D o f 23.2 percent, over five times greater than the R M S D for the bandwagon (Table 5.1). In fact, when an o p t i m i z a t i o n was performed t o choose the best m o d i f i e d p e r s u a s i b i l i t y constant f o r the name c o u n t analysis, i t was f o u n d that n o i m p r o v e m e n t s were possible b e y o n d the s i t u a t i o n o f no o p i n i o n change corresponding to the m o d i f i e d persuasibility constant having a value close t o zero (Figure 5.28, bottom frame). The persuasive force curves o f Figure 5.25 give the reason for the inaccuracy. A t essentially all times, there was a large excess o f name mentions for the m i n o r candidates (bottom frame) as compared to the t w o front runners. The actual numbers for the name counts are g i v e n i n Table 5.2. For example, Jesse Jackson was discussed i n the news at a frequency (20 percent) between that o f Glenn (17 percent) and Mondale (27 percent) due i n large part to his efforts to free a naval flier downed in Lebanon. Cranston's name mention frequency o f 13 percent was not far behind that o f Glenn, due principally to his advocacy o f a nuclear freeze. I n fact, this excess i n the mention o f m i n o r candidates was seen throughout the p o l l i n g period. Therefore, the model predicted a net movement away f r o m both major candidates toward the m i n o r ones. T h i s was seen i n the projected d r o p in support for both Glenn and M o n d a l e w i t h an accompanying rise i n the calculated popularity o f the Others (Figure 5.30). Since the projected d r o p for G l e n n was accompanied b y the w r o n g movements i n the other t w o curves, this f i t was fortuitous. F r o m the optimizations for the bandwagon analysis, the proper choice for the modified persuasibility constant gave an M S D over two-fold better than the estimate f r o m a very small modified persuasibility constant equivalent to no change i n public o p i n i o n . T h i s improvement was less than that for defense spending (fivefold) and troops i n Lebanon (tenfold). H o w e v e r , i t was unreasonable to expect as m u c h improvement because the p o l l values changed much more for those other examples, so i t was less appropriate for them to be approximated by no o p i n i o n change d u r i n g the p o l l i n g period. Since the polls changed m u c h less for the Democratic p r i m a r y , an estimate o f no o p i n i o n change gave a much better fit. Nevertheless, this t w o - f o l d increase was substantial. The chances o f obtaining such a good fit by chance were less than 0.001 (Table 5.1). The projection using the name-count analysis was quite unsatisfactory. As mentioned above, the M S D o f 540 was substantially worse than the prediction o f no o p i n i o n change ( M S D o f 39). Indeed, the estimate was so bad that i t c o u l d be obtained 34 percent o f the time f r o m random p o l l points (Table 5.1).

analysis

5.4 O P I N I O N P R E D I C T I O N S FOR T H E E C O N O M I C

CLIMATE

The economic c l i m a t e was the first o f t w o economic issues studied. A s for defense spending and the Democratic p r i m a r y , there was no need to make any adjustments by weighting the i n f o n content scores. The persuasive force curves for this issue (Figure 5.31) were computed using the usual one day persistence half-life. For this topic both the direct and sequential population conversion models were tried. A m o n g these, the sequential conversion model was better (Figure 5.32). W i t h a l l infons w e i g h t e d the same, the persistence constant and c o m m o n modified persuasibility constant were both set by least squares optimization (Figure 5.33). This procedure led to an approximately tenfold improvement in the M S D over the estimate f r o m no o p i n i o n change, as w o u l d be expected f r o m a p o l l series where the o p i n i o n values d i d vary significantly f r o m the initial value. This is seen i n the projection curves (Figure 5.34). As w i t h all other satisfactory computations, the probability o f such a fit by chance was less than 0.001 (Table 5.1).

Copyrighted material

66

Predictions

of Public

Opinion from

the Mass

Media

5.5

O P I N I O N PREDICTIONS FOR U N E M P L O Y M E N T VERSUS I N F L A T I O N

The second economic topic was unemployment versus inflation. T h i s issue was like that for troops i n Lebanon in that i t was necessary to give different infon scores different weights. By least squares optimization it was found that scores favoring the importance o f inflation should have a weight o f 1.4 w h i l e those f a v o r i n g equal importance should have a weight o f 0.5. The remaining infon group supporting the importance o f u n e m p l o y m e n t had the reference w e i g h t o f 1.0. Least squares optimization was also used to give an optimal persistence constant w i t h a one day half-life. The persuasive force curves reflecting these constants are presented i n Figure 5.35. As for the other examples where the p o l l positions ranged f r o m one extreme through the middle to the other, it was possible to test both the direct and sequential population conversion models. O f these, the best was the direct conversion model (Figure 5.36). Least squares optimizations were used to determine the best values for all constants used i n the construction o f the persuasive force functions and the modified p e r s u a s i b i l i t y constant f o r the reference i n f o n s f a v o r i n g the i m p o r t a n c e o f unemployment (Figures 5.37 and 5.38). O p i n i o n projections were then made using the o p t i m i z e d constants (Figure 5.39). W i t h the large o p i n i o n changes d u r i n g the p o l l i n g period, the estimate o f no opinion change was so poor that the ideodynamic calculation could give an improvement i n the M S D o f four- to five-fold (Figure 5.37, top frame). For this analysis also, there was a probability o f less than 0.001 that the fit could have been obtained by random poll points (Table 5.1).

5.6

O P I N I O N P R E D I C T I O N S FOR C O N T R A A I D

Contra a i d was the o n l y study in this book where o p i n i o n was f a i r l y static d u r i n g the time period o f the study. T h e text analysis for Contra aid was unique, having been performed separately both by Fan and by three graduate research assistants ( S w i m et al. i n Chapter 4 ) . Therefore, there were two sets o f infon scores (Figures 5.40 and 5.41). A l t h o u g h S w i m et al. gave scores to about t w i c e as many paragraphs, the t w o sets o f scores revealed essentially the same overall i n f o r m a t i o n structure. For both sets o f scores, least squares optimizations showed that infons opposing aid needed to have a much greater weight than infons favoring a i d . The optimized weights were similar, being 2.0 for the Fan scores and 2.4 for the S w i m et al. scores. A l s o , o p t i m i z a t i o n o f the persistence constant resulted in the best halflives having values greater than one day for both i n f o n sets. However, the one day half-life was satisfactory, so that was the value used for calculating the persuasive force curves i n Figures 5.40 and 5.41. Since there w ere only t w o positions, the o n l y reasonable population conversion model was that o p i n i o n favoring one side c o u l d convert people favoring the other (Figure 5.42). This was obviously the degenerate case where the direct and sequential conversion models collapsed into the same model. As w i t h a l l other examples studied, the optimization for the persistence constant for Contra aid had t w o m i n i m a , one at a one day half-life and the other ranging f r o m seven to over 100 days (Figure 5.43, bottom frame). Since the one-day m i n i m u m was c o m m o n to a l l six issues, i t was used as the consensus value for a universal persistence constant for all issues. The one day half-life was certainly reasonable since this was also the lowest m i n i m u m for the five examples besides Contra aid. I t was also more reasonable to set the persistence constant for the other five issues, f o r w h i c h the i n f o r m a t i o n structure and o p i n i o n both changed significantly. For Contra aid, there was little change i n either o p i n i o n ( A p p e n d i x B, Table B.6) o r the ratio o f favorable to unfavorable information (Figures 5.40 and 5.41). Therefore, much o f any different fit in the persistence constant c o u l d have been t o errors in the opinion p o l l s .

Copyrightec

Projections

of Public Opinion

67

H a v i n g j u s t i f i e d the use o f a one-day universal persistence constant based on six independent optimizations for six different topics, the final opinion projections used this half-life. These projections also used refining weights o f 2.0 and 2.4 for the t w o sets o f infons favoring less aid, as described earlier. The final curves (Figures S.44 and 5.45) showed relatively little change in opinion during the entire three-year period once the m o d i f i e d persuasibility constant was also set by least squares optimization (Figure S.43, top frame). Nevertheless, there was a substantial improvement i n the fit over projections o f almost no change at a l l after the first p o l l point since the M S D was more than three-fold lower than that for 0.001 per A P paragraph per day for the m o d i f i e d persuasibility constant. The probability o f having the quality o f fit shown i n Figures S.44 and S.4S by chance alone were again less than 0.001 (Table 5.1). It was gratifying to find that t w o quite different text analyses c o u l d both g i v e essentially indistinguishable o p i n i o n projections (Figures S.44 and S.4S). T h i s result is not surprising since both analyses showed that there were the same approximate ratios o f infons f a v o r i n g and opposing C o n t r a aid t h r o u g h o u t the polling period (Figures 5.40 and 5.41).

5.7 S U M M A R Y O F C O N S T A N T S U S E D I N P O L L P R O J E C T I O N S This chapter has presented opinion projections for six quite disparate issues. The interpretations o f these results are g i v e n i n the next t w o chapters. For those discussions i t w i l l be useful to have a list o f all the parameters w h i c h were chosen by least squares o p t i m i z a t i o n for each o f the cases (Table 5.3). A l l constants were o p t i m i z e d under the best conditions for the other constants except for the issue o f Contra aid, where optimizations were performed using a one-day persistence half-life. The constants w h i c h were optimized fell into the f o l l o w i n g categories: 1. Persistence constant-this constant measured the ability o f an A P infon to continue to exert its effect after the date o f the dispatch. As noted i n the previous section, the optimal value for this constant was a one day half-life for five out o f six issues and this same value was also satisfactory for the sixth. Thus the optimizations for the six issues yielded a universal one day half-life for the persistence constant. 2. M o d i f i e d persuasibility constants- tor modified persuasibility constants, the least squares optimal value was calculated for an arbitrarly chosen reference set o f i n f o n scores. A n y needed variations i n these constants for infons favoring different positions were incorporated into the r e f i n i n g weights. W h e n all w e i g h t i n g values were the same, as was the case for defense spending, the Democratic primary, and the economic climate, any one o f the positions could have been chosen as the reference position. 3. Refining weight the refining weight for a position was the constant b y w h i c h all infon scores for a position were m u l t i p l i e d before construction o f the persuasive force functions. T h i s w e i g h t i n c l u d e d both differences between the p e r s u a s i b i l i t y o f the p u b l i c for d i f f e r e n t positions and imperfections in the infon scoring (Chapter 1 and Appendix A ) . Therefore, when the public was as easy to persuade for all positions and when all infons were scored correctly, a l l refining weights had the value o f 1.0. I n some cases the weights were different for different infon scores. A n example was the issue o f troops i n Lebanon, where the weight was 1.6 for scores favoring less troops and 1.0 for a l l other scores. This meant that a paragraph favoring less troops was 1.6 times as effective as a paragraph favoring same or more troops.

Cc

al

68

Predictions

of Public

Opinion from tlie Mass

Media

5.8 S U M M A R Y O F S T A T I S T I C S FOR P O L L P R O J E C T I O N S Statistically speaking, the estimate for a value is the mean, w i t h deviations f r o m that mean being characterized by the standard deviation. The standard deviation for a set o f values is computed by taking the square root o f the squares o f the deviations o f the i n d i v i d u a l points f r o m the mean. S i m i l a r l y , i f the estimate f o r an o p i n i o n percentage is given by the ideodynamic prediction, then statistical deviations can be represented by the R M S D computed by taking the square roots o f the squares o f the deviations between actual p o l l measurements a n d the ideodynamics predictions. Therefore, assuming that the differences between t w o values are statistical, the R M S D is like the standard deviation, so that differences between the predicted and measured p o l l values should be w i t h i n one R M S D about 68 percent o f the time and w i t h i n t w o R M S D about 95 percent o f the time. The R M S D values i n Table 5.1 are the m i n i m u m values corresponding to the best parameters chosen. These values were 3.5 percent f o r troops i n Lebanon, 4.3 percent for the Democratic primary, 4.7 percent for Contra aid, 6.6 percent f o r the economic climate, 7.2 percent for defense spending (1977-1984), and 7.7 percent for u n e m p l o y m e n t versus i n f l a t i o n . T h e errors increased w i t h the t i m e span o f the projections. F o r instance, the most accurate computations were for troops i n Lebanon and the Democratic primary, where the p o l l series o n l y covered four and seven months respectively. The analyses for the other four examples spanned periods f r o m three t o seven years. I t is quite possible that text and its interpretation changed w i t h time so that the text analyses should have been modified as time proceeded. Also, the m o d i f i e d persuasibility constants m i g h t have been dependent o n time, changing slowly over a period o f years. For comparison, national polls frequently have reported errors due t o finite sample size i n the range o f 4 percent at the 95 percent confidence level. T h i s is equivalent t o t w o standard deviations f r o m the reported values. Therefore, their equivalent to the R M S D w o u l d have been about 2 percent instead o f the 3.5 percent to 7.7 percent i n Table 5 . 1 . However, besides sample size errors, there are also systematic errors i n the polls due to such factors as question w o r d i n g and question sequence. Therefore, the 2 percent standard error is a m i n i m u m error i n the o p i n i o n measurements. The importance o f these systematic errors is seen i n the poll data o f Figure 5.10, where p o l l s were taken at close i n t e r v a l s b y d i f f e r e n t , respectable p o l l i n g organizations. There are fluctuations i n the range o f 5-10 percent w h i c h may have been r e a l , b u t w h i c h m a y also have been due t o differences i n t h e p o l l i n g instruments. I f actual measurements o f o p i n i o n can have errors i n the range o f 5-10 percent, then the ideodynamic errors are i n the same range and may be as good as those f r o m opinion polls. F r o m the summary data i n Table 5 . 1 , i t is also clear that the best ideodynamic o p i n i o n projections had very little probability (less than 0.001) o f being obtained b y chance alone. I n a d d i t i o n , a l l projections were also better than the model o f no o p i n i o n change after the first p o l l p o i n t This was seen i n the improvements i n the M S D as the modified persuasibility constant increased above values near zero.

Copyrighted material

A l l computations used the constants i n Table 5.3 except that the persistence half-life was one day f o r a l l calculations, including the t w o for Contra aid. Table 5.1. Statistical

comparisons

for opinion projections.

Ideodynamic Estimate

1000 M o n t e Carlo Estimates (Assuming Random Poll Values)

Issue

Ideodynamic RMSD (in Poll Percentage Points)

Mean Monte Carlo RMSD (in Poll Percentage Points)

Number o f Sutndard Deviations from Ideodynamic M S D to Monte Carlo Mean M S D

Probability of Obtaining Ideodynamic MSD by Chance

Defense Spending (1977-1984) Scores M o r e , Same, Less 692 retrievals 325 retrievals 383 retrievals Scores M o r e , Less 692 retrievals

7.2 9.4 10.3

24.3 24.3 24.3

5.2 4.9 4.7

<0.001 <0.001 <0.001

8.3

24.3

5.1

<0.001

3.5 9.1

26.6 26.6

4.2 3.7

<0.001 <0.001

Bandwagon Text Analysis 4.3 23.2 Name Count Scoring

19.9 22.0

3.7 0.3

<0.001 0.34

Troops i n Lebanon W i t h T r u c k B o m b Infons N o T r u c k B o m b Infons

Democratic Primary

Economic Climate

6.6

23.3

5.4

<0.001

Unemployment vs. Inflation

7.7

27.1

4.8

<0.001

Contra A i d Fan Text Analysis S w i m et al. Text Analysis

4.7 5.1

32.9 32.9

3.3 3.3

<0.001 <0.001

Copyrighted material

Table 5.2. Candidate name counts in dispatches retrieved for the Democratic

primary.

The name counts were f o r all the retrieved A P dispatches. T h e Others category was the sum o f the results for the others.

Name Count Candidate

Actual number

Mondale Glenn Others:

959 632 2018

27 17 56

189 450 251 255 740 133

5 13 7 7 20 4

Askew Cranston Hart Rollings Jackson Mc Govern

Percent o f Total

70

Copyrightec

This table summarizes the o p t i m a l values determined by m i n i m i z a t i o n o f the M S D (Figures 5.7, 5.19, 5.20, 5 . 3 1 , 5.36, 5.40, 5 . 4 1 , and 5.46) and used f o r the o p i n i o n projections. T h e o n l y exception was for Contra a i d , where the h a l f - l i f e o f one d a y was used f o r the projections. T h e m o d i f i e d persuasibility constant k'i units were per A P paragraph Table 5.3. Optimal

constants for opinion projections.

per day and the persistence half-life was measured i n days. Unless otherwise noted. all infons had the weight of 1.0.

Persistence Persuasibility Constant Constant

Infons w i t h Weights ¿1.0

Position Favored Weight by Infons

m

Defense Spending (1977-1984) Scored More, Same, Less 692 retrievals 1.0

0.6

Troops i n Lebanon w ith Truck Bomb Infon

4.5

1.0

1.6 ^80

Democratic Primary Bandwagon Text Analysis

1.0

1.5

Economic Climate

1.0

0.09

U n e m p l o y m e n t vs. Inflation

1.0

7

0.5

1.4 Contra A i d Fan Text Analysis S w i m et al.Text Analysis

41 7

1.2 1.4

2.0 2.4

Less Troops Truck Bomb Imply M o r e Troops)

Unemployment and Inflation Equally Important Inflation Important

Oppose A i d Oppose A i d

T h i s line means that the i n f o n favoring more troops due solely to news o f the October 23, 1983, truck bombing had a value equivalent t o 80 A P paragraphs favoring more troops. a

71

Copyrighted material

Figure 5.1. Persuasive forces of AP infons scored for favoring more, same, ami less defense spending. O f the 692 retrieved dispatches, 272 had scores favoring more,

same, and less spending ( A p p e n d i x C, Section C.2-3). These i n f o n scores were converted into persuasive force curves assuming that a l l infons had (he same weight o f 1.0 and that the persistence half-life was one day. As for Figure 1.1, the curves for separate infons were added together to give the net forces i n the three directions. The heights o f the curves for individual infons, before the addition, were the sums o f the paragraph scores (Chapter 3). Infons

Favor

MORE

(Paras-Scored

More.

Same,

Less)

Favor

SAME

(Paras-Scored

More,

Same,

Leas)

Favor

LESS

(Paras-Scored

More,

Same,

LCSH)

6 5 4

3 2 1 0 Infona 6 5 4 3 2

1 0 Infons 6 5 4 3 2 1 0 80 78 79 Years 1977-1984

77

81

82

83

84

72

C

Figure 52. Persuasive forces of AP infons scored for favoring more and less defense spending. The same 692 dispatches scored for Figure 5.1 were rescored as favoring

only more and less spending. O f these 692 dispatches, 2 8 0 had scores favoring more and less spending. Tliese scores were converted into persuasive force curves using the persistence half-life o f one day and the weight o f 1.0 for all infons. Infona

Favor

MORE

(Paras-Scored

77 78 79 80 Years,1977-1984

73

81

More,

82

Less)

83

84

Figure 5.3. Opinion on defense spending from dispatches scored to favor more, same, and less spending. The projections for the three subpopulations f a v o r i n g more, same, and less defense spending ( s o l i d lines) began w i t h the o p i n i o n measurements o f the first p o l l i n M a r c h 1977 and continued using the population conversion m o d e l o f Figure 1.2 and the persuasive forces shown i n Figure 5 . 1 . A l l 272 scores favoring more, same, and less spending were used for the computations. The modified persuasibility constant was 0.6 per A P paragraph per day. Calculations were at 24-hour intervals. F o r comparison, the squares show the results o f twentyt w o published polls (Appendix B , Table B . l ) . % Favor

HORK

(272

Scores

More,

Same.

Less)

%

Favor

SAME

(272

Scores

More,

Same,

Less)

%

Favor

LESS

(272

Scores

More,

Same,

Less)

77 78 79 80 Years 1977-1984

81

82

83

84

74

Copyrighted material

Infons f r o m 272 A P messages scored as favoring more, same, and less spending were used to project opinion assuming a number o f modified persuasibility constants (k'2) (top frame) and persistence half-lives ( b o t t o m frame). For all computations, the M S D values (see Appendix A , Section A . 1 1 ) were obtained w i t h infons favoring a l l three positions having the weight o f 1.0. T h e o p t i m a l modified persuasibility constant was 0.6 per A P paragraph per day and the best persistence constant was one day. B o t h curves w ere obtained using the optimal value for the other constant. Figure

Mean

5.4. Constant optimization

Squared

curves for defense spending.

Deviation

( i n Poll

Percent

Squared)

250 200 150 100 50 0

k'2 Mean

Squared

0. 1

0.01

0.001

for ALL Infons

1

( p e r Para-Day)

Deviation ( i n Poll

Percent

Squared)

350 300 250 200 150 100 50 0 1 H a l f - l i f e

10 5 f o r Defense Spendin

100 50 in Days

75

Copyrighted material

Figure 5-5. Opinion from a subset of AP dispatches scored to favor more, same, and less defense spending. The projections (solid lines) were the same as those i n Figure

5.3 except that o n l y 325 o f the original 692 A P dispatches were used for calculating the persuasive forces. T h e same constants were used as i n Figure 5.3. O f the 325 dispatches, 131 had at least one paragraph w i t h a score favoring more, same, or less spending. For comparison, the same poll data are plotted as i n Figure 5.3. %

Favor

MORE

(131

Scores

More.

Same.

Less)

X

Favor

SAME

(131

Seoros

More.

Same.

Less)

%

Favor

LESS

(131

Scores

More,

Same,

Less)

77 78 79 80 Years 1977-1984

81

82

83

84

76

CopyriohtGcl mstcrisl

Figure

5.6. Opinion from

another subset of AP dispatches

scored

to favor

more,

same, and less defense spending. The projections (solid lines) were the same as those i n Figure 5.5 b u t used J83 essentially non-overlapping dispatches f r o m among the o r i g i n a l 6 9 2 used f o r Figure 5.3. O f the 383 dispatches, 147 had at least one paragraph w i t h a score favoring more, same, o r less spending. Actual p o l l values are shown as squares.

X F a v o r MORB (14 7 S c o r e s More. Same, L e s a )

X F a v o r SAME (147 S c o r e s More. Same, L e s s j

% _ F a v o r _ _ L E S S A1A.1.

Scores

77

78 79 80 Y e a r s 1977-1984

More,

8!

Same,

82

Lesa)

83

81

77

Copyrighted material

Figure more

5.7.

Opinion

spending

on defense

spending

at the time of the first

assuming scored

the entire population

AP

infon in January

favored

1977.

The

persuasive forces were the 272 scores used for Figure 5.3 favoring more, same, and less spending. A l l conditions for the calculation were unchanged except that the population was not assumed to begin w i t h the first actual p o l l point, but rather w i t h the a r t i f i c i a l i n i t i a l c o n d i t i o n o f 100 percent o f the p o p u l a t i o n f a v o r i n g more spending at the time o f the first scored infon. The projections based o n 100 percent initially favoring more spending (solid lines) are plotted together w i t h those based o n the first p o l l point (dotted lines, replotted f r o m Figure 5.3). %

Favor

MORE

(272

Scores

f o r

More,

Saae,

Leas)

%

Favor

SAME

(272

Scores

f o r

More.

Suae,

Less)

77 78 79 80 Y e a rs 1977-1984

81

82

83

84

78

Copyrighte

Figure

5.8.

Opinion

on defense spending from dispatches

scored to favor

more

and

less defense spending only.

The projections ( s o l i d lines) f o r the three groups favoring more, same, and less spending began with the opinion measurements o f the first p o l l i n M a r c h 1977 and continued using the 280 scores for the persuasive forces shown i n Figure S.2. As for Figure 5.3, the model was given i n Figure 1.2 and the m o d i f i e d persuasibility constant was 0.6 per AP paragraph per day, the persistence half-life was one day, and all infons had the weight o f 1.0. A c t u a l p o l l points are shown as squares.

% F a v o r MORE (280 S c o r e s f o r More, L e s a )

X Fayqr_SAME

X

Favor

LESS

(280 S c o r e s f o r More,

(280

Scores

77

78 79 80 Y e a r s 1977-1984 79

f o r

More,

81

82

Lessj

Less)

83

84

Figure 5.9. Persuasive forces more, same, and less defense

of AP infons from 1977 to 1986 scored for favoring spending. The text analysis used f o r F i g u r e 5.1 was

applied w i t h o u t change to 1,067 A P dispatches retrieved f r o m January 1, 1981 to A p r i l 12, 1986. The resulting 779 n e w scores were merged w i t h the 272 f r o m the retrieval used f o r Figure 5 . 1 . The c o m b i n e d scores were used to calculate persuasive force curves using the same one day h a l f - l i f e and the same w e i g h t o f 1.0 for a l l infons. The paragraph scores were n o r m a l i z e d to account f o r the f act that less than 100 percent o f the identified stores were collected in both retrievals. infona _Favor

MORE

(Normalized

AP

Paras)

Infons

Favor

SAME

(Normalized

AP

Paras)

I n f o n s

Favor

LESS

(Norma I i z e d

AP

Paras)

77 78 79 80 Years 1977-1986

81

82

83

84

85

86

80 Copyrighte

The persuasive force curves o f Figure 5.9 were used to calculate public o p i n i o n (solid line) using the conditions described for Figure 5.3. For comparison, sixty-two published p o l l points were plotted, including the twenty-two used for Figure 5.3 (squares). Figure

5.10.

Opinion

on defense spending from

X

Favor

MORE

(1051

Score»

X

Favor 80

SAME

(1051

AP

LESS

(1051

Scores

1977 to 1986.

f o r

Paraa

More,

f o r

Same,

More,

Same,

Less)

Less

60 40 20 0 %

Favor

77

78

Years

79

80

81

f o r

82

More,

83

Same,

84

85

Less)

8G

1977-1986

81

Copyrighte

82

CoDvriahted material

Figure 5.12. Effect of stories on waste and fraud on public opinion on defense spending. A persuasive force curve (top frame) was constructed where all paragraphs

o n defense waste and fraud (see Appendix C, Section C.2-5) had the same value as paragraphs directly advocating a position o n defense spending. The curve on defense waste and fraud (top frame) was added to the persuasive force curve for infons direcdy supporting less spending (Figure 5.9, bottom frame) to give the combined persuasive force favoring less spending (center frame). O p i n i o n favoring more defense spending was computed using the combined persuasive force (center frame), i n place o f the persuasive force l a c k i n g the paragraphs o n waste and c o r r u p t i o n ( F i g 5.9, b o t t o m frame). T h e projections both w i t h ( s o l i d l i n e ) a n d w i t h o u t (dotted l i n e ) the paragraphs o n waste and fraud are plotted in the bottom frame. T h e curve w i t h o u t waste and fraud is the same as Figure 5.10 (top frame).

Infons

on WASTE

( N o r m a l i z e d AP P a r a s )

100

80 60 40

20 0

Infons

Favor LESS

a n d o n WASTE

(Normalized Paraa)

100 80

60 40

20 0

*

F a v o r MORE ( w i t h

a n d w i t h o u t AP P a r a a o n W a s t e )

80 60 40 20 0 77

78

Years

79 80 81 1917-198.6

82

83 84

85

86

83

Copyrighted material

Figure 5.13. Persuasive forces for troops in Lebanon from AP infons only. A P dispatches were scored t o favor more, same, and less troops ( A p p e n d i x C, Section C.3). T h e dispatch scores were converted into infon forces using a persistence halfl i f e o f one day. The paragraph scores for infons favoring less troops were each given a weight o f 1.6 relative to the infons favoring more and same troops, w h i c h both had the same w e i g h t o f 1.0. T h e large spike on the horizontal axis between October and November was at October 2 3 . T h e p l o t begins w i t h the date o f the first retrieved dispatch.

Infona

Favor

May Years

MORE T r o o p a

J u l 1983-1984

(AP ParaB-Weight

Sep

Nov

1.0)

Jan

84

Copyrighted material

Figure 5.14. Persuasive forces for troops in Lebanon from AP infons with and without a truck bombing infon favoring more troops. T h e A P dispatch scores

w i t h o u t an extra truck b o m b i n g infon ( t o p frame) are the same as i n Figure 5.13 except that the plot began w i t h the date o f the first available p o l l p o i n t (Appendix B , Table B.2). I n the b o t t o m frame, a truck b o m b i n g i n f o n equivalent t o eighty A P paragraphs favoring more troops was added on October 23, 1983, the date o f the extra tic on the horizontal axis. T h e truck bombing infon also had a one-day persistence half-life.

Infona Favor 70

MOKE

Troops

(AP

MORE

Troops

(AP

Paraa)

60 40 30 20 10 0 Infona Favor 70

+

80

Bomb

Par a a )

60 50 40 30 20 10 0

Oct Yours

Nov

Dec

Jan

1983-1984

85

Copyrighted material

r 7"0

c

C £

M

I

«=

~ 81.8

"5 o

s i 4>

M

3 CO

to

4»

M

00

5 S^

1

«.E Q

S

3 -w

*

ft,


»0

^'^ E J S 3

t-

o Jr» > 6

86

CoDvriahted material

Figure 5.16. Optimizations for the modified persuasibility constant, the weight for paragraphs favoring less troops, and the value of the truck bombing infon favoring more troops. T h e best modified persuasibility constant (k'2) g i v i n g the lowest M S D

was 4.5 per A P paragraph per day (top frame); the optimal weight for infons favoring less troops was 1.6 when all other infons had weights o f 1.0 (center frame); and the most favorable value f o r the October 2 3 , 1983, i n f o n f a v o r i n g more troops was equivalent t o e i g h t y A P paragraphs ( b o t t o m frame). F o r a l l calculations, the persistence half-life was one day, and there was assumed to be no truck b o m b i n g infons favoring less troops. I n addition, all optimizations were performed at the best values f o r the other constants.

Mean S q u a r e d

Deviation ( i n Poll

k ' 2 f o r Infons Mean _S q u a r e d D e v i a t i o n

Percent

Squared)

F a v o r MORS. SAMK ( p e r P a r a - D a ^ J ( i n P o l l P e r c e n t Squared)

Weight f o r P a r a s F a v p r L E S S T r o u p a Mean S q u a r e d D e v i a t i o n ( i n P o l l P e r c e n t S q u a r e d ) 80

BO 40

20

0 0.001

0 . 0 1

1

10

Truck

Bomb News F a v o r MORE T r o o p s

100

(AP P a r a E q u i v

87

Copyrighte

Figure 5.17. Optimization curves for the persistence half-life and the value of the truck bombing infon favoring less troops. T h e lowest M S D g i v i n g the best half-life

was one day (top frame). There was essentially n o change i n the M S D f r o m almost zero to almost f o r t y A P paragraph equivalents f o r the October 2 3 , 1983, truck b o m b i n g i n f o n f a v o r i n g less troops ( b o t t o m frame). B o t h o p t i m i z a t i o n s were performed at the best values for the other constants as indicated i n Figures 5.16 and 5.17.

Mean S q u a r e d

Dcviation

( i n Poll

Mean S q u a r e d

Deviation

LLn_.PoiJLPfrcfy?JLSqviaredi

88

Percent Squared)

Figure 5.18. Opinion on troops in Lebanon assuming only AP infons. The projections beginning w i t h the first p o l l point (solid lines) used the three persuasive force curves o f Figure 5.13, the model o f Figure 5.15, and the o p t i m a l constants o f Figures 5.16 and 5.17, w i t h the exception that the truck b o m b i n g infons favoring more and less troops were both o m i t t e d . Calculations were at six-hour intervals. A c t u a l p o l l points are plotted as squares. October 23 is indicated by the tic between October and November.

% Favor

MORH T r o o p s

(no

Bomb

Paras)

%

Favor

SAME

Troops

(no

Bomb

Paras)

*_Favor

LESS

Troops

(no

Bomb

Paras_^

Oct Years

Nov 1983-1984

Dec

Jan

89

Copyrighte

Figure 5.19. Opinion on troops in Lebanon with a truck bombing infon favoring more troops. The computations (solid lines) were the same as for Figure 5.18 except that a truck b o m b i n g infon o n October 23 favoring more troops at the o p t i m a l value o f eighty A P paragraph equivalents (Figure 5.16, b o t t o m frame) was also included. Therefore, the persuasive force curve favoring more troops was that i n the l o w e r frame o f Figure 5.14. T h e other persuasive force curves are i n the b o t t o m t w o frames o f Figure 5.13. X

Favor

MORS

Troops

(AP

»

Bomb

Favor

More

Troops)

SAME

Troops

(AP

+

Bomb

Favor

More

Troops)

LESS

Troopa

(AP

+

Bomb

Favor

More

Troops1

70 60 50 40 30 20 10 0 X

Favor 70 60 50

40 30 20 10 0 %

Favor 70 60 50 40 30 20 10

0 O c t ^ e ! i

r

Nov 1983-U H 4

Dec

Jan

90

Copyrightec

Figure 520. Comparison of opinion projections with (solid line) or without (dotted tine) the truck bombing infon favoring more troops. T h e curves i n Figures 5.18 and

5.19 are plotted together.

Favor

MORE

Troopa

( + . - Bomb

Favor

More

Troops!

X Favor

SAME

Troops

( + , - Bomb

Favor

More

Troops)

Favor

LESS

Troopa

( + . - Bomb

Favor

More

Troopa)

X

X

Oct Years

Nov 1983-1984

Dec

J a n

91

Copyrighte

Figure 521. Persuasive forces from AP infons with and without a truck bombing infon favoring less troops. The A P dispatch scores w i t h o u t an extra truck bombing

infon (top frame) are the same as the b o t t o m frame i n Figure 5.13 except that the p l o t began w i t h the date o f the first available p o l l p o i n t on troops i n L e b a n o n ( A p p e n d i x B , T a b l e B.2). T h e b o t t o m frame includes the a d d i t i o n o f a truck bombing infon favoring less troops equivalent to forty AP paragraphs on October 23» 1983, the date o f the extra tic o n the horizontal axis. The truck b o m b i n g i n f o n also had a one-day persistence half-life.

Oct Nov Years_J 983-1984

92

Dec

Jan

Figure 5.22. Opinion projections with and without a truck bombing infon favoring less troops. Projections w e r e made for public o p i n i o n as i n Figure 5.19 except that

there was the addition o f a truck bombing infon o n October 23 favoring equivalent t o forty A P paragraphs (dotted lines). For this projection, the force curve favoring less troops was that i n the bottom frame o f Figure comparison, projections w i t h o u t this truck bombing i n f o n favoring less replotted f r o m Figure 5.19 (solid lines).

X F a v o r HORK Troo 70

less troops persuasive 5.21. F o r troops are

- Bomb F a v o r L e s s T r o o p s )

60 50 40

30 20 10 0 - Bomb F a v o r L e 3 3

Troops)

(•*•.- Bomb F a v o r L e s s

Troops)

X F a v o r SAME Troops 70 60 50 40 30 20 10 0 X F a v o r LESS Troopa 70 60 50 40

30 20 10 0 Oct Years

Nov

Dec

Jan

1983-1984

93

Copyrightc

•

Figure 5.23. Persuasive forces favorable AP paragraphs scored using bandwagon

to Democratic presidential candidates from words. Paragraphs were scored as favoring

Mondale, Glenn, and Others i f candidate names were close to favorable combinations o f bandwagon words (see Appendix C, Section C.4-1). T h e infon force curves were calculated using a one-day persistence half-life and a weight o f 1.0 f o r all infons.

Bandwagon

Infona

Jul Years

Favor

MONDALE (AP P a r a s )

Sep 1983-1984

Nov

Jan

94

Copyrightec

Figure 5.24. Persuasive forces unfavorable to Democratic presidential candidates from AP paragraphs scored using bandwagon words. Paragraphs were scored as

unfavorable t o M o n d a l e , G l e n n , and Others i f candidate names were close to combinations o f unfavorable bandwagon words (see A p p e n d i x C, Section C.4-1). The i n f o n force curves were calculated using a one-day persistence half-life and all infons w i t h a weight o f 1.0 as i n Figure 5.23. Bandwagon

Infons

D i s f a v o r

MONDALE

Infons

Disfavor

GLENN

(AP

Paras)

10 8 6 4 2 0 Bandwagon

(AP

Paras)

12 10 8 6 4 2 0 Bandwagon._Infons_Disfavor

OTHERS_

(AP_.Paras)

10

8 6 4 2 0

Jul

Sep

Nuv

Jan

VvjairB _ 1 «IH:» -1 Utf 4

95

CoDvriahted material

5.25. Persuasive farces of AP infons scored by name count only. Paragraphs mentioning Mondale, Glenn, and Others were scored as discussed i n A p p e n d i x C, Section C.4-2. The i n f o n force curves were calculated w i t h a one-day persistence halt life and all infons w i t h a weight o f 1.0.

Figure

I n f o n a Mention MONDALE (AP P a r a s )

Infons

Mention

Jul

OTHERS

(AP

Sep

Paras)

Nov

Jan

Vi*aj!LB__lfl83-19_84

96

Copyrights

Figure 5.26. Population conversion model for actions of infons scored using bandwagon words. The boxes denote the subpopulations under consideration. The

words i n the boxes begin w i t h "B" to refer t o those " believing" o r having an o p i n i o n favoring Mondale, Glenn, and Others or having N o O p i n i o n . T h e persuasive forces at any particular time were read f r o m the curves i n Figures 5.23 and 5.24 and were favorable o r unfavorable t o M o n d a l e , G l e n n , o r Others. W o r d s describing these forces begin w i t h " G , " w i t h favorable information denoted b y Pro and unfavorable b y Con. T h e candidate o r group o f candidates is indicated by the first three letters o f the name or group. T h e arrows indicate opinion conversions due t o the persuasive forces shown i n Figures 5.23 and 5.24.

***************** *

GProGle

*****************

*

* BMondale * * * *****************

> *

<

* BGlenn * * * *****************

GProMon

A\

A \

*

/A

/A

GConGle // //GProGle

GProMon \\GConMon

\\ \\/ \// *****************

\\ \\ \\

// // //

//

w

* \\ GProOth * BNoOpinion * \\ ***************** GProMonW

GProOth// //GProGle

\\

GProOth

GCoriOth

w

\\ w

// // //

//

// // Y \// \\/ \ ***************** / *

* *

Rüthers *

*****************

97

Copyrighte

sir O 5

4*

*

*•

*4

§ g oe Sİ S >

-*

I f S -S 8 ri

4* •4

44

i

—

44

**•

44 H

c

C

•4

2

44 44 •4 •M-

S 1'

4* 4* 44 44

4*

*4

44 44 +4 ••4

*#-

*•

44 44 44 44 •M-

S 5 CO a;

V

4* +4 +4 4444 44 44 •H

«44* 44 44

M

Q

•**

44 44 «• 44 4*

4*

44

4*

M

¡> u *- — JS > € £fi r •¬ § 2 2 2 S c -2 X *5 -

o

o y .

44 4* •M44

44 44

-

S

f

5

>

»e a « O C

4*

98

Figure 5.28. Optimization curves for constants for the Democratic primary. Bandwagon analysis: Using the population conversion model o f Figure 5.26 and the

persuasive forces o f Figures 5.23 and 5.24, the best modified persuasibility constant \k'2) g i v t o g the lowest M S D was 1.5 per A P paragraph per day (top frame), and the o p t i m a l persistence half-life f o r the same analysis was one day (center frame). T h e t w o o p t i m i z a t i o n s were each performed under the best c o n d i t i o n s f o r t h e other constant. Name-count analysis: The bottom frame shows that the best value f o r the modified persuasibility constant was very close to zero using a persistence half-life o f one day, the persuasive forces o f Figure 5.25, and the population conversion model o f Figure 5.27.

Mean S q u a r e d D e v i a t i o n

( i nPoll

Percent Squared)

Bandwagon k 2 f o r A L L I n f ona ( p e r P a r a - D ayj^ Mean S q u a r e d D e v i a t i o n ( i n P o l l P e r c e n t S q u a r e d ) 9

H a I f - l i f e f o r Bandwagon T e x t A n a l y s i s ( i n D a y a ) Mean S q u a r e d D e v i a t i o n ( i n P o l 1 P e r c e n t s q u a r e d )

Naae C o u n t

k ' 2 f o r A L L I n f ona

( p e rPara-1)a

99

Copyrighte

Figure 5.29. Opinion bandwagon analysis.

on Democratic

candidates

when infons were scored

by the

T h e projections (solid lines) f o r the three groups favoring Mondale, G l e n n , and Others began w i t h the o p i n i o n measurements o f the first p o l l o n June 19, 1983, and continued using the persuasive forces shown i n Figures 5.23 and 5.24, the m o d e l o f Figure 5.26, a n d the o p t i m i z e d m o d i f i e d persuasibility constant o f 1.5 per A P paragraph per day. Calculations were at six-hour intervals. Actual p o l l points are plotted as squares.

X F a v o r MONDALK_(Bandwagon T e x t A n a l y s i s )

% F a v o r GLENN (Bandwagon T e x t A n a l y s i s )

% Favor OTHERS (Bandwagon T e x t A n a l y s i s )

Jul Years

Sep 1983-1984

Nov

Jan

100

Copyrighted material

Figure 5,30. count only.

Opinion

on Democratic

candidates

when infons were scored

by

name

T h e projections (solid lines) for the three groups favoring Mondale, Glenn, and Others began w i t h the o p i n i o n measurements o f the First p o l l on June 19, 1983, and continued using the persuasive forces shown i n Figure 5.25, the model o f F i g u r e 5.27, and the same m o d i f i e d persuasibility constant o f 1.5 per A P paragraph per day used for Figure 5.29. Calculations were at six-hour intervals. Actual poll points are plotted as squares. %

Favor

MONDALE

%

Favor

OTHERS

J u l Years

(Name

(Naae

Count

Count

Sep

A n a l y s i s )

A n a l y s i s )

Nov

Jan

1983-1984

101

c

Figure 531. Persuasive forces from AP paragraphs favoring better, same, and worse economic conditions. Paragraphs favoring these three positions were scored (see

Appendix C, Section C.S) and used for the construction o f persuasive force curves employing a one-day half-life and equal weights o f 1.0 for a l l infons. Infona

Favor

Economic

C l i m a t e

B e t t e r

(AP

I n f o n s F a v o r Economic C l i m a t e Same (AP

Paras)

Paras)

8

Infona

Favor

Economic

C l i m a t e

Worse

(AP

Paras)

2

I

- 5 © - C O Si V3 l/S

0

>1

2 f i e *

** *-

V5 O .

•*

4>

** *

*•

-*

**-

§ 2 T § 4> . O •OcQ °

55

o

4> — £

O

o

o

+4—

4i

3 •2 i t/3

n i/1

5 CO :

y 2

1-2

= =

3 a,

S f— .2 '2


>

1

I

S S -&.S?

103

CoDvriahted material

Figure 5.33. Optimization curves for constants for the economic climate, Usinj: the population conversion model of Figure 5.32 and the persuasive forces of Figure 5.31, ihe best modified persuasLoiliry constant (¿'2) giving the least M S D bad a value of 0.09 per AP paragraph per day (top frame). Trie optimal persistence half-life was one day (bottom frame). Both optimizations were performed under the most favorable conditions for the other constant. Mpari

3quar_GDeviation

1—1

0 "1

Perccnt_Squarcd)

1—1

o.oi

0.001 Mean

( i n Poll

1—r

0.1

1

k ' Z F o r A L L Inform ( p e r Para-Day) Squared D e v i a t i o n ( i n Pol1 Percent. Squared)

40

0 T

!

2 Half-life

1

1 4

1—r—1 6

10

1

1 40

for Econoiic Cliiate

1—r—r 100 ( i n

Dftya)

101 MaTepnan, sauiMmeHHbiH ¿1« re |> - . " < : •• .' •

105 Figure 5.34. Opinion on economic climate. The projections (solid lines) for the three groups feeling that the climate was better, same, or worse began with the opinion measurements of the first poll on March 6, 1981, and continued using the persuasive forces of Figure 5.31, the population conversion model of Figure 5.32, and the optimal constants from Figure 5.33. The computations were performed every 24 hours. Actual poll points are plotted as squares.

%

Believe

Egonniic

Conditions

It

Jji'Hevt!

Ecofloinic

Conditions

i

1

Tear*

1

Rl^ 1981-IBS4

1

Ratter

Worse

1

B3

1

T

#4

Figure 5.35. Persuasive forces of AP infons favoring important, equal importance, and inflation more important.

unemployment

more

Paragraphs favoring these three positions were scored (see Appendix C, Section C.6) and used for the construction o f persuasive force curves employing a one-day persistence half-life and the f o l l o w i n g weights for infons supporting different positions: 1.0 for infons favoring unemployment more important, 0.5 for infons favoring equal importance, and 1.4 for infons favoring inflation more important. Infons

Favor

UNEMPLOYMENT

Infons

Favor

EQUAL

InIons

Favor INFLATION

Jul

78

Import,

(AP

Jul

106

(AP

Paras-Weight

1.0)

(AP P a r r i s - W e i j g h t

Paras-Weight

79

Jul

80

0.5)

1.4)

Jul

S £ £ * s ** ^ t*

í 3

ills i S-sgB h d

?

*

=

I

e^^

Ife" i •ill S

Q H

V

c

È5

_

.s s Işı

c

Figure 5 J7. Optimization curves for the modified persuasibility constant and the infon weighting constants for unemployment versus inflation. U s i n g the population

conversion model o f Figure 5.36 and the persuasive forces o f Figure 5.35, the best m o d i f i e d persuasibility constant g i v i n g the least M S I ) had a value o f 7.0 per A P paragraph per day (top frame). T h e o p t i m a l w e i g h t f o r infons f a v o r i n g equal importance was 0.5 (center frame), a n d that f o r infons f a v o r i n g i n f l a t i o n more important was 1.4 (bottom frame). A l l optimizations were performed under the most favorable conditions for the other constants and w i t h a one-day persistence half-life. Mean

Squared

Deviation ( i n Poll

Percent

Mean

Squared

Deviation ( i n Poll

Percent Squaredl

Weight Mean

f o r Paras Favor

Suuared D e v i a t i o n ( i n P o l l

Weight

f o r P a r a s Pavor

EQUAL

Squared)

Impor tanot?

Percent

INFLATION

Squared)

Iaportant

10K

Copyrighted material

Figure versus

5.38. Optimization curve for the persistence constant for unemployment inflation. U s i n g the same conditions and the o p t i m a l constants f r o m Figure

5.37, a persistence constant o f one-day gave the least M S D .

Mean S q u a r e d

Deviation { in P o l l

Percent

Squared)

H a l f - 1 i f e .for Uncmploy. va. I n f 1 a t . ( i n Days)

109

Copyrighted material

Figure 5.39. Opinion favoring unemployment more important, equal importance, or inflation more important The projections (solid lines) for the three opinions began

w i t h the measurements o f the first p o l l on M a r c h 22, 1977, and continued using the persuasive forces shown i n Figure 5.35, the population conversion model o f Figure 5.36, and the o p t i m a l constants f r o m Figures 5.37 and 5.38. The computations were performed every 24 hours. Actual poll points are plotted as squares. X_Bo1 i o v e

X

Be 1 i e v e

% Believe

UNEMPLOYMENT

EQUAL

Important

Impor tance

INFLATION

J u l Years

more

more

78 J u l 1977-1980

Important

79

J u l

80

J u l

110

Copyrighte

Figure 5.40. Persuasive opposing Contra aid.

forces

of AP infons scored

Paragraphs for

by the author as favoring

Fan

and

these t w o positions were scored by (see Appendix C, Section C.7) and used for the construction o f persuasive force curves employing a one-day persistence half-life, a weight o f 1.0 for infons favoring aid, and a weight o f 2.0 for infons opposing aid.

InfonH

FAVOR__Contra _ A i d

( AP

Paraa— Fan

Scores)

16 30 20

10

li

0

I n f o n s OPPOSF C o n t r a Aid (AP ParaB^-Fan

Scores)

40

30 2 0

E

10

0

83.5 84 Y«»H r-K

84.5

85

85.5

86

111

C

I

Figure 5.41. Persuasive forces of AP infons scored by Swim, Miene, and French as favoring and opposing Contra aid. Paragraphs far these two positions were scored (see Appendix C, Section C,7) and used for the construction of persuasive force curves employing a one-day persistence half-life, a weight of 1-0 for infons favoring aid* and a weight of 2A for infons opposing aid. Infojiri

FAVOIt A i d _[ AP _ P a r i i a _ - - S w i m n t

a [

SuoresJ

60

83.5 Years

HA

H4.5

85

85.5

Hi)

112 M i l T C P H a n . 3iiUJHLJJCHHblH ilR TC[I

L

" ' i

I

•

Figure 5.42. Population conversion model for actions of infons favoring and opposing Contra aid. The boxes denote the subpopulations under consideration. The

words i n the boxes begin w i t h " B " to refer to those " b e l i e v i n g " or having an o p i n i o n favoring o r opposing Contra aid. The arrows indicate o p i n i o n conversions due to the persuasive forces (beginning w i t h " G ) shown i n Figures 5.40 and 5.41. H

*************** *

GOpi*>se

*

* BFavor * * * ***************

*************** >

< GFavor

113

*

*

* BO| >pose * * * ***************

Figure 5 A3. Constant optimization curves for Contra aid. The M S D was calculated for o p i n i o n projections using both infon scores obtained by the author as plotted i n Figure 5.40 ( s o l i d lines) and b y S w i m , Miene, and French as plotted i n Figure 5.41 (dotted lines). For all calculations, the population conversion model was the one i n Figure 5.42. A one-day persistence half-life was used for determinations o f the best m o d i f i e d persuasibility constants (k'2) favoring Contra a i d (top frame)--1.2 p e r AP paragraph per day for Fan infons and 1.6 for S w i m et al. infons—and calculations for the optimal weights for infons opposing Contra aid (center frame)--2.0 for Fan infons and 2.4 f o r S w i m et al. infons. T h e o p t i m a l values f o r the other constants were used for c o m p u t i n g the best persistence half-lives (bottom frame)--forty-one days f o r Fan infons and seven days for S w i m et a l . infons.

Mean Squart?d D e v i a t i o n ( i n P o l 1 P r e c e n t 100

Squared)

80 60 40 20 0 0.01

0.001

0.1

1

10

k'2 f o r FAVOR C o n t r a A i d ( p e r P a r a - D a y ) Mean S q u a r e d D e v i a t i o n ( i n P o l l P r e c e n t S q u a r e d ) 350 300 250 200 150 100 50 0

J 1.5 2 2.5 3 Weight f o r P a r a a OPPOSE C o n t r a A i d Mean S q u a r e d D e v i a t i o n ( i n P o l l P r e c e n t S q u a r e d ) 120 100 80 60 40 20 0

1

2 4 6 8 10 20 40 Lt.-JL TÇ f o r C o n t r a A i d _ ( j n Dayn) 1

114

Copyrightec

115 Figure 5.44. Opinion favoring and opposing Contra aid using infon scores by the author. The projections (solid lines) for the two opinions began with the opinion measurements of the first poll on August 20, 1983, and continued using the persuasive forces shown in Figure 5.40, the population conversion model of Figure 5.42, a persistence half-life of one day, and optimal values for the modified persuasibility constant favoring Contra aid and the weight for infons opposing aid ( Figure 5.43, solid lines). Computations were performed every 24 hours. Actual poll points are plotted as squares.

Figure 5.45. Opinion favoring and opposing Contra aid using infon scores by Swim, Miene, and French, The projections (solid lines) for the two opinions began with the opioion measurements of the first poll on August 20, 1983. and continued using the persuasive forces shown in Figure 5A\, the population conversion model of Figure 5.42. a persistence half-life of one day, and Optimal values for the modified persuasibility constant, favoring Contra aid and the weight for infons opposing aid (Figure 5.43 doited lines). Computations were performed every 24 hours. Actual poll points are plotted as squares. % PA Villi q o n t . r i i

A i d _£Sw_i m_ rt_

n 1 __Soprriii_}

(it) •

20 T

0 % OPPOSE HO 60

Cn_pt.rn.

Aid

(Swim

el. a l

St-nros)

-

10 • 20

'

0 83.5

T

"I

P

1

84

H4.5

85

85.5

T86

M a TOP n a n , 3iiiiinLiicnHbin a B T o p c « w i * • i

:••

•

6

Methodological Significance of Work

The previous three chapters have applied computer lent analyses and ideodynamics equations to project expected public opinion. The present chapter examines the methodological implications of the InfoTrend procedures. The next chapter will consider the theoretical significance of the results. The studies in this book focused on public opinion where there was relatively little social or economic cost to persons changing their minds. A deliberate choice was made to avoid economic issues, including product purchase, because such activities required weighing such complicating factors as competing uses for financial resources. Messages due to these important factors were difficult to study directly. The introduction of innovations into society was also not studied because they frequently involved not only economic considerations but also social factors. While economics was obviously important for the classical studies of the adoption of hybrid com (Ryan and Gross, 1943), investigations on the acceptance of family planning (Berelson and Freedman, 1964) also had to contend with complex societid forces related to sexuality and reproduction. In contrast, people were unlikely lo have deep convictions for the issues studied in this book. Members of the population as a whole clearly have no good idea how much should be spent for defense. Troops were only briefly in Lebanon, with little time to form ingrained prejudices. None of the Democrats running for president had ever held that office, so none had the advantages of incumbency. The public was well aware that the economy could change, so there was no reason to feel that the economic climate should always be good or bad or that unemployment should always be more or less important than inflation. As with Lebanon, the Contras were in a distant land with which few Americans had personal contact and about which there was little inherent opinion. These theoretical arguments for opinion malleability and volatility were bolstered by actual findings that there were .substantial opinion movements in all examples except that of Contra aid. In fact, one of the reasons for studying these other ins lances was that opinion did Change, thereby providing the most critical tests of the methodology. However, it was also useful to demonstrate that the calculations could project unchanging opinion for Contra aid when constancy was actually observed The model would definitely have been weakened by data showing that there was a change in the ratio of favorable to unfavorable information while there was no simultaneous change in opinion. Although these studies showed the applicability of the calculations to issues where people hold shallow and changeable convictions, the same methodology should Theoretically succeed even for more firmly held beliefs i f all the relevant messages

M a T e p n a n . saLaumeHHbiti aBTopc«wi* • :

• •

/18

Predictions

of Public

Opinion from

the Mass

Media

available t o the p u b l i c can be coded. F o r f i r m l y held beliefs, the persuasibility constants w o u l d s i m p l y be decreased so that more i n f o r m a t i o n w o u l d be needed t o cause an o p i n i o n s h i f t

6.1 V A L I D A T I O N O F I D E O D Y N A M I C S Since n o applications o f ideodynamics have been described previously, i t was important to validate the model using not o n l y logical argument but also empirical tests. Such tests benefited f r o m one o f the unusual capabilities o f the m o d e l - i t s ability to consider the time dimension d i v i d e d i n t o infinitesimally small intervals. A s noted i n Chapter 2, this was possible because o p i n i o n calculations were not based solely o n the i n f o r m a t i o n available t o the p o p u l a t i o n , b u t also o n the o p i n i o n i n the previous t i m e i n t e r v a l . T h e use o f s m a l l t i m e intervals, l i k e the 6- o r 2 4 - h o u r periods i n this book, permits the tracking o f r a p i d changes l i k e that f o r troops i n Lebanon, where opinion favoring more troops changed f r o m under 7 percent to over 30 percent and back d o w n t o under 9 percent, all w i t h i n a few days. T h e use o f small time intervals is important t o e m p i r i c a l tests o f ideodynamics because the number o f o p i n i o n predictions increases and the t i m i n g o f the computed o p i n i o n values becomes more precise as the intervals f o r the o p i n i o n calculations decrease i n size. The result is more o p i n i o n projections at more closely defined times. These very precise predictions can then be tested against measured poll data. I n the actual cases studied, the number o f times o p i n i o n values were calculated for each time series ranged from approximately 400 f o r troops i n Lebanon to over 3,000 f o r defense spending. Therefore, the model c o u l d be tested b y i t s a b i l i t y t o m i m i c poll data for hundreds to thousands o f time points. For each o f the cases i n this book, it was necessary t o assign one to three i n d e p e n d e n t p a r a m e t e r s i n a d d i t i o n t o the persistence constant, w h i c h was set t o h a v e a one day h a l f - l i f e f o r a l l studies. Therefore, the m i n i m u m number o f p o l l points needed t o set the parameters s h o u l d be those s u f f i c i e n t t o g i v e one t o three independent opinion measurements. A t each p o l l time, there were t w o t o four polled positions. O p i n i o n f o r one o f the positions c o u l d be determined by subtraction f r o m 100 percent, so the number o f independent o p i n i o n measurements was one less than the polled positions. A s a result, a time series w i t h three p o l l measurements w o u l d y i e l d a m i n i m u m o f three independent p o l l percentage values (the Contra case, w i t h o n l y t w o positions, f o r or against aid). Therefore, three well-spaced polls w i t h t w o o r more positions per measurement should be sufficient to establish the three parameters i f the polls had n o errors. I n addition, the p o l l percentages at the earliest time i n the series were used as the starting point for the calculations, so o p i n i o n values at this t i m e were the boundary conditions also needed f o r setting the parameters. Therefore, i n the worst case, a p o l l series w i t h f o u r t i m e points should uniquely define the three parameters. O n e o f these poll measurements w o u l d correspond t o the i n i t i a l p o l l conditions, a n d the other three p o l l points w o u l d be used t o set three parameters. I n the best cases, w i t h o n l y one parameter aside f r o m the consensus half-life o f one day (defense spending. Democratic primary, and economic climate), o n l y t w o accurate and well-spaced poll points w o u l d be needed, in principle, to define the parameters. Once the parameters were set, any additional p o l l points c o u l d n o longer be f i t by adjusting the parameters and w o u l d c r i t i c a l l y test the model e m p i r i c a l l y . T h i s estimate o f needing p o l l series o f t w o t o f o u r points t o establish one t o three parameters is only approximate because the o p i n i o n values are not truly independent Instead, o p i n i o n at any time is dependent o n o p i n i o n at earlier times. Nevertheless, these arguments d o indicate that series w i t h eight (Democratic p r i m a r y ) t o s i x t y - t w o (defense spending) time points d i d indeed provide meaningful empirical tests o f the model.

Copyrightec

Methodological

Significance of Work

119

Since ideodynamics uses a number of apprarirnations, it will not always be clear which are faulty if the model does not give satisfactory predictions. In contrast, generally accurate calculations for a number of issues will render plausible the total constellation of approximations* The quality of die Tit shown in the tables and figures of Chapter 5 indicates that the methodology has good predictive powers. It was also significant that the methods in this book were successful for si* oui of s i l cases tested. In consequence, the collection of approximations used for the computations stands a reasonable chance of being valid. Since the empirical tests for ideodynamics involved comparisons between poll results and opinion computed from information available to the population, it was essential that scores for persuasive messages be obtained independently of opinion measurements. For this reason, four precautions were taken: 1. The conditions for the infon scoring of Chapter 4 were developed by examining AP stories in random order so that the analyst would not be tempted-consciously or unconsciously-to bias scores with the goal of fitring F * " results. 2, The same computer scoring was applied to all stories, so any changes in scoring rules could not be applied p referen liai I y to a chosen subset of Stories in order to achieve scores which would result Ln good opinion predictions, 3- The vocabulary and scoring rules had to be logically defensible. For instance, word combinations implying less spending could not be used to score for more spending. A. For the example of Contra aid, the Roper Center was first contacted to establish that there was likely to be enough polls to construct a reasonable time series for the model testing. However, no poll values were actually obtained until the text analysis was finished both by Fan and by Swim, Miene, and French. Only after completion of the text analysis were actual poll percentages obtained frnm the Roper Center. Therefore, for this example, there was absolutely no way for the analysts to adjust the scoring to match measured poll values. 0

6.Z

DATA AND ISSUES CALCULATIONS

FOR

SUCCESSFUL

1DEODYNAMIC

Since methodological development is one of the significant aspects of this book, it is useful to consider the appropriate conditions for applying the method"logicsOne important suggestion from the studies presented above is that ideodynamics is applicable to a wide variety of issues. After all, the model was successful for issues drawn from areas as diverse as foreign policy (troops in Lebanon and Contra aid), economic issues (economic conditions and the importance of unemployment versus inflation), domestic policy {defense spending), and political campaigns (Democratic primary). The accuracy of ideodynamics derives from the fact that the calculations include all relevant persuasive messages. All the issues just mentioned shared the condition that the mass media were likely to contain the majority of them. Other informational sources such as books or éducation in schools could not keep abreast of the pertinent news as it was being generated. The main other sources which had the potential to inform as rapidly were personal experience and underground means of communication such as ru mors. Personal experience was unimportant for all the issues studied. For the Democratic primary in the early stages, defense spending. Contra aid, and troops in Lebanon, it was plausible that the mass media were likely to be the primary sources of all pertinent communications. For the economic climate and the importance of inflation versus unemployment, on the other hand, personal experience and observation might have been expected to provide significant messages. However, the empirical resting showed that reasonable opinion time trends could be calculated

Ma

TOP Ha n , sawHiueHUbiH a B T o p c i w i * • :

•

•

¡20

Predictions of Public Opinion from ike Mass Media

without modifications to account for personal experiences. In other words, personal experience messages could be ignored. This result suggested that information in the media might color substantially an individual's interpretation of personal experiences. For instance, m both the best and worst of limes, people will be aware of others out Of work. When the economy is considered to be good, people might interpret a person's unemployment to be his own fault. However, with bad economic news, people might well shift the blame from the individual to general conditions. Word-of-mouth communication such as rumors would have been important if the mass media were not trusted by the population. Indeed, if a government controlled press is perceived by the public to be biased and untrustworthy, then alternative persuasive messages will become important. Examples would range from underground publications in totalitarian states to rumors during time of war. Reliance on rumor and an underground press is probably minimal in the major democracies at this time. Another implication of these calculations is that successful calculations can frequently be made using only AP messages. The tests in this book deliberately made the extreme simplifying approximation that AP stories could represent all relevant persuasive messages for the topics studied. If this approximation is valid for a large number of diverse topics, then the AP İs indeed likely to be representative of most of the news in the mass media in agreement with the finding of similarity in much of the mass media by Paletz and Entman (1981), Furthermore, opinion calculations will have been shown to be independent, generally speaking, of special considerations for special events. Indeed, among all six examples in this book, it was only necessary to add one non-AP infon once. That addition was the infon favoring more troops being sent to Lebanon after the truck bombing of American Marine headquarters and was needed to calculate the great increase in opinion just after the truck bombing. In general, the success of the empirical tests argues for the appropriateness of assuming ihat (he AP could represent both the print and electronic media except in very rare cases Like the truck bombing incident. Previous investigators have frequently used the New York Times or the Vanderhi 11 summaries of television news (e.g., MacKuen, 1981; Ostrom and Simon, 1935: Page, Shapiro, and Dempsey. 1985, 1987). There might have been minor differences between the news content İn the AP and the New York Times or television broadcasts. However, the variations were probably not large. Nevertheless, there was likely to be a significant difference between the news stories used for this book and those identified as relevant by human judges. With human judges, there is probably preferential identification of stories concentrating on the topic under study. Stories with oblique inclusion of pertinent messages are Likely to be ignored. The retrievals for this book did not have this bias since the full texts of all dispatches in the Nexis data base were searched using combinations of key words chosen by the investigator (Chapter 2). AH phrases relevant to the topic were identified even if they were minor components or a story mainly discussing some other topic. Since only text in the region of discussions of the relevant issue was collected, the number of retrieved words per dispatch gave a good idea of the fraction of the typical dispatch devoted to the question. The typical AP dispatch had 400-900 words. About 420 words were retrieved per average dispatch for troops in Lebanon, consistent with the observation that a large number of these dispatches were devoted mainly to this topic. For the Democratic primary and Contra aid, those numbers were 310 and 258, respectively, already significantly less than a full story, while the equivalent values for defense spending, the economic climate, and unemployment versus inflation were all under 200, meaning that less than half of the typical story was on these topics. This inclusion of articles mainly about other topics was appropriate since members of (he general public were frequently not looking for (ex( on particular polled topics and thoughts were probably absorbed from whatever was read regardless of whether the ideas were surrounded by similar or dissimilar information.

Ma

TOP n a n , sauiHmcHHbiH aBTopc«wi* • :

•

•

Methodological

Significance

121

of Work

The computer searches not o n l y selected articles in which other topics were the m a i n focus, but such a search also guaranteed consistency. U n l i k e a human j u d g e w h o m i g h t have been distracted when the major emphasis was on another subject, the computer always found the programmed w o r d combinations. The thoroughness o f the data base searches was seen i n the identifications ranging f r o m 1,156 dispatches for Contra aid to 12,393 for the economic climate. S i m p l y to read these numbers o f articles w o u l d have been a daunting task for any human investigator. F r o m a methodological standpoint, i t is very useful that the A P alone was able to represent all persuasive messages aside f r o m the truck b o m b i n g example. T h i s f i n d i n g indicates that calculations can usually succeed w i t h o u t r e q u i r i n g ad hoc adjustments to account for special events. The a p p l i c a b i l i t y o f the methods to six quite varied topics suggests that many future computations are also l i k e l y t o be generally v a l i d i f o n l y A P messages are analyzed.

6.3

P O S I T I O N S FOR W H I C H P E R S U A S I V E M E S S A G E S A R E S C O R E D

Once the issues have been defined and the relevant persuasive messages have been assembled, it is then necessary to obtain scores for the infons i n messages. As noted i n A p p e n d i x A, there w i l l be overlap between the positions that infons are favor and those that people are to support i n o p i n i o n polls. However, the overlap need not be complete. For instance, i n the bandwagon scoring for the Democratic p r i m a r y , there were people w i t h N o O p i n i o n and no infons f a v o r i n g this p o s i t i o n , w h i l e there were infons unfavorable to M o n d a l e and no persons polled to have this viewpoint. For the other cases, the overlap was m u c h greater. For Contra aid, both infons and subpopulations either favored or opposed aid. In the remaining cases, there were three p o l l positions, r a n g i n g f r o m one extreme t h r o u g h the center to the other extreme. A n example was positions o f more, same, and less spending for defense. Here, infons were also scored to favor either the same three positions o r the t w o positions o f more and less spending. Since all messages were considered to contain different infons, each w i t h a content score r a n g i n g f r o m zero t o some positive n u m b e r o f paragraphs, some A P dispatches were m i x e d messages w i t h positive scores for t w o or more infons. F o r issues w i t h t w o extremes and one central p o s i t i o n , a neutral message f a v o r i n g the central idea was d i s t i n g u i s h e d f r o m a message w i t h t w o equal components favoring the t w o extreme positions. In the First case, the dispatch w o u l d have had a positive score for the neutral position. In the second case, the story w o u l d have had t w o positive scores, one i n favor o f each o f the extremes. Both types o f scores were found and provided a more subtle means for extracting information f r o m messages than w o u l d be possible by scoring a message w i t h pro and con components as being equivalent to a neutral message. As a r g u e d i n Chapter 2, a m i x e d pro and con message should not have the same persuasive effects as a neutral message. Since dispatch scores were given i n paragraphs, stories w i t h more relevant paragraphs had higher scores. Therefore, the dispatch content scores were weighted i n proportion to message salience.

scored to

measured

6.4 C O M P U T E R T E X T S C O R I N G A l t h o u g h human j u d g i n g can be used to obtain position scores for A P stories, this book relied on a computer method to guarantee u n i f o r m i t y o f scoring. Since the techniques and ramifications o f the text analysis have already been presented i n Chapter 4 and are i n A p p e n d i x C, they w i l l not be repeated here. However, as an o v e r v i e w , there are several important novelties i n the I n f o T r e n d computer text analysis.

Copyrighte

122 Predictions

of Public Opinion from the Mass Media

One feature is the decision to return to the idea o f t a i l o r i n g the analyses to individual issues instead o f using a generalized computer content analysis program with a fixed dictionary and invariant scoring rules. The extra rime and effort spent i n making up specific dictionaries and rules were compensated for by the lack o f need to interpret the results, a situation opposite f r o m that with predefined dictionaries and rules. The method was also quite general, being applicable to any texts, i n c l u d i n g those not examined f o r o p i n i o n f o r m a t i o n . T h e possible analysis o f letters o f recommendation has already been mentioned i n Chapter 3. T h e generality o f the method was not at the level o f the dictionary or rules, b u t rather at the level o f strategy, including text nitrations as key elements. Precision i n the analysis was greatly aided by the use o f repeated text nitrations to r e m o v e irrelevant text. The r e m a i n i n g text was more homogeneous, thereby s i m p l i f y i n g subsequent steps and p e r m i t t i n g the use o f words w h i c h w o u l d be ambiguous i n a general setting. These f i l t r a t i o n steps were also useful for incorporating relationships between words i n the input text for f o l l o w i n g steps. The InfoTrend procedures were also surprisingly robust i n t h a i apparently important changes i n the dictionaries and rules d i d not affect the basic shape o f the curves fur predicted opinion. Thus, dispatches for defense spending could be senred either to favor more, same, and less spending or just more and less spending. Also, nuclear arms reduction c o u l d either be interpreted to favor Less defense spending or not. A number o f words in the scoring dictionaries c o u l d be changed. Yet the basic opinion projections stayed more or less the same. Therefore, there was no need to be overly concerned that words or rules were either o m i t t e d or misassigned d u r i n g construction o f the dictionaries and rules. A n equally compelling argument for the robustness Of the text analysis was the fact that Fan and S w i m el al. could independently arrive at different dictionaries and rules for the text analysis for Contra aid (Chapter 4), The dictionaries and rules were quite different for the t w o analyses, w i t h S w i m et aL using quite different strategies and assigning scores to twice as many paragraphs as Fan. Despite the differences, the resulting o p i n i o n projections were essentially the same, suggesting that the text presented the same thoughts in a number o f alternate ways. I n addition* strong conclusions c o u l d be made about the message components critical for persuading the public since the computer applied the dictionaries and rules b l i n d l y . This was an important advantage o f analyzing the text using a structure where all dictionary words and rules had to be j u s t i f i e d o n logical grounds. N o artificial considerations c o u l d be introduced for the sole purpose o f f i n i n g desired output results. The dictionaries and rules for the text analyses were a l l constructed o n l y b y l o o k i n g at die retrieved t e x t without the application o f expert knowledge specific to the topics under study. This strategy was reasonable since the goal was to study o p i n i o n i n the general public, comprised mainly o f nonexperts. The success o f the calculations for six out o f six issues, despite the absence o f expert knowledge, suggests that appropriate d i c t i o n a r i e s a n d rules can be made i n general by nonspccialists, Besides the omission o f possibly pertinent information outside o f the messages themselves, portions o f the text indirectly supporting a position were also ignored w i t h t w o exceptions. One was the interpretation o f waste and fraud infons to support less defense spending, and the other was the i n c l u s i o n o f indirect messages i n the scoring for Contra a i d by S w i m et al. Otherwise, i t was sufficient to include o n l y •hose phrases w i t h text directly arguing for a position, phrases analogous to the verv b r i e f extracts o f f i l m reviews i n advertisements (e.g., best o f the year..New York

Times'), I t is l i k e l y that there was usually a h i g h correlation between indirect and direct messages, since the same opinion calculations for Contra aid were obtained f r o m text analyses by S w i m et al* using many indirect messages and by Fan scoring o n l y those w o r d clusters directly taking a position on the desirability o f aid (Chapter 3), The

MaTepnan,

aauiMmeHHbiH

as rt p - . <•

npasoM

Methodological

Significance

of Work

123

inclusion o f indirect messages meant that about t w i c e as many paragraphs were scored b y S w i m et al. The similarities i n the o p i n i o n projections using either set o f scores indicated that the message structure was the same either i n c l u d i n g or e x c l u d i n g many indirect messages. The most plausible explanation is that direct statements advocating a position are usually coupled w i t h justifications c i t i n g information bearing indirectly on the issue. I n this case, the result w i l l be the same i f either the advocacy o r the j u s t i f i c a t i o n s are scored. A p r o b l e m w i l l o n l y arise i f statements f a v o r i n g one position are j u s t i f i e d w h i l e those favoring another are not. H o w e v e r , given the general effort o f the American mass media to use a neutral tone, a consistent bias o f this sort is u n l i k e l y . The o n l y case i n c l u d i n g a n o n - A P infon was that for troops i n Lebanon, where the very unusual terrorist truck b o m b i n g o f October 23 generated an i m p o r t a n t indirect i n f o n favoring more troops. I n the case o f the Democratic primary, indirect messages due t o name counts were not g o o d predictors o f p u b l i c o p i n i o n . Furthermore, i t was quite satisfactory to ignore the positions taken b y the candidates o n campaign issues. The display o f computer message scores as i n f o n persuasive force curves is a convenient pictorial method for visualizing message structure. T h i s was very useful for s h o w i n g that i n f o r m a t i o n o n defense waste and fraud was n e g l i g i b l e when compared to d i r e c t statements f a v o r i n g less defense spending ( F i g u r e S.12). Persuasive force curves also illustrated the a b i l i t y o f the computations i n this book to assess the relative influences o f different types of messages.

6.5

IDEODYNAMIC CALCULATIONS OF OPINION T I M E TRENDS

Once the issue and its positions are defined and the persuasive messages scored, i t is t h e n possible t o c o m p u t e expected o p i n i o n t i m e trends. F o r these computations, a number o f approximations are made. T h e i r v a l i d i t y derives f r o m the e m p i r i c a l tests i n this book c o m p a r i n g o p i n i o n t i m e trends w i t h measured p o l l values. One reasonable approximation is that the A P can be assigned a u n i f o r m v a l i d i t y for all stories on an issue. The validity score in ideodynamics refers to the credence given by the public to the m e d i u m carrying the message. The approximation for these studies was that the A P and the mass media in general were highly credible and had the same high v a l i d i t y score for all stories for any one issue throughout the time periods o f the opinion trend calculations. A more contestable approximation was that all sources quoted i n the A P have the same weight. A l t h o u g h ideodynamics permits different senders o f infons to be given different weights, this w e i g h t i n g was not needed for acceptable t i m e trend calculations. Instead, successful opinion computations c o u l d be made w i t h die same degree o f persuasiveness assigned to all infon sources, including news commentators, members o f Congress, and the President o f the U n i t e d States. In contrast. Page and Shapiro (1984) and Page, Shapiro, and Dempsey (1987) have suggested that such weights m i g h t be useful because popular presidents and news commentators have greater influence. H o w e v e r , the data i n this book suggest that even a popular president like Ronald Reagan was less persuasive than his opponents for issues like defense spending and Contra aid. I n both these cases, a better f i t t o the o p i n i o n data w o u l d have required that there be a l o w e r w e i g h t for messages f r o m Reagan and his administration (Chapter S). F o r Contra aid, infons opposing the administration's position were more man twice as strong as infons favoring that position. A methodological difference might explain the differences between the results o f this book and those o f Page, Shapiro, and Dempsey. As discussed earlier, the manual searches used b y these other investigators m i g h t not have systematically included messages where o n l y a small fraction was relevant to the issue. T y p i c a l l y , infons f a v o r i n g different positions o f a given issue are about as powerful. O n l y f o r h a l f the i s s u e s - t r o o p s i n L e b a n o n , C o n t r a a i d , a n d

Copyrighte

124

Predictions

of PMic

Opinion

from

the Mass

Media

unemployment versus inflation—was it necessary to give different weights for i n f o n scores favoring different positions (Table 5.3). B y least squares o p t i m i z a t i o n , i t was s h o w n that the m a x i m u m r a t i o between the m i n i m u m and m a x i m u m r e f i n i n g weights for any one o f the six issues was less than threefold. Part o f these weight differences was probably due to different degrees o f effectiveness for infons supporting different positions. This was especially l i k e l y for Contra aid, where different scoring methods by Fan and S w i m et al. both showed that infons opposing aid were twice as effective as infons favoring aid. However, the need for different weights for different groups o f infon scores might also have resulted i n part f r o m imprécisions i n the computer text analyses. No computer p r o g r a m , however good, can be expected to reflect a l l the subtleties o f natural language. T h u s i t is possible that infons f a v o r i n g some positions were consistently over- or underscored and that scores for some positions should actually have been assigned to two or more infon persuasive force functions (Appendix A ) . In the studies i n this book, no i n f o n scores were assigned to t w o different persuasive force functions. One o f the most interesting conclusions is that an A P infon's audience size typically decreases exponentially w i t h a persistence half-life o f one day. As discussed in Chapter 1, the persistence constant describes the rate at w h i c h mass media infons like those f r o m the AP lose their effectiveness. F r o m Chapter 5, this half-life had a consensus value o f one day. Since message action disappeared so rapidly, i t seems reasonable that all opinion reactions were immediate. The memory o f messages was probably o n l y important i n the very small subpopulation o f individuals caring deeply about particular issues. For each issue, that subpopulation was liable to be different since any one person o n l y has the t i m e to be v i t a l l y concerned about a very smalt number o f the issues for w h i c h polls are taken. As noted i n Chapter 1, the number o f converts depends on an infon's content, validity, and audience size. The strength o f an i n d i v i d u a l infon at any particular time is given i n a time-dependent persuasive force function constructed by m u l t i p l y i n g the content score, the v a l i d i t y score, and the audience size at that t i m e . Ideodynamics postulates that the number o f people persuaded is p r o p o r t i o n a l to this persuasive force function. A l t h o u g h not e n t i r e l y e x p e c t e d , i t was e x t r e m e l y useful that o p i n i o n reinforcement and message saturation can be ignored. W h e n the population receives more than one i n f o n , i t is necessary i n principle to consider possible interactions among infons. As discussed i n Chapter 1, such interactions c o u l d arise f r o m o p i n i o n reinforcement due to the resolution o f cognitive dissonance i n favor o f reinforcing messages. A l s o , interactions c o u l d lead to infons b e c o m i n g less effective at saturating densities. For s i m p l i c i t y , the calculations i n this book w e r e performed assuming that there were n o such interactions. A s a result, persuasive force functions for i n d i v i d u a l infons favoring a position were added to şive the net persuasive force function i n favor o f that position. In a natural extension o f the case for i n d i v i d u a l infons, the fraction o f the population converted is assumed to be proportional to the persuasive force function for all infons favoring the position under consideration. Ideodynamics permits the d i v i s i o n o f the p o p u l a t i o n i n t o different types o f subpopulations depending o n the issue. The details o f the o p i n i o n change are specified by a population conversion model detailing the population conversions due to the calculated persuasive force functions. For each issue there is a different model, w i t h the m o s t c o m m o n models i n v o l v i n g people expressing o p i n i o n s o n a continuous scale f r o m pro to c o n . I n this case, the t w o major variants are the sequential (defense spending. Figure 1.2; and economic climate, Figure 5.32) and the direct conversion (troops i n Lebanon, F i g u r e 5.15; name-count analysis for the Democratic p r i m a r y , Figure 5.27; and unemployment versus i n f l a t i o n , Figure 5.36) models. For the more complicated bandwagon analysis for the Democratic primary, a m i x e d model was employed (Figure 5.26), and for the very simple case o f Contra aid (Figure 5.42), the m o d e l was the sequential and direct conversion models collapsed into one.

Copyrighte

Methodological

Significance of Work

i 25

Usually the sequential variant was preferred whenever there was a large number of adherents lo the intermediate position between two extremes without a large number of messages favoring that position. High frequencies for die central position could then result from traffic between the extremes with some persons having the intermediate opinion en route. Besides being dependent on the persuasive force functions, the number of converts also increases with the sizes of the susceptible target subpopulations. If everyone is alreadv convinced, there would be no target population and no more recruits could he obtained regardless of the effectiveness of the messages. At the other extreme, if no one favored the position, then the number of potential converts is large and an infon favoring the position could convert a large number of people even if that was a very small fraction of the total possible recruits. As discussed for the sin examples studied, the subpopulations corresponded to the possible responses to poll questions. The more the response categories, the more were the population subdivisions. There would have been no difficulty in dividing the population into many different positions if there were many possible poll responses. However, there is always the practical limit that the sample sizes will be too small for accurate poll measurements if the category number becomes too large. For this reason, the poll data for the minor candidates were pooled for the Democratic primary. In computing the number of people converted, the constant of proportionality is the modified persuasibility constant adjusted using refining weights. As discussed in Chapter 1 and Appendix A, the modified persuasibility constant is the multiplier for the persuasive force function and the subpopulation size in the equations for computing opinion time trends. A number of important variables are all incorporated into the modified persuasibility constant for mass media infon induced opinion change (Appendix A, Equation A25). First of all, the modified persuasibility constant includes the validity of the medium. As stressed in Chapter 1, the validity of a message is restricted to that of the medium. As argued above, the rapidity of the population response makes it unlikely that opinion Leadership was of great importance for the issues in this book. Therefore* the medium for these studies was the mass media in general and the AP in particular. For opinion calculations* infon validity need not have any relationship to whether the public thinks the press is credible. The mass media need only have a sufficient reputaiion that other sources of information, such as rumor, do not play an important role. If press credibility drops so far that rumor needs to be included, then ideodynamics still should be able to mate reasonable predictions if the rumors can also be coded. Secondly, the modified persuasibility constant includes the initial audience size, For mass media infons, the initial audience size refers to the number of people exposed shortly after the message broadcast. The approximation in this book was that all AP messages had the same approximate audience size. It was only critical for the predictions that the audience size stay constant for each issue analyzed. The projections for any one issue would not have suffered if the audience sizes were generally lower for some issues than others, even for mass media messages, since the audience size is also incorporated into the modified persuasibdity constant. Finally, the refining weights temper the modified persuasibility constant. Chapter 1 and Appendix A note that every issue is characterized by a persuasibility constant describing ihe extent to which the issue is close to the core beliefs of the population- The more central the issue, the lower is the oersuasibility constant. However, it was also noted that there could be differences in ihe persuasive powers of different infons for different target populations even for a single issue. These differences in the persuasibility constant were modeled by multiplying the persuasibility constant for a target population and a corresponding infon persuasive force function by an appropriate refining weight [n obtain the final constant by which persuasive force functions and target population sizes were multiplied. Based on the studies in this book, there were probably only small differences among the

M a T e p n a n , aauiHmcHHbiH aBTopc«wi* • :

•

•

126

Predictions

of Public

Opinion from

the Mass

Media

persuasibility constants for any one issue since the ratio o f the largest to the smallest refining weights for an issue d i d not exceed threefold. Besides minor differences in the persuasibility constant, the refining weights also corrected for any variations i n i n f o n scoring. For example, i f some infons were consistently over- or underscored, then those differences could be normalized using r e f i n i n g weights. A g a i n , since die r e f i n i n g weights were n o t large, message misscoring was probably also not great. Since the r e f i n i n g weights were usually not large, the o p i n i o n calculations t y p i c a l l y began by assuming that a l l infons were scored p r o p e r l y a n d that all persuasibility constants were the s a m e - i n other words, that the refining weights were 1.0 for all combinations o f persuasive forces and target populations leading to o p i n i o n conversion. Fortunately, this simple approximation c o u l d be used for three o f the six cases studied. For the other three examples, one or t w o r e f i n i n g weights had values different f r o m 1.0. Deviations i n the refining weights f r o m 1.0 were only necessary when the o p i n i o n projections for one position were consistently too high or too low. Leaving refining weights and returning t o the modified persuasibility constants, it should be noted that these constants should have changed w i t h the issue, w h i l e the infons' validities, i n i t i a l audience sizes, and persistence constant should have stayed the same for a l l A P dispatches across a l l topics. Since the m o d i f i e d persuasibility constants v a r i e d f r o m 0.09 to 7.0 per A P paragraph per day, the superficial interpretation is that the public can be m u c h more easily persuaded for some topics than others. T h e value o f 0.09 was f o r the economic climate study where the number o f articles was over 12,000, w i t h most mentioning the issue o n l y i n passing. The greatest ease for changing public o p i n i o n was f o r i n f o n s f a v o r i n g i n f l a t i o n b e i n g more i m p o r t a n t f o r the issue o f unemployment versus inflation (7.0 per A P paragraph per day). I t is perhaps premature to draw f i r m conclusions f r o m the modified persuasibility constant I t w o u l d be dangerous t o say that the low value for the economic climate actually meant that people were very fixed i n their o p i n i o n for this question, so that many more paragraphs were needed to cause a change o f m i n d . I t is possible that people d i d not notice the relevant paragraphs in articles o n the economic climate because their minds were o n other portions o f the article. A l s o , the persuasive strength o f a paragraph on the economic climate might have been weaker than one for unemployment versus inflation. There was also the stochastic p r o b l e m o f having sampled different amounts o f text for the six examples. Comparison o f Figures 5.3, 5.5, and S.6 shows that the fluctuations i n the projections increased as fewer dispatches were i n c l u d e d i n the analysis. S i m i l a r l y , the predictions became worse as the m o d i f i e d persuasibility constant increased due, not t o the inherent s u i t a b i l i t y o f higher values for the constant, but to a l i m i t e d sample o f messages g i v i n g such large fluctuations that predictions deteriorated. Therefore, all the modified persuasibility constants might be underestimates, w i t h the problem being most severe for the economic climate where o n l y 3.7 percent o f the identified dispatches were retrieved, and least severe for unemployment versus inflation where 99 percent o f all stories were collected. Y e t another complication was the approximation (Appendix A, Equations A . 2 1 and A.22) that all relevant dispatches were identified d u r i n g the Nexis search. I f a significant number were left out, the interpretations o f the m o d i f i e d persuasibility constants would also need to be altered. T h e i n c o r p o r a t i o n o f a n u m b e r o f d i f f e r e n t constants i n the m o d i f i e d persuasibility constant and i n the r e f i n i n g weights is both an advantage and a disadvantage. The advantage is that o n l y one small set o f constants is needed f o r accurate opinion predictions. The disadvantage is that the constants incorporate not o n l y audience persuasibility but also audience size and v a l i d i t y o f the A P as w e l l as possible i n f o n misscoring. As a result, i t is d i f f i c u l t to dissect the contributions o f the individual factors. The relative constancy o f the modified persuasibility constants for any one issue over time also meant that the parameters changed o n l y s l o w l y i f at all w i t h time.

Copyrighte

\fethodological

Significance

of Work

127

although it was possible that the constants v a l i d i n early times were less appropriate at later times w h e n studies were c o n t i n u e d for as l o n g as nine years (defense spending, Figure S.10). W h e n the r e f i n i n g weights were the same f o r a l l positions, the suggestion was that i t was as easy to sway opponents as proponents or an idea, given infons w o r t h the same number o f paragraphs.

6.6

INSENSITIV1TY VALUES

OF

PREDICTIONS TO

THE

STARTING

OPINION

One reassuring observation was the convergence o f o p i n i o n calculations t o values dictated by the message structure, regardless o f the i n i t i a l opinions used for starting the computations. Therefore, errors i n the first o p i n i o n time point used for the c o m p u t a t i o n w e r e n o t i m p o r t a n t . F o r example, the projection assuming a population w i t h everybody favoring more defense spending (Figure 5.7) showed that even this very inaccurate i n i t i a l p o l l point still led to the same projections after a delay. Consequently, i f i n i t i a l p o l l values o n l y have small errors, the projections w o u l d home very rapidly to the proper values. There is another illustration o f this point i n Figure 5.20 for troops i n Lebanon. In this case, projections were made both w i t h and w i t h o u t the truck b o m b i n g infon in favor o f more troops. A g a i n , the wide disparities seen just after October 23, 1983, disappeared w i t h i n t w o to three months.

6.7

I N T E R P R E T A T I O N S FOR A L L I D E O D Y N A M I C

PARAMETERS

A major strength o f ideodynamics is the fact that it was derived deductively f r o m k n o w n phenomena. As a result, a l l the constant parameters i n the m o d e l were interpretable i n terms o f social interactions. There was n o appeal t o arbitrary parameters o n l y added to fit the data. The ready explanation o f a l l constants demonstrates that ideodynamics does not impose an a r b i t r a r y mathematical structure inappropriate for the analysis o f o p i n i o n formation. I n summary, these parameters for mass media messages are: 1. The persistence constant describing the exponential loss in availability o f a message to the population and having a universal half-life o f one day. 2 . The m o d i f i e d persuasibility constant, i n c l u d i n g the v a l i d i t y o f the m e d i u m , the size o f the typical mass media i n f o n (when o n l y one m e d i u m l i k e the A P is used), and the malleability o f the population for the issue dependent o n the closeness o f the issue to the core beliefs o f the public. 3. The refining weights, one for each conversion i n the population conversion model. The weights describe the extent t o w h i c h the infons favoring one position are more or less powerful than infons favoring other positions and the extent to w h i c h messages may have been systematically misscored so that unusually high or l o w content scores may have been g i v e n to infons favoring one or more positions. 6.8 S I G N I F I C A N C E O F N O O P I N I O N

CHANGE

T h e fact that o p i n i o n w i l l reach the values dictated by the information structure is o f interest i n cases where little o p i n i o n change is predicted and little is observed. Relatively speaking, that was the case for Contra aid and for the Democratic primary, where the changes were not nearly as large as for the other examples. U s i n g the bandwagon analysis for the Democratic primary, the observed values were the predicted ones. W i t h the name-count scoring o n l y , the projections were for a steady decline i n o p i n i o n supporting Glenn and M o n d a l e w i t h a corresponding rise i n opinion favoring the others to a final level i n the range o f 50 percent. Indeed, thai was the total percent name count for this group (Table 5.2). The name-count

Copyrighte

128

Predictions

of Public

Opinion from

the Mass

Media

analysis illustrates h o w an inappropriate i n f o r m a t i o n structure w i l l lead t o a prediction worse than that o f n o o p i n i o n change because the p o l l values w i l l be projected to change to those dictated by the erroneous information.

6.9

ANALYSIS OPINION

OF PERSUASIVE

MESSAGES

ACTING

ON

PUBLIC

The ability o f ideodynamics to relate percentage values for public o p i n i o n w i t h i n f o r m a t i o n available t o the p u b l i c , i n a time-dependent fashion, has obvious practical applications. It is useful t o recall that the calculations were d i v i d e d i n t o t w o parts. I n the first, messages were scored by computer for their support o f i n d i v i d u a l positions o r ideas. In the second, the scores were entered into the ideodynamic equations for calculation o f o p i n i o n after an i n i t i a l p o l l date. I t is necessary t o stress the two-step nature o f the process because i t is possible to enter message scores obtained from alternative methods into the o p i n i o n projections. I t w o u l d be quite appropriate for human j u d g e s , f o r example, t o assess i n f o r m a t i o n instead o f using a c o m p u t e r program. Such a procedure might be appropriate for calculating public responses to advertising campaigns i n w h i c h manufacturers w i s h t o improve their market share. This w o u l d be especially true i f the advertisements were more complex than text, including visual and aural cues. It has already been mentioned that the computer text analysis has the advantages o f being robust, easy t o use, and capable o f g i v i n g accurate o p i n i o n projections. These same comments apply to opinion calculations using i n f o n scores. A m o n g the important advantages o f the ideodynamic o p i n i o n calculations is the paucity o f parameters. I n the simplest cases, i f the universal persistence half-life is used and i f the r e f i n i n g weights are a l l 1.0, it is o n l y necessary t o o p t i m i z e the modified persuasibility constant. Since this simplest case was sufficient f o r t h r e e o f the six examples i n this book, the m o d i f i e d persuasibility constant is probably the o n l y parameter w h i c h w i l l need to be optimized for a significant fraction o f all cases. The scarcity o f parameters means that they can be set using very few measured poll values.

Copyrightec

Significance of Work to Theories of Opinion Formation

In discussing ihe ramifications of ihe studies in this book, me argument was made that the approximations for the computations are plausible because empirical tests with six diverse sets of data yielded reasonable opinion time courses in all cases (Chapter 5). If additional examples confirm the methods, then the theory will be on sounder footing still. In a recent and thorough review of the persuasion literature with an emphasis On time trends, Neuman (19S7) did not find any other répons in which accurate time courses of opinion could be computed from mass media messages. Furthermore, as noted in Chapter 2, ideodynamics does have unique features. For example, this model can be used to compute an opinion time series using arbitrarily small time intervals such as the six or twenty-four hours in this book. This use of very short lime intervals has not been reported in the past and demonstrates that mass media messages have a duration of less than a week. The small intervals of calculation also provide enough precision to study very rapid increases and decreases in opinion, as was found for troops in Lebanon. This and other unusual features, such as the ability to explore opinion reinforcement mathematically, means that no other models have been reported with the same powers as ideodynamics. In consequence, further empirical tests can point to limitations of ideodynamics but will not necessarily support competing models for the responses of opinion to information since none have been reported with the same capabilities. In addition to providing empirical evidence for opinion being calculable from persuasive messages, the studies in this book also give insights into the processes of persuasion as discussed below. 7,1 MASS MEDIA M E S S A G E S AND OPINION L E A D E R S H I P Ideodynamics can be adapted to model all news reports first persuading opinion leaders, who then rebroadcast the messages and influence the public at large (Chapter 2). Such a two-step process would have had three important effects: (1) The mass media messages would have amplified audience sizes and credibilities, (2) the effect of mass media messages on the public as a whole would have been delayed as the opinion leaders absorbed and retransmitted news, and (3) continued rebroadcast by the opinion leaders would also give messages an apparently longer duration than if they acted on the population directly. Data in mis book on the last two points suggest that opinion leadership of this type is unlikely. The first line of evidence is that the Lag between media messages

Marepnan, 3auinmeHHbin aerape*HH npaacM

130

Predictions

of Public Opinion from the M ass

Media

and o p i n i o n response was at most a few hours, since o p i n i o n i n favor o f more troops i n Lebanon increased f r o m under 7 percent t o over 30 percent w i t h i n t w o days after news o f the truck b o m b i n g o f A m e r i c a n M a r i n e headquarters i n Lebanon in 1983. T h a t w o u l d g i v e o p i n i o n leaders too l i t t l e time t o process and retransmit messages f a v o r i n g more troops in any reasonable interpretation o f the two-step process o f persuasion. The second set o f data arguing against this sort o f opinion leadership is related to the duration o f mass media messages. As discussed i n Chapter 5, least squares optimizations for five o f the six examples gave a one-day persistence half-life for A P messages. The exception was for Contra aid, where i t was inherently d i f f i c u l t to perform a v a l i d o p t i m i z a t i o n since there was little change i n either o p i n i o n or the message structure. Therefore, a one day half-life is a good consensus value. F r o m a practical standpoint, this is advantageous because a one day h a l f - l i f e means that infons lose their effect w i t h i n about a week. Therefore, o p i n i o n calculations o n l y need i n c l u d e i n f o r m a t i o n w i t h i n a week o r t w o before the b e g i n n i n g o f the c o m p u t a t i o n . T h i s persistence h a l f - l i f e is substantially shorter than the s m a l l number o f months used by other investigators (e.g. H i b b s , 1979; K e r n e l ] , 1978; M u e l l e r , 1970a, 1970b; O s t r o m and S i m o n , 1985; and Zielske and H e n r y , 1980). Therefore, i t is important that this consensus half-life was assigned after least squares optimizations for six independent issues. T h i s short d u r a t i o n o f messages means that l o n g - t e r m m e m o r y is p r o b a b l y relatively unimportant, so far as persuasion o f the public as a whole is concerned, for the issues studied. The majority o f the public are casual observers for most p o l l e d issues and w i l l probably either undergo o p i n i o n changes rapidly or not be persuaded at all after becoming acquainted w i t h a news i t e m . I f their o p i n i o n stays the same, they w i l l soon forget messages as new ones arrive to take their places in our modern society where i n f o r m a t i o n is saturating. Perhaps, i n a society w i t h fewer new messages, o l d information m i g h t have a longer effect since i t w o u l d be displaced i n a s l o w e r fashion. O b v i o u s l y , there w i l l b e a f e w e x p e r t s i n the population w h o w i l l remember o l d messages and use them for decision m a k i n g . However, they w i l l only comprise a tiny fraction o f the people polled. I n brief, both the rapidity o f the public response and the short message duration argue against a two-step process o f persuasion w o r k i n g through opinion leaders. Since the concept or o p i n i o n leadership was formulated f r o m survey data where the respondents were asked to reconstruct their information sources (beginning w i t h Lazarsfeld, Berelson, and Gaudet, 1944), i t w o u l d be interesting to perform a survey asking people for the reasons they changed opinions for the issues i n this book. I t is entirely possible that the result w o u l d be somewhat different than that suggested by the measurements here, studying actual media messages and broadcast times. In fact, m e m o r y is often significantly distorted ( M a r k u s , 1986), w i t h people misrecalling p r i o r attitudes. The unreliability o f m e m o r y is further demonstrated i n Menzel's (1957) study s h o w i n g that physicians are l i k e l y t o remember that they began prescribing an antibiotic at a time significantly earlier than the t i m e they actually d i d so. W h e n challenged to give the reasons for an o p i n i o n change, the typical person may w e l l feel obliged to provide a more coherent reason and sequence o f events than actually occurred. The unimportance o f opinion leadership has important consequences for persons interested i n generating persuasive messages. The i m p l i c a t i o n for efforts to change public o p i n i o n d i r e c t l y is that the messages s h o u l d actually be d i r e c t e d at the populace and not at some m y t h i c a l groups o f elites, w h o w i l l , i n turn, convince the public as whole. However, f r o m a propaganda or public relations standpoint, it may indeed be useful to try first t o convince elites, w h o can persuade the mass media to broadcast additional messages favoring the propagandist. These elites m i g h t indeed have inordinate influence on the public, but that influence is most l i k e l y to be due to their access to mass media rather than to their ability to persuade followers d i r e c t l y by retransmitting media broadcasts i n a two-step process. I t should be realized, however, that this direct importance o f the mass media is l i k e l y to be restricted to national issues where the media discuss the issues

Copyrighte

Significance

of Work

¡31

extensively and where local leaders are u n l i k e l y to have access to p r i v i l e g e d information. For local issues, not subject to wide media debate, it is quite possible that the local elites can have more o f a monopoly on privileged information and can have significant influences on their followers without use o f the mass media.

7.2 R E I N F O R C I N G R O L E O F P E R S U A S I V E M E S S A G E S In Chapter 1 and Appendix A , there is an extensive discussion o f how the reinforcing role o f persuasive messages can be entered into the construction o f the persuasive force curves. A refinement was further proposed for the dulling effects due to frequent repetition o f persuasive messages. Fortunately, it was unnecessary to invoke either phenomenon for the studies i n this book. F i n d i n g that o p i n i o n reinforcement can be ignored is not i n contradiction w i t h previous data f r o m surveys in w h i c h people are asked about their impressions o f persuasive messages (e.g. Lazarsfeld, Berelson, and Gaudet, 1944; Trenaman and M c Q u a i l , 1 9 6 1 ; Noelle-Neumann, 1973, 1977, 1984). People can say i n a survey that their opinions were reinforced w i t h o u t such reinforcement actually taking place. I d e o d y n a m i c s argues that i t makes n o difference i f people c l a i m o p i n i o n reinforcement so long as their minds are actually as easy t o change as that o f other persons n o t subject t o r e i n f o r c i n g i n f o r m a t i o n . I f there is no difference i n persuasion, then the ideodynamic interpretation is that the subjects are f o o l i n g themselves into thinking that their views are being strengthened. H a v i n g just noted that time-dependent reinforcement was not observed, i t is still formally possible that there was a constant level o f time-independent reinforcement due t o favorable infons. I n the calculations i n this book, constant reinforcement cannot be distinguished f r o m generally lowered scores for infons. However, when the actual infons favoring different positions were examined (Chapter 5), it was clear that infons favoring i n d i v i d u a l positions were n o t always present at the same levels. Furthermore, the differences i n i n f o r m a t i o n were s u f f i c i e n t l y large that p u b l i c opinion d i d shift. Since it seems rather forced to argue that reinforcement is constant w h i l e i n f o r m a t i o n favoring change is not, the most plausible c o n c l u s i o n is that reinforcement is usually relatively u n i m p o r t a n t - w h e n measured by its effects o n opinion change. One theoretically possible explanation for the small amount o f reinforcement observed is that information densities were so l o w that most people only received one u n m i x e d message. Under these c o n d i t i o n s , i t w o u l d make a difference i f most messages were m i x e d , containing infons favoring more than one position, or i f most messages were pure, having o n l y one i n f o n w i t h a non-zero content score. W i t h m a i n l y pure messages, i t is conceivable that reinforcement c o u l d be very large but invisible in the ideodynamic computations since people receiving reinforcing infons w o u l d not have received any infons persuading t h e m t o change t h e i r m i n d s . S i m i l a r l y , people exposed to infons capable o f converting their opinions w o u l d not have been exposed to reinforcing infons. Therefore, reinforcement can be very strong but irrelevant because people whose minds are being changed are not subject to the reinforcing infons. T h i s explanation is u n l i k e l y , however, since a significant number o f the actual messages were m i x e d . T h e sample text analyzed i n A p p e n d i x C, for example, supported three positions. A n y person exposed to infons f a v o r i n g one o f these positions is l i k e l y t o come i n contact w i t h infons f a v o r i n g the other positions as well. A distinction must also be made between time-dependent and time-independent r e i n f o r c e m e n t I f reinforcement is constant w i t h t i m e , then i t w i l l be i n v i s i b l e i n ideodynamics. The effects w i l l be absorbed i n t o the persuasibility constant. The c r i t i c a l p o i n t for the calculations i n this book is that reinforcement should not be h i g h e r at some times than others due t o changes i n the levels o f r e i n f o r c i n g information.

Copyrighte

132

Predictions

of Public

Opinion from the Mass

Media

Besides ignoring reinforcement, the satisfactory calculations i n this book also d i d n o t need adjustments f o r i n f o r m a t i o n o v e r l o a d . T h e e n d result was the mathematically agreeable conclusion that persuasive force curves could be constructed simply by adding the forces for individual infons.

7.3

C U M U L A T I V E EFFECTS OF I N F O R M A T I O N R A T H E R T H A N EFFECTS OF T H E M E D I A

MINIMAL

H a v i n g j u s t discussed reinforcement, i t is appropriate to comment o n the l o n g standing, but not undisputed, concept o f the " m i n i m a l effects" o f the media (Cook et al., 1983; E r b r i n g , Goldenberg, and M i l l e r , 1980; Funkhouser, 1973a, 1973b; Iyengar and Kinder, 1986; Iyengar, Kinder et al., 1984; Iyengar, Peters et al., 1982; M a c K u e n , 1 9 8 1 , 1984; Page and Shapiro, 1982, 1983a, 1983b, 1984; Page, Shapiro, and Dempsey, 1985, 1987; Patterson a n d M c C l u r e , 1976; Robinson, 1976; Rogers, 1983; Shaw and M c C o m b s , 1977; Wagner, 1983). One interpretation is that the media have m i n i m a l consequences on o p i n i o n change because the m a x i m a l consequence is reinforcement As noted i n the previous section, reinforcement was not detected for the examples i n this book. However, reinforcement may occur i n other circumstances. Nevertheless, under any circumstance, the essential question f r o m the standpoint o f social change is whether reinforcement is so strong that n o change occurs at a l l . For issues where opinions do change, reinforcement cannot be so o v e r w h e l m i n g as to block all movement. For these issues, the crucial element i n d e t e r m i n i n g public o p i n i o n is the residual amount o f persuasive force, h o w e v e r s m a l l , w h i c h can override the reinforcing information, since those are the effects w h i c h w i l l cause the social alterations. Another interpretation o f the law o f m i n i m a l effects is that mass media messages are small relative to other messages T o draw this conclusion i t is necessary that all relevant messages favoring all positions be included i n the model, and that is one o f the p r i n c i p a l features o f idevxlynamics. For instance, ideodynamics can show that certain messages m i g h t have apparently m i n i m a l effects because the opposing messages are o v e r w h e l m i n g rather than the favorable messages being ineffectual. W i t h fewer opposing messages, the favorable c o m m u n i c a t i o n s m i g h t be q u i t e persuasive. Due t o this importance o f preexisting and other contemporaneous messages f a v o r i n g all positions, i t is d i f f i c u l t to assess the importance o f natural messages such as those f r o m a presidential debate on voter preference (e.g. Katz and Feldman, 1962; Mueller, 1970b; Sears and Chaffee, 1979), unless all other important messages relevant to the campaign are also entered i n t o the calculations. T h e strength o f a message w i l l appear m i n i m a l " i f there are many other relevant messages supporting the same position. O n the other hand, i f there are few other messages o f this type, the identical message can seem to have a " m a x i m a l " e f f e c t B o t h possibilities were graphically demonstrated for the case o f troops i n Lebanon. I n the absence o f the truck b o m b i n g i n f o n supporting more troops, the o p i n i o n projection was poor w i t h a large M S D and entirely missed the increase i n people favoring more troops. W h e n eighty A P paragraphs favoring this p o s i t i o n equivalent to about 10 percent o f all A P paragraphs analyzed-were added on October 23, 1983, there was a dramatic improvement i n the M S D , w h i c h decreased eightfold. The effect was obvious upon inspection o f the projected time trends and was expected because there was very little other i n f o r m a t i o n favoring more troops at any t i m e . W i t h a l o w b a c k g r o u n d , there was a dramatic and " m a x i m a l " effect w h e n the persuasive force function favoring more troops included these eighty paragraphs. The situation was quite different for the addition o f a truck b o m b i n g i n f o n favoring troop w i t h d r a w a l w i t h a strength anywhere i n the entire range f r o m zero to about forty A P paragraphs. Such an addition had essentially no effect on the f i t as seen by least squares optimization. This was again expected since there was a large amount o f additional information in this direction even i n the absence o f this truck

Cc

Significance

of Work

133

bombing infon. Therefore, even a very large amount o f additional information w o u l d have a " m i n i m a l " effect on opinion calculations. Another example o f the m i n i m a l effects o f the media was the small effect o f stories o n defense waste and corruption on o p i n i o n favoring less defense spending. This d i d not indicate that such stories were inherently negligible, but only that their effects were small in the total sea o f infons in that direction. W i t h fewer other infons directly favoring less spending, the waste and fraud stories w o u l d have been m u c h more important. Therefore, i n d i v i d u a l mass media messages or groups o f infons can appear to have m i n i m a l o r m a x i m a l effects depending o n the prevalence o f other information, both pro and c o n . The example w i t h Lebanon has shown how m i n i m a l and maximal effects can be seen even w h e n a l l the relevant i n f o r m a t i o n is restricted to the mass media. A l t h o u g h the mass media are l i k e l y to be the major source o f persuasive messages for some issues, ideodynamics can also include i m p o r t a n t amounts o f information f r o m non-mass media sources i f those messages can be coded as infons. In fact, these other messages might sometimes even swamp out the effects of mass media infons. A n illustration m i g h t be the adoption o f antibiotic usage, w h i c h was correlated w i t h physician interactions and not media messages (Coleman, Katz, and M e n z e l , 1966). Mass media messages were not l i k e l y to have contained all the technical information a doctor w o u l d have desired before making such a professional judgement. For this i n f o r m a t i o n a doctor was more l i k e l y to seek advice f r o m knowledgeable colleagues. I n this case, the mass media m i g h t have had a m i n i m a l effect due to the primacy o f messages transmitted i n alternative media. I n contrast, for an issue l i k e defense spending, there were very few experts or o p i n i o n leaders to w h o m the public at large c o u l d t u m for additional i n f o r m a t i o n . Therefore, the mass media was l i k e l y to have contained the p r i n c i p a l persuasive messages, as was found i n this study (Comstock et al., 1978). The mass media w i l l also have m i n i m a l effects when the media have such a low reputation that the more prevalent method o f information f l o w is rumor, as discussed in the introduction and i n Chapter 6. It is the quantitative nature o f ideodynamics w h i c h permits this model to assess different contributions o f selected messages under different circumstances, pointing to the usefulness o f a quantitative approach. I n addition, throughout the discussion i n the present section, comments about message subsets have consistently been related to other relevant messages. A n y one message can be m i n i m a l l y o r m a x i m a l l y important depending on the prevalence and importance of other messages and the extent to w h i c h the issue is close to the svstem of beliefs o f the i n d i v i d u a l . Since ideodynamics includes the malleability o f the population, the v o l a t i l i t y o f o p i n i o n , and the effects o f a l l messages i n a quantitative way, this model is able to highlight examples showing that i n d i v i d u a l messages have different effects under different circumstances. Furthermore, ideodynamics provides a convenient framework f o r accounting for many different messages simultaneously. The summary conclusion o f this consideration o f the effects o f mass media stories is that there is actually no conflict between the ideodynamic results and earlier reports o n the m i n i m a l effects o f the media. As just noted, this book has considered examples where individual messages can make contributions o f different magnitudes. However, the only case where a single infon was able to have an enormous effect was the i n f o n favoring more troops being sent to Lebanon. I n t w o other cases, the effects were s m a l l . A very strong infon favoring less troops i n Lebanon had a negligible effect. A l s o , the entire list o f stories on waste and fraud contributed m i n i m a l l y to opinion o n defense spending. Nevertheless, when all A P messages were considered for these t w o topics and the four others studied, the accumulated power o f mass media messages was f o u n d to determine o p i n i o n so strongly that accurate o p i n i o n time trends c o u l d be calculated f r o m mass media infons alone. Therefore, i t is l i k e l y that i n d i v i d u a l mass media stories w i l l indeed have m i n i m a l effects most—but not a l l — o f the t i m e . I t is w e l l k n o w n that isolated droplets f r o m a river have little effect i n changing geography w h i l e massive gorges

Copyrights

134

Predictions

of Public

Opinion

from

the Mass

Media

can be cut w i t h the cumulative effects o f the persistent and steady coursing o f a l l the water. I n the same way, as stated at the beginning o f this book, it is more useful to think o f a series o f persuasive messages having a powerful cumulative effect rather than individual messages having m i n i m a l effects. Therefore, the law o f the m i n i m a l effects might be more f r u i t f u l l y replaced by the concept o f the cumulative effects o f information.

7.4 C A V E A T S FOR L A B O R A T O R Y E X P E R I M E N T S The necessity o f taking i n t o account a l l persuasive messages acting o n the population also points to an important caveat i n evaluating the results o f studies where the population is measured just before and after an intervention w i t h persuasive messages. Such studies m i g h t i n v o l v e individuals being exposed to messages such as television news in a laboratory setting, w i t h measurements on attitude being made before and after the treatment (e.g. Iyengar et al., 1982; Iyengar and Kinder, 1986). W i t h o u t k n o w i n g the informational residues from relevant messages received before the laboratory intervention, i t is not easy to predict whether the added messages w i l l have a large o r small effect as j u s t discussed. Ideodynamics also notes that o p i n i o n can continue to change after message exposure w i t h i n the concerned and motivated subpopulation influenced by remembered information. Although this effect was not large for the p u b l i c at large not concentrating o n the issue, remembered information may play a more crucial role in a laboratory setting where the subjects might devote concentrated thought to an issue. Therefore, different o p i n i o n changes m i g h t be found at different times after the time o f information exposure. A single time point might give a misleading result.

7.5 L A W O F T H E 2 4 - H O U R D A Y A t this point, i t m i g h t be useful to mention again the law o f the 24-hour day discussed earlier i n the introduction and in Chapter 1. This law simply acknowledges that time moves inexorably forward, leaving most people w i t h insufficient time to consider most issues in depth. As a result, the public at large is l i k e l y to make most decisions based only o n superficial i n f o r m a t i o n . T h i s unavoidable superficiality is probably central to the success o f ideodynamics and stresses the fact that an analysis o f the passage o f real time is crucial to any understanding o f mass behavior. The law of the 24-hour day means that caution must be exercised in interpreting data f r o m one-shot surveys where people are asked to reconstruct the events and reasons i n v o l v e d in decisionmaking (Lane and Sears, 1964; Robinson and Clancey, 1984). The original decision may w e l l have been made w i t h o u t m u c h reasoning and have been based largely on infons merely advocating a position. However, a person asked to reconstruct events is l i k e l y to try to find a logical explanation, as discussed above. I t was argued i n the i n t r o d u c t i o n and i n Chapter 1 that there is a certain i n e v i t a b i l i t y i n the superficiality w i t h w h i c h the bulk o f the population w i l l absorb information about an issue. Indeed, the greater the diversity o f decisions made by an i n d i v i d u a l , the greater w i l l be the variety o f relevant i n f o r m a t i o n . A s a result, the less w i l l be the care w i t h w hich i n d i v i d u a l decisions can be made. This consequence is seen at the l e v e l o f the public as a w h o l e , l e a d i n g to the a b i l i t y t o use ideodynamics to predict opinion f r o m the structure o f persuasive messages. H o w e v e r , the same arguments o f t i m e l i m i t a t i o n apply to the g o v e r n i n g , business, and other elites w i t h i n the population. B y definition, these people are the ones w i t h the most power. The more powerful the person, the broader the range o f responsibility and the more superficial must be the decisionmaking due to t i m e constraints. These comments stress the importance o f having a competent staff, w h i c h can subdivide the responsibility so that different individuals have small enough domains that they have the time to assess the primary information w i t h some care.

Copyrights

Significance

7.6 I N T E R P R E T A T I O N S OF I D E O D Y N A M I C

of Work

135

PARAMETERS

A s discussed i n Chapter 6, the three types o f ideodynamic parameters for mass media messages are the persistence constant, the modified persuasibility constant, and the refining weights. The relationship between the persistence constant and opinion leadership was considered at the b e g i n n i n g o f this chapter. The modified persuasibility constant was discussed at length i n Chapter 6. The r e f i n i n g weights also have interesting interpretations. These weights reflect, i n part, corrections needed to account for systematic overo r underscoring o f certain classes o f infons (Chapter 6). One example m i g h t be the case o f unemployment versus inflation. Here, i t was necessary to reduce the content scores for infons favoring the equal importance o f unemployment and i n f l a t i o n to h a l f o f the value for infons supporting the importance o f u n e m p l o y m e n t A t the same t i m e , infons stressing the importance o f i n f l a t i o n needed t o be augmented by 140 percent I t was not easy to design the text analysis for this issue so that infons scored for the equal importance o f unemployment and i n f l a t i o n exactly favored that position. I t was quite conceivable that a significant portion o f these i n f o n scores actually had a consistently higher component supporting the importance o f inflation. I f a p o r t i o n o f the i n f o n scores favoring equal importance were transferred to the position supporting the importance o f i n f l a t i o n , a l l refining weights c o u l d actually have been the same, consistent w i t h equal malleability o f the pubiic for all positions o f the issue. Therefore, for u n e m p l o y m e n t versus i n f l a t i o n , the r e f i n i n g weights might have mainly reflected scoring errors. I f unemployment versus inflation can also be characterized by the same refining weight for all positions once the messages have been scored properly, the use o f a c o m m o n refining weight may be the r u l e rather than the exception, since i t w o u l d have applied to four o f the six examples tested. Besides u n e m p l o y m e n t versus i n f l a t i o n , this c o m m o n a l i t y was also seen f o r defense spending, the Democratic primary, and the economic climate (Table 5.3). The use o f a c o m m o n r e f i n i n g w e i g h t w o u l d correspond to strength o f feeling being correlated more closely w i t h the issue than its positions. A s the issue comes closer and closer to the population's system o f beliefs, the persuasibility constant w i l l decrease and the population w i l l be more and more refractory t o messages w i t h opposing ideas. However, once people have adopted a new position for this issue, any position, they w i l l again be as d i f f i c u l t t o dislodge. For core beliefs, i t may be very d i f f i c u l t to change any o p i n i o n . Again, though, once changed, i t w o u l d be just as d i f f i c u l t to cause a reversion to the o r i g i n a l position—if the refining weights are the same for the t w o conversions. T h e a b i l i t y to dispense w i t h r e f i n i n g weights f o r three o r f o u r o f the six positions has the aesthetically pleasing result o f reducing the number o f parameters needed i n the calculations. Therefore, whenever a choice is available, i t seems desirable t o adjust the analysis so that refining weights are not used. T h i s was an important reason w h y the name-count analysis for the Democratic p r i m a r y was considered inferior t o the bandwagon analysis (Chapter 5). T h e p o l l points c o u l d have been fit m u c h better for the name-count analysis i f r e f i n i n g weights different f r o m 1.0 were permitted. T h a t w o u l d have meant g i v i n g M o n d a l e a m u c h higher refining w e i g h t than that for the Others w h i l e the refining weight f o r G l e n n w o u l d have stayed the same. A l t h o u g h interpretations m i g h t be given for w h y those weights are reasonable for the name-count analysis, no variations at all were required Tor the bandwagon analysis. The dispensibility o f refining weights for the bandwagon analysis means that this analysis w i l l probably be m o r e useful f o r future predictions o f the popularity o f p o l i t i c a l candidates since n o weights need to be o p t i m i z e d . I f the bandwagon analysis continues to be better than the name-count analysis, the i m p l i c a t i o n w o u l d be that people do have the sophistication t o look beyond mere names for their candidate preferences-when the names become established. However,

Copyrights

136

the additional implication is that people may not look beyond bandwagon words to the candidates' positions or activities. In general, if two competing analyses require different numbers of parameters, the most powerful model will be the one with the fewest, so this criterion of avoiding additional parameters will be used to choose among different ideodynamic analyses whenever possible. In contrast to the case for unemployment versus inflation, it is unlikely that the only problem for the Contra aid issue was infon misscoring, because both Fan and Swim et al. found that infons favoring fewer troops were more than twice as effective as infons favoring more troops. Two rather different computer scoring schemes both led to this conclusion. Similarly, infons favoring less troops in Lebanon were also found to be 1.6 times as powerful as messages favoring more troops or troop retention at the same levels. Therefore, i f the scoring for troops in Lebanon is also reliable, there is the hint of the generalization that the American public has certain isolationist tendencies leading to a reticence in foreign adventurism regardless of whether the locale is Nicaragua or Lebanon. Given only two issues, this thought must be considered just a suspicion. However, consistent differences in the refining weights for more foreign policy issues might solidify the hypothesis. This analysis illustrates how important differences in refining weights might give significant insights into differences in the response of the population to different types of information. 7.7 NATURE OF EFFECTIVE PERSUASIVE MESSAGES IN THE MASS MEDIA The successful applications of ideodynamics give insights into the important aspects of persuasive messages. For instance, the superficiality forced by multiple demands on limited time is probably also at the heart of the success of the big lie in propaganda, where totally implausible assertions are made baldly and with no apologies. These lies will be believed since most of the population will be preoccupied with other concerns and will not take the time to pause and reflect carefully on the situation. However, in order for the big lie to work, the medium transmitting the propaganda must have a good reputation because the populace will assign a validity to the big lie in proportion to the reputation of the medium. The importance of the credibility of the medium was recognized even in early fairy tales, such as the one about the little boy who cried wolf. From time to time, however, the public can check the reliability of any medium against other information in the same way that the villagers were able to assess the validity of the wolf alarms. I f the medium shows signs of inaccuracy or unbelievability based on the public's comparisons with other information, then its reputation will fall and few reports by the medium, regardless of topic, will be believed. Although the media cannot afford to tell lies and still retain credibility, the press does have very broad latitude in the choice of messages disseminated. Members of the public usually will not fault the media for omitting items because they realize that the press cannot transmit all the information received, so some items must be omitted. The press might even be able to retain its trust in cases where the public might like to know about a suppressed item, so long as reported stories are accurate and the press or its censors can efficiently propagate the idea that the omitted item is one which is too delicate for transmittal to the public at large. One possible rationale is national security. However, the censors must be careful to suppress only those items which the public will forgive the press for suppressing. This may mean that the censorship should be accompanied by simultaneous messages training the public to tolerate the removal of certain items. Such tolerance is seen by public acceptance of censorship during wartime. Furthermore, the public will sometimes even favor overt

137 manipulation of the available information. This is usually true when the public provides unabashed support for public health and anti-drug abuse campaigns. It is the rare person who will insist on the publication of balanced stories presenting both the disadvantages of drug use and the glories of transient highs. This analysis shows that a credible medium can transmit the big lie in the short term and the law of the 24-hour day can prevent close scrutiny of medium content. However, in the long term, such a strategy has the problem that all which is said, even i f true, will be suspect once credibility is destroyed. Public distrust will be important as soon as the population resorts to alternative media. Ideodynamics asserts that the effects will be minor i f a medium loses credibility but is still the only source of pertinent information. In this case, the medium will influence the public more slowly, but will still move the public in the directions specified by the media messages. I f the public resorts to other media such as rumor and an underground press, however, then the effects of the original medium can be drastically decreased. In brief, then, the mass media together comprise a critical instrument in a modern democracy. As discussed above, the media are probably crucial for shaping both the agenda and opinion within the agenda for a large number of issues. However, the analysis may be misleading unless it includes all information bearing on the issue. In the same way, a tugboat pulling an ocean liner to port can have maximal effects in calm seas but minimal effects in a hurricane, where the nontugboat forces on the liner are overpowering. It is only by considering the cumulative effects of all messages relevant to an issue that the true impact of the media can be established. One of the most sobering aspects of the importance of the mass media is the small number of people involved in determining press contents. Indeed, the concepts of press scoops and exclusivity explicitly acknowledge that a single reporter or small group of reporters for one news organization can have an unusually large effect on the news structure. The small number of people involved is related to the limited number of journalists covering any given issue. Most local press organizations, both electronic and written, do not have the resources to cover national and international events. For these events, most reports in the press are second-hand, derived in large part from wire services like the Associated Press. This press agency in turn will assign only a few reporters to any one story. These persons then have the capacity--and obligation due to limitations of time and space--to select what they feel is newsworthy. As a result, news from journalists, however unbiased, is still colored by the inevitable omissions. In addition to the reporters are the editors who also exercise the power of agenda and set, together with the journalist, the tone of mass media messages. Although editors at the Associated Press and local news organizations can both remove information, they will not be able to add much unless they have access to other news sources. Certainly, there will be a few alternatives such as reports from other newspapers, other wire services, and television. But, again, the number of alternative sources is not large, with each of these sources only having a few primary news gatherers. In fact, newspapers typically give prominent treatment to stories identified by the wire services as important (Chapter 3), thereby ceding even more power to the wire services. Since a limited number of journalists and wire service editors are responsible for much of the original news, the public receives its news from a rather restricted group of individuals for any one issue. The ideodynamic studies in this book also emphasize the need for continual responses to the opposition. The very malleability of public opinion as described above means that opinions are not permanent, so it is always necessary to keep an eye on the activities of the opponents. There are no permanent victories or defeats. I f the opponents generate more information, then the proponents must do likewise. As seen for defense spending (Chapter 5), the increase in public opinion in 1979 favoring more spending corresponded to increased information in this direction without important changes in the messages favoring less spending. Then, opinion

138

Predictions

of Public Opinion from the Mass

Media

was brought back down not by a decrease i n messages favoring more spending but by an increase i n messages f a v o r i n g less spending. The total " d e c i b e l " level o f the debate was therefore higher from 1982 to 1984 than f r o m 1977 to 1979, even though public opinion about defense spending was about the same in both time periods. B y m u l t i p l y i n g persuasive force curves by the sizes o f the relevant target populations, ideodynamics recognizes that m i x e d messages w i t h strong unfavorable components can be favorable. F o r example, largely unfavorable i n f o r m a t i o n can have a significant positive effect when most o f the population is in the opposition. Obviously, i f only a minuscule amount o f favorable information is included w i t h the unfavorable, then o n l y a tiny fraction o f the opposition w i l l be converted. However, i f everybody is in the opposition, then a small fraction o f a large number can still be sizable i n terms o f absolute numbers. T h i s is one key to the observation that name mention for a product or political candidate, even i f i t is unfavorable, can s t i l l lead to increased purchases or popularity. However, that strategy must be abandoned as soon as the product or candidate gathers a large number o f supporters because the negative publicity w i l l then cause a loss i n loyalty. S i m i l a r l y , terrorists can expect to gain favor for their position i f the terrorized population is o v e r w h e l m i n g l y opposed t o their v i e w p o i n t . T h e y w i l l not lose a significant number o f sympathizers since they had few to begin w i t h . B u t i f even a small mention o f their cause is included i n the news, that m i g h t be able to convert a s m a l l number o f opponents and hence give the terrorists a significant, but s m a l l number o f sympathizers. Therefore, at l o w sympathy, t e r r o r i s m can indeed be effective for recruiting converts t o the terrorists cause. I n fact, the tool o f terrorism can be made more effective by condemnations w h i c h o n l y serve to h i g h l i g h t that cause. This gain t o the terrorist w i l l be significantly greater i f the news is neutral, w ith a sizable component mentioning the terrorists' cause, rather than being violendy antiterrorist w i t h a very small pro-terrorist c o m p o n e n t T h i s analysis f o l l o w s f r o m the argument (Chapter 2) that e v e n l y m i x e d messages are n o t neutral unless the population is evenly split on an issue. As soon as terrorists have an appreciable number o f supporters, however, they should shift their tactics so that their actions do not drive their adherents to the other camp. They s h o u l d behave m o r e responsibly, w h i c h by then they can probably afford to do, because they are also likely to have easier access to the media. 1

7.8 C A U S E A N D

EFFECT

Ideodynamics claims that o p i n i o n can be c o m p u t e d f r o m an analysis o f the c o n t e n t reputation, and t i m i n g o f the messages a r r i v i n g at the p o p u l a t i o n . A n alternative is to postulate that messages reflect o p i n i o n . O f these t w o possibilities, the studies i n this book suggest that the more p o w e r f u l and general model involves opinion f o l l o w i n g messages because this phenomenon was observed for a l l six cases. The reverse concept o f messages tracking o p i n i o n clearly cannot alw ays be true. Some messages are c l e a r l y e v e n t - d r i v e n . F o r example, the truck b o m b i n g i n Lebanon was a newsworthy event w h i c h c o u l d not have been predicted f r o m the o p i n i o n structure. A l s o , there was n o w a y to predict f r o m o p i n i o n on defense spending that net information favoring more spending was going to increase i n 1979 and decrease i n 1981. The relationships between message generation and impact m i g h t be explained by postulating that o p i n i o n always reflects messages w h i l e messages reflect both o p i n i o n and unusual events. Therefore, i n the absence o f extraordinary news, i t m i g h t be possible to m o d e l both message generation and message i m p a c t by exploring the relationships between the t w o . T l i i s modeling can be performed i n the context o f ideodynamics, w h i c h has a structure for separating information generation from information impact A t the core o f this structure are infons w h i c h code essential features o f persuasive messages. This book has already examined the impact o f infons. T o extend the

Copyrights

Significance

of Work

139

model, i t is o n l y necessary to postulate mathematical models for h o w o p i n i o n affects i n f o n generation. Persuasive force functions can then be computed based on the expected infons Using these functions, i t w i l l be possible to compute both o p i n i o n f r o m messages and messages f r o m o p i n i o n to g i v e a t i m e trajectory o f o p i n i o n without any need to measure messages directly. The calculation c o u l d proceed by starting w i t h a set o f o p i n i o n values at a particular time. I f the opinion structure causes certain messages to be generated, then those messages should be computable since the starting o p i n i o n structure w i l l have been specified. The new messages can then be used to predict the opinion structure at the next t i m e i n t e r v a l . The resulting o p i n i o n can then be used t o calculate the messages i n that time interval. These messages and the calculated o p i n i o n for that t i m e interval can be used to compute o p i n i o n i n the f o l l o w i n g t i m e i n t e r v a l . This autoregressive process can be repeated ad i n f i n i t u m to calculate a final t i m e trend independent o f actual message measurements. A s discussed i n Chapter 2, this has already been done for the case where information favorable to an idea is broadcast i n proportion to the number o f believers and where there is no opposition to the idea. I n this case, ideodynamics predicts that the idea w i l l increase i n a logistic fashion until the entire population accepts the idea (Fan, 1985a), as has been f o u n d for the adoption o f many innovations ( H a m b l i n . Jacobsen, and M i l l e r , 1973). O n the other hand, i f proponents and opponents have the same powers o f message emission, then their ratios should not change, regardless o f what those ratios are, even though both groups may increase i n size (Fan, 1985a). In future studies, other models w i l l be explored, i n c l u d i n g those t o explain h o w fads can come and go w i t h great r a p i d i t y . Such models m i g h t require a boredom effect where the adopters o f a fad diminish their message broadcast as time proceeds. S i m i l a r l y , there might be a social pressure effect due to people observing others w i t h an o p i n i o n , fad, or habit. I n this case, the more the people w i t h the trait, the greater w o u l d be the social pressure. The result w o u l d be social pressure infons, w h i c h have already been incorporated i n t o message generation models for habits l i k e s m o k i n g (Fan, 1985b). F o r habits, there m i g h t also be a r e c i d i v i s m effect due to people who have j u s t changed a habit being momentarily euphoric at having made the change and then being d r i v e n back to the old habit by nostalgia. Part o f this nostalgia m i g h t be b i o l o g i c a l , reflecting a desire to return to the physiological state o f an addiction. This recidivism effect could be entirely due to personal experiences, i n w h i c h case the ideodynamic equations w o u l d include personal experience infons due both to euphoria and nostalgia (see Fan, 1985b for equations). It is also possible that people whose ideas are i n the ascendancy w i l l be more vociferous i n their dissemination o f favorable messages, as has been proposed by Noelle-Neumann (1984) i n The Spiral of Silence. Such a phenomenon m i g h t be modeled b y persons being more w i l l i n g to send favorable messages w h e n they perceive that their position is gaining progressively more converts. Unfortunately, N o e l l e - N e u m a n n (1984) does not propose h o w the spiral o f silence ends, so the model w i l l probably lead to the position i n ascendancy gradually becoming the o n l y acceptable position, a circumstance w h i c h w o u l d be inaccurate for social issues where controversy does not die completely. For the study w i t h the Democratic p r i m a r y i n this book, the most successful analysis used bandwagon words, assuming that people w i l l be swayed s i m p l y by news that a candidate's campaign was proceeding w e l l or p o o r l y . T h i s adoption o f favorable o p i n i o n when a bandwagon starts has been modeled mathematically by Brams and Riker (1972) and Straffin (1977). Their proposal that people preferentially j o i n ascendant groups c o u l d be adopted to ideodynamics by assigning higher i n f o n generation powers to subpopulations for w h i c h the bandwagon had started to r o l l . Obviously, these examples are but a few o f the w i d e variety o f models i n w h i c h message generation is dependent on prior o p i n i o n . Further w o r k is therefore planned to explore these and other models. A l t h o u g h calculations o f messages f r o m o p i n i o n have the advantage that they permit assessments o f social response trajectories w i t h o u t the need to measure real

140 messages, there is the corresponding weakness that there are no inclusions of the event-driven messages which do in fact occur in a significant number of instances. Besides the examples just mentioned for Lebanon and defense spending, it might be interesting to consider the example of the habit of smoking. Here, models not permitting the introduction of unexpected infons would not have been able to account for the surgeon general's report in the early 1960's on the health hazards of smoking. Therefore, in further studies it may be found that satisfactory opinion time trends can be projected by adding infons calculated from opinion to a minimum number of measured messages from unusual events. In brief, this book marks a halfway point in a mathematical examination of persuasion, showing that this process can indeed be analyzed in the two separate steps of message generation and message impact. The fact that models of message generation have not yet been fully explored does not reflect a weakness of ideodynamics since, as noted above for the logistic equation, ideodynamics does provide a convenient framework for including models for information generation. To return to the analogy with ballistic missiles, the purpose of this book was to examine the effect of messages once launched. To the extent that an understanding of war requires an understanding of the devastation due to weapons, an understanding of message impact also is useful for analyzing persuasion. A satisfactory method for analyzing the effects of messages and weapons provides a solid foundation for considering message and missile launching, other key steps in processes of war and persuasion. This book has demonstrated a consistent predictability of the public in the face of persuasive information. This finding has certain policy implications for democratic societies in which the main messages are from a trusted press. Since public opinion is heavily influenced by the press, opinion is not likely to be a direct check on the powers of individual groups of elites-¬ like the government, or portions of the government. Instead, the main mechanism may be a number of elites successfully transmitting different sides of a story, thereby checking the powers of other elites. For its part, public opinion may just reflect the messages sent, whatever they may be.

Appendix A

Mathematics of Ideodynamics

T h e essence o f the m a t h e m a t i c s o f i d e o d y n a m i c s has been presented p r e v i o u s l y ( F a n ,

1984, 1985a, 1985b). H o w e v e r , new insights were drawn f r o m the studies i n this book and an updated version o f the model is presented below. I n ideodynamics, social changes f o r a single issue are characterized by senders transmitting persuasive messages w h i c h have impacts o n receivers. Therefore, k e y elements i n the analysis are the issue w i t h i n w h i c h change is occurring, the messages transmitted and the structure o f the population.

A.l

STRUCTURE OF IDEAS

The first structure to consider is the issue and its associated positions, also called ideas. Each issue- essentially a q u e s t i o n - i s denoted b y Qq, where Q refers t o all possible issues and index a describes a particular issue. In this book, six issues were studied, so index a can range f r o m one t o six for the issues. F o r example, a « 1 m i g h t refer t o the issue o f whether there s h o u l d be more, same, o r less defense spending. W i t h i n any issue Q , the public can hold one o f several positions or ideas indexed by lettery. T h e individual ideas are denoted b y Q L w i t h a referring t o the and j t o t h e idea w i t h i n that issue. F o r defense spending, j = 1,2 J c o u l d correspond w i t h the positions o f more, same, and less spending. a

issue A.2

a

STRUCTURE OF THE POPULATION

I n the general case, P refers t o the p o p u l a t i o n w h i c h is l i k e l y t o receive information relevant t o issue Q - For defense spending, i t was assumed that the N o t Sures and Don't K n o w s were a subpopulation w h i c h d i d not receive such information and hence stayed oblivious o f the topic. I n this case, P referred to the remaining population actually holding opinions—approximately 90 p e r c e n t Ideodynamics, as described i n this b o o k , assumes that the p o p u l a t i o n is constant i n size a n d composition d u r i n g the entire time period o f any study. A t any time /, P is d i v i d e d i n t o t w o subpopuladons, PaA(0 made u p o f people aware o f the issue and P U(0 comprised o f persons unaware o f the issue but able t o receive pertinent i n f o r m a t i o n . The fraction o f awares PaA(0 w i t h i n P is defined as Aaft), so the unawares Palf(0* m a k i n g u p the balance, constitute l-A^t) o f the total p o p u l a t i o n Pq. P o p u l a t i o n constancy permits fractions o f the public t o be used interchangeably w i t h numbers o f people. a

a

a

a

a

a

Copyrights

142

Appendix

A

A m o n g the awares P /[ it is convenient to define the population favoring or believing i n p o s i t i o n Q j as PaAjThe p r o p o r t i o n o f believers P Aj a m o n g the awares P A is B \j(t), so the fraction of P Aj w i t h i n the total population P is a

t

a

a

a

aj

a

a

AaiO'BoAjlO-

W i t h this structure, the number o f subpopulations is one more than the number o f positions h e l d by the aw ares w i t h the unawares comprising the extra subgroup.

A.3

STRUCTURE OF MESSAGES

Each persuasive message is denoted A/* where index k is an arbitrary but unique number referring to a particular message. Each message A/^ has three i m p o r t a n t properties c a r r y i n g the same index k : a v a l i d i t y characteristic o f the m e d i u m , an audience size, and a number o f components called infons favoring different positions: 1. V a l i d i t y v£ - reputation score for the medium used for transmitting message A/fc. In the present studies, all AP messages were assigned the constant validity o f kvAP characteristic o f the A P m e d i u m so that (A.1)

vk =

k Ap. v

Fortunately, this simple assumption gave g o o d o p i n i o n calculations for a l l f i v e examples. 2. Audience size afrfi) - mathematical function of time t describing the audience size for message Af* as time proceeds. Since AP messages, among many others, are characterized by a high audience size when the message first arrives at the population, i t is convenient to define t w o subsidiary message properties: a. T i m e /£= the t i m e at w h i c h message A/£ first arrives at the p o p u l a t i o n . For A P messages, i t was assumed that '£ was the date o f the A P story. A l t h o u g h there is a delay o f several hours before an A P message a c t u a l l y appears i n the press, the date o f the dispatch is probably not m u c h different f r o m the time o f broadcast o f television and radio news carrying equivalent information. b. I n i t i a l audience size atftk) = the number o f people exposed to i n f o n message Afjt at t i m e w h e n the message first reaches the population. T h i s audience size is o b v i o u s l y larger f o r mass m e d i a messages t h a n i n d i v i d u a l conversations. For this book, the i n i t i a l audience size was assumed to be the same for all A P messages relevant to an issue. T h i s initial audience size was designated as constant k^p so, for A P messages, (A.2)

akftk) - kaAP •

I n contrast to A P infons, the audience sizes for some messages m i g h t actually increase after This might happen w i t h a book, for example, where sales w o u l d increase before an eventual decrease. For mass media messages such as those i n the A P , the audience size is l i k e l y to decrease r a p i d l y at the same rate for each message after the i n i t i a l date. Therefore, arft) c o u l d be described by an exponential decrease w i t h persistence time constant p so that

(A.3)

atft)-akttfrfP( ' k)f€*t2tk i t

= 0 for t <

I f mass media infons were all retransmitted by o p i n i o n leaders o r other persons w i t h approximately the same kinetics each time, then the analysis can be performed b y incorporating the effect o f the o p i n i o n leaders i n t o the audience size function for the mass media infons (Chapter 2). The effect w o u l d be to prolong the audience size function aidt). I f a significant amount of time was required for the o p i n i o n leaders to

Copyrights

Mathematics

of Jdeodynamics

143

begin the retransmission, then the audience size function c o u l d begin w i t h a short lag after initial receipt o f the message by the opinion leader and then be f o l l o w e d by an increase the leaders began convey the new i n f o r m a t i o n before final decline as the o p i n i o n leaders stoprebroadcasting the i n f o r m a t i o n . Therefore, there w o u l d be no need to include opinion leaders explicitly in the model i f their effects can be absorbed into the audience size function. 3. Infons I ijk components o f message M¿ f r o m different sources and directness, indexea by i , and favoring different positions Q j* Indices a and j refer to position Q j w i t h i n issue Qq and are the same i n Iaijk * Qaj* Index j corresponds to the first i n f o n dimension i n Chapter 1. Index i refers both to the directness o f the message and to the source o f an idea as deduced f r o m the message itself. O d d indices i = 1JJ5 ... are used when the source indexed by / d i r e c t l y advocates position Q ; w h i l e even indices i ~ 2.4,6,... are employed when support for Q j is only inferred due to information f r o m the source indexed b y L The specification o f odd and even indices corresponds t o the second i n f o n dimension (Chapter 1). Different index numbers are used when the population can and does distinguish among infon sources. These individual index numbers, aside f r o m their oddness and evenness, index the third infon dimension o f Chapter 1. Index i can code for both the second and third dimensions w i t h o u t any ambiguity since the second dimension only has t w o possibilities and these can be represented by oddness and evenness. The result is a less cumbersome terminology. The same o d d index number can be used for direct support o f Q j when the population cannot o r does not d i s t i n g u i s h among sources. S i m i l a r l y , the same even index n u m b e r is used for information indirectly favoring Q j when the population does not take the source i n t o account In this book, almost all infon scores were based on phrases explicitly espousing i n d i v i d u a l positions and were therefore o f the direct variety (odd values o f i ) . Also, the scoring t y p i c a l l y d i d not depend o n source. Thus statements by the president o f the U n i t e d States were given the same weight as quotes f r o m anyone else. The result for these cases was all infons being direct and having the same i - 7. The o n l y exceptions were the waste and fraud infons i n d i r e c t l y f a v o r i n g less defense spending, and the truck bombing infon supporting more troops in Lebanon. T h i s i n f o n carried the even index i - 2 since the population had to interpret the news to support more troops. Both the truck bombing and waste and fraud infons were in addition to other infons directly supporting these infons' positions. A l t h o u g h the positions w h i c h infons can favor often coincide w i t h the polled positions, as happened for the defense spending example, this need not necessarily be the case. F o r example, for the Democratic p r i m a r y , the polled positions were: Pro¬ M o n d a l e , Pro-Glenn, Pro-Others, and N o O p i n i o n . H o w e v e r , the infons i n the bandwagon analysis (Chapter 4) supported the six positions of: Pro-Mondale, ProGlenn, Pro-Others, Con-Mondale, Con-Glenn, and Con-Others. Therefore, index / refers to a l l ideas Q j under consideration, i n c l u d i n g those favored by corresponding subpopulations P j a n d / o r t h o s e supported by corresponding infons laijh Polls m i g h t show no persons favoring some ideas Q j. F o r instance, there were no p o l l measurements for persons opposed to Móndale (Con-Mondale). S i m i l a r l y , there were no infons associated w i t h some polled positions. For example, no infons were scored as f a v o r i n g N o O p i n i o n i n the Democratic p r i m a r y study. H o w e v e r , there w i l l usually be a substantial number o f positions w h i c h both people and infons w i l l favor. T o continue w i t h the Democratic primary, Pro-Mondale was both a polled position and a position w h i c h some infons supported. Index k - the same index k as that i n message A/¿ containing i n f o n Iaijk* Thus k indexes the fourth dimension o f Chapter 1. I n summary, i n f o n Iaijk refers to the i n f o n f r o m persuasive message Af¿, w i t h source and directness i , and supporting position Q j o f issue Q . W i t h this structure, a persuasive message is analyzed as a c o l l e c t i o n o f i n f o n s a l l a c t i n g o n the population. There can be a number o f different infons f r o m different sources indexed by i , all supporting the same position Q j.

as

opinion a

to

a

m

a

anc

a

9

a

a

a

a

a

a

a

a

a

a

144

A.4

Appendix

A

NOMENCLATURE

SIMPLIFICATION

For the rest o f this appendix, the nomenclature w i l l be s i m p l i f i e d corresponding to o n l y one issue being studied at a time, as was true for each case i n this book. The restriction t o a single issue means that index a referring t o the issue is always the same i n any one analysis and can be dropped f r o m a l l terms d u r i n g calculations for that issue. A l s o , believers PaAj >n position Q \ were necessarily aware o f that position, so index A specifying awareness is redundant and is therefore dropped f r o m PaAj and their associated percentages fl^y. W i t h the omissions o f subscripts a and A, the subpopulations and their percentages are simply denoted as Pj and By. A f t e r removal o f subscript a, the remaining subscripts for infons are i f o r i n f o n source, j for i n f o n position and k for the message carrying the i n f o n . a

A.5

I N F O N PROPERTIES

Since ideodynamics treats infons //yt as components o f message A/*, each i n f o n also has a validity characteristic o f the medium, an audience size, and a content score: 1. M e d i u m v a l i d i t y v/yjt o f i n f o n //yt = v t f o r the parent message A f t - T h i s equality holds regardless o f the source indexed by i o r the position favored Qu because the m e d i u m is the same for all infons o f a message. 2. A u d i e n c e size ayk(0 o f i n f o n /jyt = Qldt) regardless o f i and j since a l l corresponding infons are portions o f the same message A/£. 3. Content c/fjfc o f i n f o n //yjfe = content score for the i n f o n . T h i s score gives the extent to w h i c h the i n f o n favors position Qj I n this book, A P infons were given content scores i n terms o f typical A P paragraphs. T h i s provides a method for comparing the content strengths o f different infons. m

A.6

INFON PERSUASIVE FORCE

These i n f o n properties are used t o calculate f u n c t i o n s fijk d e s c r i b i n g "immediate persuasive force" o f i n f o n / y t at time t:

(A.4) fijk(t) =

the

cijwakV).

This function states that the population's exposure to an infon's persuasive power at time / is proportional to the infon's content and v a l i d i t y scores and to the audience size. The exposure o f the p o p u l a t i o n to all persuasive infons / / i t is g i v e n b y a " c o m b i n e d immediate persuasive force" function F//, assumed to be the sum o f the persuasive force functions fijktt) fot all infons received before time t. Therefore, (A.5)

Fijit)

«

£

fi (t) jk

for a l l k w i t h * t < T h i s summation o f i n d i v i d u a l functions w i t h exponential time dependent decays was also used by Hibbs (1979, see his Equation 2) to explore the issue o f unemployment versus inflation, a topic also studied i n this book.

A.7

INFORMATION INFLUENCING THE UNAWARES

O n l y infons direcdy supporting idea Qj (odd index numbers i ) should be able to raise the consciousness o f those unaware o f the issue. In contrast, indirect infons (even index numbers i) do not make a direct statement about the issue and w i l l be ignored by the unawares. Since only F n w i t h odd numbers i act o n the unawares, i t

Copyrights

Mathematics

of Ideodynanucs

145

is convenient to define function Fj summarizing the total information i n favor o f idea Qj available to the unawares where (A.6)

F/t)

=I

Fi/O

I

for all odd i .

A.8

INFORMATION INFLUENCING THE AWARES

T h e unawares should not be able to remember information w h i c h d i d not raise their consciousness. Those unawares w h o are able t o remember w o u l d already have become aware. I n contrast, the aw ares should be able t o remember. Therefore, information w i l l act on the awares and unawares w i t h different time courses. T o model this effect o f memory, consider an infon arriving at the population at time /£. T o compute the effect o f this i n f o n at a later time t i t is convenient t o d i v i d e the t i m e between tk and t i n t o very small intervals. Consider one such interval between time /' and (t' + dt ). T h e chances that information w i l l have been received b y the awares at this time w i l l be fijkit'). I f this i n f o r m a t i o n is lost f r o m the consciousness o f the awares i n an exponential fashion e" ( ' ' w i t h memory constant m . then the persuasive force due to information remaining at time f w i l l be the chances that the information was received at time t* times the probability that the i n f o r m a t i o n h a d n o t been forgotten, i n other words fijk*e However, the number o f awares m a y have been different at different times /' i n the time interval f r o m ik to time /. I n this case, the more awares there are at a given time the more w o u l d have been the amount o f information received i n that time interval r' to W + dt') and the greater w o u l d have been the persuasive force at time /. Therefore, the persuasive effect o f information f r o m that time interval is also proportional t o the fraction o f awares A(t') at time /' (see Section A . 2 above). A s a result, the residual persuasive influence at time / o f information absorbed at time t' is A(t')*fijk e~ ' ~ ' The total effect o f a l l information at time t f r o m i n f o n Ijjk is then the integral f r o m the time o f infon receipt by the population to the measurement time: t

0

m t t

mm

m

(A.7) gijtfO-

i AUWijkiO-e-nt'-O 'lc

m t

i

di-

where gijk(0 is the "remembered persuasive force" function describing the remaining information f r o m infon /y* available for action o n the awares at time t. I n a l l the polls i n this book, the total percentages o f N o t Sures, N o O p i n i o n s , and D o n ' t K n o w s were a l l typically less than 10 percent, so that awareness c o u l d be considered to be essentially 100 percent, i n w h i c h case A(t') -/ . T h i s assignment f o l l o w e d by substitution o f Equation A . 2 i n A . 3 , insertion o f that result together w i t h Equation A . l i n Equation A . 4 , and then further substitution i n Equation A . 7 yields

( A . 8 ) gijkft)-

\

hcijk+vAp-kaAp-e-Pi^

E x p l i c i t evaluation o f this integral yields

( A . 9 ) gijkO)

=

cijk'kvAP'kaAP 1 m - p

, • [e-PC'tk)

-

/ #

, e'^k)].

Copyrights

¡46

Appendix

A

T h i s function gutf!) w i l l approach a single exponential decay w h e n either (m» p) o r (p » m). I n the case o f much larger m , the second exponential i n Equation A.9 w o u l d q u i c k l y become negligible and w o u l d effectively mean that an infon is able to influence the population only a very short time after i t is received. I n other words, there w o u l d be very little memory o f an infon for the purposes o f o p i n i o n change. S h o u l d p be much greater than n% the first exponential in Equation A . 9 w o u l d be n e g l i g i b l e shortly after T h i s w o u l d be equivalent to the infon appearing and disappearing almost instantaneously, w i t h most o f an infon's effect being due to continued persuasion f r o m remembered infons. As m and p approach each other, the gijlc function has significant values at progressively longer times. Function G{j(0 can be constructed to describe the "total remembered persuasive force" acting o n the a wares due t o information from a l l previous infons As for Equation A . 5 , this function is the sum o f the persuasive forces f r o m the i n d i v i d u a l infons so that

Gijt)=

(A.10)

I gijkft)

for all k w i t h < t Since the awares, u n l i k e the unawares, are susceptible to infons w i t h both o d d and even indices i , the total information Gj i n favor o f position Qj w i l l be the sum o f the G functions over a l l indices i . That i s (A.I1)

Gj(t)

- I T

Gi/O

for a l l i . A g a i n , this equation has a f o r m similar to H i b b s ' (1979) Equation 2. A s noted i n Chapter 1, E q u a t i o n A . 1 0 , a l t h o u g h successful for t h i s b o o k , i g n o r e d the hardening o f the v i e w p o i n t s o f the various subpopulations due to reinforcing infons. I n the concept or reinforcement, a subpopulation P supporting position Q w i l l be r e i n f o r c e d b y infons l%fk f a v o r i n g the same p o s i t i o n Q. Therefore, the persuasive force o f infons lijk p u l l i n g subpopulation members away f r o m p o s i t i o n Q a n d t o w a r d p o s i t i o n Q y w i l l no longer be G / b u t w i l l be d i m i n i s h e d . " D i m i n i s h e d persuasive force" j u n c t i o n s Hjfit) can be constructed to account for this decreased e f f e c t These functions H; include the persuasive force o f i n f o n s s u p p o r t i n g p o s i t i o n Qj being d i m i n i s h e d oy r e i n f o r c i n g i n f o n s f a v o r i n g position Q . r

r

r

r

r

r

One function w h i c h w o u l d have the right properties w o u l d be

H
(A.12)

jy

+ l)

r

where dj is a d i m i n u t i o n constant describing the decrease in the persuasive forces o f infons lijk due to the reinforcing effects o f infons I f dj is very s m a l l , there is no reinforcement and Hj = Gj As dj increases, the reinforcing powers o f infons Ifrk also g r o w . A c c o r d i n g to Equation A.12, the greater the reinforcing power G the lower w o u l d be the conversion force due to Gj Corrected persuasive force functions //y>can be further m o d i f i e d to take i n t o account people becoming saturated due to continued and frequent repetitions o f the same infons favoring position Qj: r

r

r

t

r

n

(A.13)

Hjr
(dj^GriO

+ dffGjit)

+ 1).

As for dj , djj is a constant for the saturation effect due to over repetition o f infons hjk- The addition o f the djfGj t e r m i n the denominator again means that there is little saturation due to more infons lijk >f djj is small. O n the other hand, i t is easy to saturate i f djj is large. r

Copyrights

Mathematics A.9

of Ideodynamics

147

E F F E C T OF I N F O R M A T I O N O N T H E P O P U L A T I O N

Ideodynamics permits any number of positions Qy For example, one position may be aware but u n c o m m i t t e d . Some o f the Don't K n o w s measured i n o p i n i o n polls may have been o f this type w h i l e others may have been unaware. In all the studies i n this book except that for the Democratic primary, i t was assumed that all the Don't K n o w s were unaware and stayed unaware throughout the p o l l i n g period. For the bandwagon analysis for the Democratic primary, i t was assumed that the N o Opinions were actually aware but uncommitted. The actual situation may have been somewhere in between. However, the total percentage o f the population involved was typically less than 10 percent, so the calculations are not strongly dependent o n assumptions for the Don't K n o w s . A general differential equation can be w r i t t e n to describe the time dependent changes in the number o f people A(t)»Bj(t) belonging to subpopulation Py

(A.14)

d

[A(0*Bj(t)I

.

_

«I

dt

fyrfHyrityMthBrit)

j',r

-

k jyllfj{thM^Bp)

I

2

klj'j.Fj'(tMl-A(0)

I

-

u*A(t)*Bj(t).

I n the terms o n the right, the first double s u m o v e r / a n d r gives the gain i n subgroup />ydue t o recruitment f r o m other subpopulations Pr* w h i l e the second double s u m o v e r / and r gives the loss o f members f r o m due t o i n f o r m a t i o n favoring other positions. The single sum over / gives the g a m i n subgroup P. due t o r e c r u i t m e n t f r o m the unawares, and the last t e r m reflects loss f r o m P y t o unawareness due to forgetting. The detailed explanation of these terms is as follows;

Recruitment f r o m Other Subpopulations-First Double Sum in Equation A.14 Members o f a particular subpopulation P can be persuaded by persuasive force functions Hyr t o j o i n subgroup Py Resistance t o change due to reinforcing infons I irk and saturation w i t h infons I n't has already been incorporated into Hy (Equation A . 1 3 ) . T h e n u m b e r persuaded at t i m e t is p r o p o r t i o n a l t o the pool o f p o t e n t i a l converts P The size o f this pool f r o m Section A . 2 is A(t)*Br(t). T h e number converted is also p r o p o r t i o n a l to the persuasive force function Hy describing the effectiveness o f infons iij'k i n the face o f r e i n f o r c i n g infons I irk* T h e constant o f proportionality is *2yy, the "persuasibility" constant w i t h subscript 2 indicating that r

r

r

r

the c o n s t a n t i s a p e r s u a s i b i l i t y c o n s t a n t .

The summation is over a l l possible f and r w i t h constant ¿2/77 having values o f either * 2 texo. Constant * 2 has the same value f o r eacn *2/>y i n any one equation where persuasive force f u n c t i o n //y>can actually persuade members o f subpopulation P to change their opinions f r o m position Q and j o i n subpopulation / j f a v o r i n g idea Qy T h i s constancy for ¿2 values is consistent w i t h Chapter 1 postulating that the persuasibility constant measures the closeness o f the issue to the core beliefs o f the population. However, ¿2 is a constant w h i c h can change f r o m issue to issue. Constant * 2 / r / P° ' ^ he positive for any c o m b i n a t i o n o f indices. F o r instance, *2/V/ * 2 is possible for t w o d i f f e r e n t / but the same r a n d j . This w o u l d mean t h a t t w o types o f persuasive forces can persuade members o f the same target p o p u l a t i o n P t o j o i n the destination population Py A n example w o u l d be o

r

r

r

c a n

D e

s t u

a t e

t o

m

r

Copyrights

148

Appendix

A

i n f o r m a t i o n favoring both more and same defense spending persuading people f a v o r i n g less spending to support same spending. I n contrast, kij'rj - 0 if a transition is not permitted. For example, a persuasive force function favoring less defense spending should usually not persuade people favoring less spending t o favor m o r e spending. The details o f the * 2 / r y array correspond to the postulated " p o p u l a t i o n conversion models for the awares" (see Figures 5.2, 5.18, 5.29, 5.30, 5.35, 5.39, and 5.45 o f Chapter 5 for examples).

Loss o f Believers-Second Double S u m i n Equation A . 14 T h i s sum merely reflects the fact that the population stays constant i n size. I f people are persuaded b y persuasive force ///'/to leave p o p u l a t i o n Pj to j o i n population P w i t h *2/;> • *2« then there w i l l not o n l y be a gain i n population P but also a loss f r o m population Py T h i s loss results i n the second sum i n Equation A.14 h a v i n g a negative sign. The magnitude of the loss is the same as the gain by p o p u l a t i o n P , so the terms i n the first t w o sums o f the equation have the same form. r

r

r

Conversions o f Unawares to Awareness-Single

Sum i n Equation A.14

These conversions are based o n the argument that the unawares cannot remember i n f o r m a t i o n so they learn about the issue through infon persuasive force functions Fjit) (Equation A.6). The rate at w h i c h the unawares P\j move to hold position Qj due t o Fjit) w i l l be proportional both t o the number o f people (/ - A(t)) i n Py and to Fj% w i t h an "attentiveness" constant o f p r o p o r t i o n a l i t y kjjj. The subscript / denotes an attentiveness constant (Chapter 1). Constant kjjj has the same structure a s constant * 2 i W - However, since the o n l y t a r g e t population under consideration is the unawares Pfj, there is no need t o specify the target population i n the c o n s t a n t It is sufficient t o specify the index / f o r the persuasive force F / and index j for the destination population Pj. As w i t h * 2 / > / ¿7/7 either has a constant value denoted by kj or a value o f zero, depending on a chosen "population conversion model for the unawares." The sum i n Equation A . 1 4 is over all / . I f there is no contribution f r o m a function F/, the consequence w o u l d be kjjj * 0 for the corresponding attentiveness constant. For instance, i f the m o d e l proposes that a l l functions Fj' first m o v e the unawares into a population Pj o f aware but uncommitted, kjjj - kj o n l y for this one value o f / A l l other kjjj - d .

Forgetting o f a P o s i t i o n - L a s t T e r m o f Equation A.14 I n the reverse process, i t is assumed that any aware can forget the issue and become unaware. The unlearning o f the issue is characterized by constant u w i t h the rate o f conversion o f awares favoring position Qj to unawareness being u*A(t)*Bj(t). I n this expression, the chances o f forgetting are the same f o r all i n d i v i d u a l s so the t o t a l loss f r o m awares f a v o r i n g idea Qj is p r o p o r t i o n a l t o the size o f the corresponding subpopulation A(t) Bj(t). As argued for Equation A.7, the D o n ' t K n o w s i n this book were usually less than 10 percent, so that A(t)*4 T h e n Equation A.14 becomes m

m

dB/0

(A.15)

V di

lk2frrHj'r(t)*Br
-

k jjr*HjjUWj(t)}. 2

j ,r

w i t h n e g l i g i b l e r e c r u i t m e n t o f persons f r o m the unawares and essentially forgetting o f the issue.

no

Copy rights

Mathematics

of Ideodyruimics

I f At is a small time interval, then integration o f Equation A.IS yields (A.16)

B/0

149

f r o m t-At t o /

-Bj(t-&)

t-At

j\r

I f the H and B functions do not change substantially d u r i n g the time interval t-At to / , then Equation A . 1 6 c a n be approximated by (A.17)

Bj(t)=Bj(t-At)

W

/*2/r////r<0^

I If

This deterministic equation has no stochastic terms because i t was shown (Chapter

S,

Figures S.10 and S.23) that calculations o f Bj several t i m e intervals At after the beginning o f the c o m p u t a t i o n are relatively independent o f i n i t i a l values f o r the variously. A.10

MODIFICATIONS

FOR A P I N F O N S A S S U M I N G N O U N A W A R E S

For all examples i n this book i t was assumed that the A P was representative o f mass media messages w i t h o u t pretending that this w i r e service included all mass media communications. Furthermore, the analysis used only random samples rather than all A P dispatches identified as appropriate t o the issue being studied. Therefore, the equations needed to be m o d i f i e d to account for the facts that o n l y some o f the relevant dispatches were examined and that some relevant messages may not have been identified by the methods used to find the pertinent communications. I n Chapter it was demonstrated that functions g¡jk for all A P infons gave good projections for a l l six examples using an exponential decay w i t h a h a l f - l i f e o f approximately one day. W i t h such a short h a l f - l i f e , it is l i k e l y that the persistence c o n s t a n t p is the dominant term, w i t h forgetting being extremely rapid such that m » p so that the second exponential i n Equation A.9 can be neglected, g i v i n g

S

(A.18) where

gijkO) -

( A . 19)

kj\p = k AP*kaAP

kAP*ci *e-P("k) jk

(m-ph

1

v

D u r i n g the total t i m e for w h i c h o p i n i o n calculations were made, let T- total number o f A P dispatches identified as potentially relevant, 7/ ~ number o f irrelevant A P dispatches identified erroneously using the search command passed to the Nexis data base, 7¿/ - number o f pertinent A P dispatches missed during the identification process, R - n u m b e r o f dispatches actually retrieved, /?/ = number o f irrelevant dispatches a m o n g the number retrieved, R\{ = number o f pertinent b u t missed dispatches, 5 -/?-/?/ - number o f dispatches among the retrieved w h i c h were shown to be relevant d u r i n g the f i n a l scoring process, and G'j « persuasive force function calculated f r o m the random subset o f all relevant infons. Recall that Gj is defined as the persuasive force function i f all relevant infons and not a random subset is used. I f i t can be assumed that dispatches i n tfie Tm group w h i c h were not identified as relevant had the same infon persuasive forces as those in the identified group T. then ( A.20)

G j = IS/(T-T

F

+

T )hGj M

Copyrights

150 since G' represents the fraction of G corresponding to the truly relevant number of dispatches S divided by the number of dispatches from which the sample should have been drawn, namely the relevant dispatches actually identified (T - T ) together with the ones which were missed during the search (T ). If the retrieval is random, then the fraction of irrelevant dispatches should be the same among the retrieved stories and the identified dispatches from which the random set was drawn, so that (A.21) R I / R = T I/ T. From Equations A.20, A.21, and the relationship S = R - R , the result is (A.22) Gj = (T/R)-([S + (R/T) T ]/S)-G'j. In other words, the G function neededfor the ideodynamic equations can be replaced by the G' functions calculatedfrom a random subset of the retrieved messages if the G' is multiplied by (T/R), the ratio of the total dispatches to those retrievedfor the analysis, whenever T = 0, meaning that the procedures used for the message identification did not miss any important messages. Thus the assumption of no missing messages means that Equation A.22 becomes (A.23) G j = (T/R)'G'j. Besides sampling mass media infons from AP dispatches alone and then retrieving only a random set of stories identified as relevant, it was further assumed that d = d = 0 for all H so that H = G from Equation A.13. This assumption was used because it gave good calculations for all the cases in this book. Therefore, in applications to a random sample of relevant messages, (A.24) H j = (T/R)'G'j. It is further convenient to define a "modified persuasibility constant" k of the form (A.25) k = k -k = k -kv -ka / (m-p) using Equation A. 19. Substitution of Equations A.9, A.18, A.24, and A.25 in Equation A.17 gives (A.26) I

M

1

M

M

j r

jj

j r

j r

j

r

2

2

for all (A.27)

2

j

G"f«)

f

A P

2

and

r

= I

A P

AP

where

G'j

are

"skeleton

persuasive

force"

functions

of

form

cij-fe-PC-'k)

¡Je

The summation is over all i andfor k with tk < t. For consistency with Equation A.26, Equation A.27 uses index j ' instead of index j employed in earlier persuasive force functions. Empirically measured scores for

cij'k are denoted

Sij"k Scores

Sij"k carry the same index i for the source of the information and k for the message for which the score was obtained. However, index j " for the position which a score favors might not always coincide with index j ' for the position of the infon

lij'k . The reason is the occasional difficulty in assigning scores to a particular position. For instance, an exhortation to cut an increase in defense spending might be interpreted by some

Mathematics

of ¡deodynamics

151

people as still favoring more defense spending, although less o f an increase, w h i l e others might feel that defense spending should not be increased but held the same. Therefore, this exhortation might be scored as favoring same spending but might actually support both more and same spending. I f the average score favoring same defense spending reflects a significant component favoring more spending, then there should be a mechanism t o permit the score favoring same spending to contribute t o persuasive forces for both same, and more spending. T h i s is done by i n t r o d u c i n g " r e f i n i n g w e i g h t " constants H^I/V". These weights specify the c o n t r i b u t i o n o f a score syk to content scores for infons lifk f a v o r i n g positions Qt\ Weights permit a measured score to contribute to the content score o f more than one infon using the f o l l o w i n g equation: (A.28)

cij'k - Z

"ij'j'Sif'k-

The summation here is over a l l positions indexed by / ' w h i c h the scores can favor. In fact, some o f these positions may not even coincide w i t h the positions Qf o f the infons used i n the analysis. For example, i n Chapter 5 scores favoring waste and fraud b y defense contractors were considered t o contribute to infons f a v o r i n g less defense spending. I n the most s t r a i g h t f o r w a r d case, however, every score measured to favor position Qy w o u l d also contribute o n l y to the content score o f an i n f o n supporting the same position Qf. I n this case, wij'j" w o u l d o n l y be positive when / « j ' \ and positions o f infons indexed by / w o u l d coincide w i t h the positions o f the scores indexed by J Substitution o f Equation A.28 i n Equation A.27 yields (A.29)

GyY/J-

X

wijy*sifk*e-P< k). M

ij"Jk

for a l l * and / ' and for all k w i t h tk
2

n

Copyrights

152

Appendix

A

issue (see Chapter 1). The r e f i n i n g weights then reflect the r e m a i n i n g m i n o r differences i n the persuasibility constant and/or the effectiveness o f statements f r o m different sources supporting different positions. Since the persuasibility constant * 2 - - a n d hence the m o d i f i e d persuasibility constant Jt'2--can vary f r o m issue to issue (see Equation A . 2 6 ) , i t was necessary to f i n d the o p t i m a l value for this constant. I t was also i m p o r t a n t to o p t i m i z e the persistence constant and the r e f i n i n g w e i g h t s iVy* A l l o p t i m i z a t i o n s were performed for each polled topic by m i n i m i z i n g the differences between the calculated Bj(t) and the BjU) f r o m published polls. The method was a least squares optimization where a number o f possible values for the variable parameters were chosen by the analyst and computer simulations were made using Equation A . 2 6 . F o r each set o f values, a public o p i n i o n time trend was computed over the entire time interval o f the polls and deviations were measured between the actual p o l l points and the calculated o p i n i o n curves for all points o f v i e w . There were as many o p i n i o n projection values as positions p o l l e d . T h e deviations between measured and projected p o l l percentages were computed for all curves, and squared and averaged to g i v e the mean squared d e v i a t i o n ( M S D ) . Optimizations were performed by plotting various trial constants against this M S D to f i n d the value w i t h the m i n i m u m M S D . These plots show the sensitivity o f the fit to values o f the o p t i m i z e d constants. I f the M S D plots are relatively flat for increasing values o f the parameter, then the calculations are relatively insensitive t o changes in the constant i n the region o f the m i n i m u m M S D . Sometimes, the square root was taken o f the M S D to give the root mean squared deviation ( R M S D ) . T h i s R M S D can be compared directly w i t h the standard error calculated by taking the square root o f the reciprocal o f the sample size times the poll percentage times the quantity 100 m i n u s the p o l l percentage. I f the p o l l points deviate randomly f r o m the calculated opinion, the differences between the calculated o p i n i o n and calculated p o l l values should f o l l o w a n o r m a l d i s t r i b u t i o n w i t h the standard deviation being equal to the R M S D . Unfortunately, i t is not permissible to perform the usual regressions-without making assumptions w h i c h are difficult to j u s t i f y rigorously—between o p i n i o n p o l l values and o p i n i o n s calculated f r o m Equations A.26 and A . 2 9 . T h i s is because o p i n i o n s at later times are dependent o n o p i n i o n s at earlier t i m e s . Since the independence conditions needed for the regressions are not met, the more descriptive M S D and R M S D were used f o r statistical comparisons. These calculations also provided a convenient set o f numbers for o p t i m i z i n g the parameters o f the model. w

A . I I COMPARISON WITH UNIFORM

DISTRIBUTION

O n e o f the most fundamental questions is whether the o p i n i o n projections are any better than those obtained by p i c k i n g p o l l points at random. Trie straightforward test i n this book used M o n t e C a r l o simulations. A t each p o l l t i m e , a set o f r a n d o m p o l l points a l l adding to 100 percent was calculated f r o m a set o f r a n d o m numbers. Then squared deviations were calculated between each o f the measured p o l l points and its corresponding value d r a w n at random. These squared deviations were then averaged to give the precise equivalent to the M S D computed using optimal constants i n the ideodynamic equations. Therefore every M o n t e C a r l o s i m u l a t i o n based o n r a n d o m numbers also y i e l d e d the exact analog of the M S D calculated f r o m A P dispatches. A thousand o f these s i m u l a t i o n s were p e r f o r m e d for each set o f o p i n i o n projections y i e l d i n g 1,000 M S D ' s based on r a n d o m l y calculated p o l l points. The fraction o f these MSD's w h i c h were less than the MSD's calculated by ideodynamics gave the p r o b a b i l i t y that the ideodynamic estimates c o u l d be obtained by chance alone.

Copyrighte

M ut ^mattes A.12

MODIFICATIONS UNAWARES

FOR

of Ideodynamics

AP INFONS A S S U M I N G

153

NON-NEGLIGIBLE

In Section A.10 above, the assumption was made that there were essentially no unawares. I f this g r o u p cannot be neglected, then o p i n i o n calculations are made using Equation A . 1 4 . I f both sides o f Equation A.14 are summed over all j the result is (A.30)

dáíÜ-II

k]fj-FjitHl-A(t))l-u.A(0

dt

jj'

since the sum over a l l Bj(t) - / by the definition o f Bp). Also, the terms i n the t w o d o u b l e sums o f E q u a t i o n A . 1 4 c a n c e l after s u m m a t i o n o v e r j . L i k e the approximation o f the solution o f Equation A . I S by Equation A . 17, the solution o f Equation A.30 can be approximated by (A.31)

A(t)

+ (I X

= A(t-At)

kij'j'FjVMl-AU-At))}

-

u*A(t-At))'At.

jj'

So long as functions Fjit) can be computed, this equation can be solved i f constant u, an i n i t i a l value for A(t) and all ¿7/7 are also provided. F r o m Equations A . l , A . 2 , A.3, A . 4 , A . 6 , and A . 2 8 , t

(A.32)

Fjit)

=

£

WkaAP^ijr^ij're'P

( M

k>

for a l l k w i t h t < i¿ and for a l l o d d i. Since ku*j - kj o r zero (see discussion f o l l o w i n g Equation A . 1 4 ) , and since kjjj i n Equation A . 3 1 is always m u l t i p l i e d by kvAP kaAP> convenient to define a " m o d i f i e d attentiveness" constant a n d

li

(A.33)

k'j=kj*k kaAP vAr

T h e n , Equations A . 3 1 and A . 3 3 , together w i t h analogs to Equations A . 2 0 - A . 2 3 , yield (A.34)

A(t)

. A(t-At)

+ U(TIR>%

k']jj'F'jHHl-A(t-At))l

-

u*A(t-At))*At

jj'

for all j and f where (A.35)

F'jit)

-

£ ij\k

w jj--*sifk*e-P'' ' k) i

t t

for all odd i . all7", and k w i t h t < í¿. W i t h Equations A.34 and A.35 and postulated values o f k'jjj-m addition to the values o f Section A . 1 0 a b o v e - i t is possible to calculate A(t) at intervals o f At i f an i n i t i a l value o f A(t) corresponding to those w h o had not yet heard o f the issue, is available f r o m a p o l l . Then, w i t h the postulated values o f k'jjj and calculated A(t) it is possible to calculate the fraction o f the total population favoring position Qj by approximating the solution o f Equation A.14 by %

9

(A.36)

A(t)*Bj(t) + (TfR)*Af(

+ (TlR)*AfI

-

A(t-At)»Bj(t-At)

I

Gy(/>/*'2/r/*f^^^ k'lffF'jitHl-A(t~At))l

X

-

wA(t-At)*Bj(t-At).

J

Copyrighte

154

A.13 EXTENSIONS TO VERY LONG TIMES As noted in Section A.2 above, the assumption was that the population was constant during the time period of the calculation. In other words, birth, death, and migration into and out of the population were ignored. These assumptions are likely to be valid for the periods under a year used for the Lebanon and Democratic primary examples. However, as times increased to over nine years for the case of defense spending, this assumption may start to break down. To account for death, it is possible to introduce another term like that at the end of Equation A.14. To account for birth, it is necessary to add additional terms to Equation A.14 adding members to the population of unawares. Then their conversion to awareness can follow the discussion in Section A.12 above. Furthermore, in the absence of population constancy, all the equations of this chapter will also need to be modified so that all computations are in terms of absolute number and not percentages.

There is another place where the model may need to be changed. The constants in Equation A.14 and its derivative equations may slowly change with time as society changes. For instance, the reputation of the medium may slowly drift. Therefore, for truly long term studies, the constants may have to be converted into time dependent functions reflecting the accumulated experience of the population. A.14 MODELS WITH NO DEPENDENCE ON SUBPOPULATIONS Unlike ideodynamics, some models do not calculate opinion percentages based on a subdivided population. Instead, only forces on the population as a whole are taken into account. Ideodynamics can also be used to calculate opinion based solely on persuasive forces by making the assumption that opinion change is sufficiently slow so that dBj/dt ~ 0. In this case, Equation A.15 converts to (A.37) if

With dBj/dt ~ 0, all the Bj are constants and calculable given only the persuasive forces Hjr and the persuasibility constant kjr. Equation A.37 is actually a system of simultaneous equations, one for each position with subscript j . With j equations, there is a unique solution for each of the Bj. Given no rapid change in Bj, the H functions which drive the change must also be reasonably constant at calculation time t.

Appendix B

Data for Calculating Opinion Change

The data used for projecting public o p i n i o n were o f t w o types: (1) t i m e series o f public o p i n i o n polls f r o m published data and (2) A P dispatches relevant to the polled topics retrieved f r o m the Nexis electronic data base sold by Mead Data Central, 9393 Springboro Pike, P.O. B o x 9 3 3 , D a y t o n , O h i o , 4 S 4 0 1 . l i t i s data base contained a l l A P dispatches since January 1, 1977. Polls and A P dispatches were obtained f o r six issues. A l l retrievals were restricted to text w i t h i n f i f t y words both before and after one o f the k e y w o r d s used i n the o r i g i n a l search. T h e f i f t y - w o r d l i m i t e l i m i n a t e d irrelevant sections o f the dispatches a n d was chosen because the w o r d s at the beginnings and ends o f the retrieved regions t y p i c a l l y showed transitions t o other topics. Articles concentrating on an issue t y p i c a l l y had the key search words w i t h i n 100 words o f each other (fifty words after one search w o r d and fifty words before another). These articles were automatically retrieved i n their entirety. A l l A P searches began before the first p o l l p o i n t i n order t o account for the residual effects o f p r i o r messages. T h e search was made f o r a l l dispatches u p to six months before the first point i n the p o l l series unless the s i x - m o n t h period extended before the beginning o f the data base on January 1, 1977. I n that case, the search began w i t h this date. A l l searches stopped at the end o f the p o l l i n g period.

B.l

DEFENSE

SPENDING-1977-1984

Four variant p o l l series were found f r o m 1977 t o 1984 for the issue o f whether more, same, o r less s h o u l d be spent o n defense (Table B . l ) . A l t h o u g h earlier polls existed, they were n o t studied because the N e x i s data base o n l y contained A P dispatches back to 1977. I n a l l polls, the vast majority o f the population had definite opinions and were d i v i d e d i n t o three groups: those f a v o r i n g more, same, o r less defense spending. There was also a group o f Don't K n o w s o r N o t Sures, t y p i c a l l y i n the range o f 5-10 p e r c e n t T h i s subpopulation was subtracted f r o m the total and the p o l l data was renormalized among all those w i t h an o p i n i o n . F o r ideodynamics, this step effectively assumed that the s m a l l n u m b e r o f persons w i t h no o p i n i o n stayed i n that category. Even i f this assumption was not entirely v a l i d , the numbers were sufficiently small that the results w o u l d not have been much affected. Fortunately, the same t i m e trend was seen f o r a l l four polls after the D o n ' t K n o w s were removed (Figure 2.1). F o r this figure, n o adjustments were made

Copyrights

156

Appendix

B

beyond the r e m o v a l o f those w i t h no o p i n i o n . G i v e n the agreement between the different poll series, the data were pooled. Relevant i n f o r m a t i o n i n the Nexis data base was identified by searching the full texts o f a l l dispatches using combinations o f key words chosen by the investigator. For defense spending, the search was for ( D E F E N S E o r M I L I T A R Y o r A R M S ) w i t h i n five words o f ( B U D G E T ! or E X P E N D I T U R E o r S P E N D ! o r F U N D ! ) f r o m January 1, 1977, t o A p r i l 1, 1984. The T permitted the trailing characters to be a n y t h i n g , so that both budgeted and budgetary w o u l d have been f o u n d w i t h BUDGET!. The search command yielded 9,314 dispatches w i t h the data base numbering the dispatches i n reverse chronological order, number one being the most recent and number 9,314 being the earliest. F r o m a random 692 o f these dispatches, text was retrieved i f i t was w i t h i n f i f t y words o f one o f the seven key search words given i n the previous paragraph. I f t w o key words occurred w i t h i n 100 words o f each other, the entire intervening text was collected. The total retrieval was 820,000 characters of text A sizable number o f dispatches were not about American defense spending. As soon as this became clear, the retrievals were stopped. The analyses using the 692 A P dispatches f r o m 1977 t o 1984 and the poll data in Table B . l were extended in t w o ways. First, additional p o l l data were collected f r o m the Roper Center at the University o f Connecticut (see Table B . l below for more details) f r o m January 1977 to A p r i l 1986. A time series o f s i x t y - t w o separate polls c o u l d be obtained by p o o l i n g these additional polls w i t h those i n Table B . l . Besides polls f r o m the National O p i n i o n Research Center, N B C News, and the Roper organization, pooled polls also contained results f r o m A B C N e w s , C B S N e w s , and the G a l l u p and Harris organizations. The same commands used for i d e n t i f y i n g the 9,413 dispatches f r o m 1977 t o 1984 were used again to locate 10,451 stories f r o m January 1, 1 9 8 1 , to A p r i l 12, 1986. O f these, 1,067 w e r e retrieved randomly for extending the Study to 1986. T o determine the importance o f stories on waste and fraud on o p i n i o n o n defense spending, the Nexis data base was further searched for ( D E F E N S E or M I L I T A R Y o r A R M S ) w i t h i n five words o f ( W A S T E or F R A U D or C O R R U P T I O N ) from January I , 1977, to A p r i l 12, 1986, y i e l d i n g 878 dispatches o f w h i c h 512 were retrieved at random for text w i t h i n fifty words or one o f the search words.

B.2

TROOPS I N L E B A N O N - - 1 9 8 3 - 1 9 8 4

A single p o l l series p r o v i d e d o p i n i o n for whether more, same, or less troops should be sent to Lebanon i n 1983-1984 (Table B.2). As for defense spending, the N o Opinions were i n the 5-10 percent range and were subtracted f r o m the total. The other opinions were renormalized to 100 percent and the resulting values were used for the remainder o f the calculations. Pertinent A P dispatches were again retrieved f r o m the Nexis data base, searching for ( L E B A N ! and ( ( A M E R I C A ! or U.S. o r U N I T E D S T A T E S ) preceding b y t w o words or less the words ( T R O O P o r M A R I N E or F O R C E ) , f r o m M a r c h 26, 1983, to January 17, 1984. The search began six months before the first p o l l p o i n t and ended w i t h the last p o l l date. The search yielded 1,517 dispatches among w h i c h 467 were retrieved at random for 1,570,000 characters o f t e x t As for defense spending, the retrieval was for text w i t h i n fifty words of one o f the search words.

B.3

DEMOCRATIC

PRIMARY-1983-1984

The polls for candidate preference before the Iowa caucuses were f r o m A B C News and ran f r o m June 19, 1983, to February 15, 1984 (Table B.3). D u r i n g this t i m e , the major candidates were John G l e n n and W a l t e r M o n d a l e . There were other

Copyrighte

157 candidates, but none of their percentages ever exceeded 15 percent so those percentages were all pooled. AP dispatches were retrieved i f they contained the name of at least one of the candidates tallied in the polls. The search was for (REUBEN preceding ASKEW by two words or less) or (ALAN preceding CRANSTON by two words or less) or (JOHN preceding GLENN by two words or less) or (GARY preceding HART by two words or less) or (ERNEST preceding HOLLINGS by two words or less) or (JESSE preceding JACKSON by two words or less) or (GEORGE preceding MCGOVERN by two words or less) or (WALTER preceding MONDALE by two words or less) from December 19, 1982, to February 15, 1984. As for troops in Lebanon, the search began six months prior to the first poll date. Although the last names would probably have been sufficient for relatively rare names like Cranston and Mondale, it was necessary to include the first names due to more common names, of which Jackson would have been the most ambiguous. Therefore, the search used the condition that the first name of every presidential candidate must precede the last name by no more than two words. With this condition, a middle initial could be present in the names some of the time and missing in others. The search yielded 2,435 dispatches of which 425 (1,100,000 characters) were retrieved at random for text within fifty words of a search word. B.4 ECONOMIC CLIMATE--1980-1984 The polls for public opinion on the economic climate were taken by ABC News and covered a three-year period from March 1981 to January 1984 ( Table B.4 ). The No Opinion category was very low at all times, never exceeding 3 percent, so this fraction was subtracted and the other percentages were renormalized to 100 percent for the calculations. AP dispatches in the Nexis data base were searched for (ECONOM! within twenty-five words of (CONDITION! or HEALTH or PROSPECT! or FUTURE or FORECAST! or OUTLOOK! or PROJECT!)) from September 6, 1980, to January 17, 1984. This search also began six months before the first poll date. A total of 12,393 dispatches were identified of which 461 (730,000 characters) were retrieved at random for text within fifty words of a search word. B.5 UNEMPLOYMENT VERSUS INFLATION--1977-1980 Polls from NBC News asked about the relative importance of unemployment and inflation ( Table B.5 ). The 3 percent or less of the population who were not sure were subtracted and the remaining percentages were renormalized. AP dispatches on this topic were identified searching for (UNEMPLOY! within twentyfive words of INFLATION) from January 1, 1977, to August 23, 1980. The search began with the beginning of the data base in January 1977, about three months before the first poll date. The search identified 1,591 AP dispatches of which most (1,582 with 2,300,000 characters) were randomly retrieved for text within fifty words of a search word. B.6 CONTRA AID--1983-1986 Polls on the topic of whether aid should be sent to the Contras fighting the government of Nicaragua were obtained from the four organizations listed in Table B.6. Despite significant wording differences from poll to poll, there was very little change in opinion during the entire polling period, so all polls were pooled. The criteria for choosing the polls was that they ask about American opinion on either aid with no qualifiers or on both military and nonmilitary aid. No published polls found

158

Appendix

B

before the last p o l l i n Table B.6 were excluded i f they met these criteria and were i n the P O L L data base at the Roper Center. T h e CBS-New York Times p o l l was obtained independently f r o m C B S N e w s and was the o n l y a d d i t i o n a l p o l l found meeting the criteria given above. For a l l tables, the p o l l i n g date was assumed to be the m i d p o i n t between the beginning and the end o f the p o l l i n g period. Where no date was given, the midpoint o f the polling month was used.

Copyrights

The data were f r o m four different variants o f polls on defense spending compiled w i t h the aid o f B. I . Page and R. Y . Shapiro, and their colleagues at the N a t i o n a l O p i n i o n Research Center. The symbols are the ones used i n Figure 3 . 1 . W h e n o n l y the month was available, the poll date was assigned to be the middle of the month. P O L L V A R I A N T N B C 1 : Source: N B C N e w s , 30 Rockefeller Plaza, New Y o r k , N Y 10020. Question: D o y o u t h i n k that the defense budget for next year should be increased, decreased or should i t be kept the same as i t is now? Responses: (1) Increased, (2) Kept the same as now, (3) Decreased, (4) N o t sure. P O L L V A R I A N T N B C 2 : Source: N B C N e w s . Question: D o y o u t h i n k the federal government's spending next year on defense and the m i l i t a r y s h o u l d be increased, decreased, or kept about the same? Responses: (1) Increased, (2) Kept about the same, (3) Decreased, (4) N o t sure. P O L L V A R I A N T GSS: Source: General Social Survey, N a t i o n a l O p i n i o n Research Center, 6030 E l l i s Ave., Chicago, I L 60637. Question: W e are faced w i t h m a n y problems i n this c o u n t r y , none o f w h i c h can be s o l v e d easily o r inexpensively. I ' m going t o name some o f these problems, and for each one I ' d l i k e you to tell me whether y o u think we're spending too m u c h money o n i t , too little m o n e y , o r about the right a m o u n t . The m i l i t a r y , armaments a n d defense. Responses: (1) T o o little, (2) About right, (3) T o o much, (4) Don't k n o w . P O L L V A R I A N T ROPER: Source: Roper Center f o r Public O p i n i o n Research, P.O. Box 440, Storrs, C T 06268. Question and responses: Identical to those for variant GSS above. Table BJ.

Polls on the desirability

Symbol and Poll Source A V A a V • V O • • • A 0 0 0 V A 0 V A 0 A

GSS ROPER GSS NBC1 ROPER NBC1 ROPER NBC1 NBC1 NBC1 NBC1 GSS NBC2 NBC2 NBC2 ROPER GSS N BC2 ROPER GSS NBC2 GSS

of increasing

defense spending.

Percent Response Date

03/ /77 12/07/77 03/ /78 10/17/78 12/06/78 12/12/78 02/05/79 09/ /79 12/12/79 01/18/80 01/30/80 03/ /80 01/22/81 02/ /81 11/17/81 12/09/81 03/ /82 03/30/82 12/08/82 03/ /83 0 1 / /84 03/ /84

0)

a)

(3)

(4)

23.6 23 27.0 28 31 24 41 38 51 63 69 56.3 65 63 34 29 29.4 24 19 24.1 23 17.3

45.4 40 43.6 45 35 47 35 36 31 21 19 25.7 23 25 47 38 35.8 47 37 37.8 46 41.2

22.9 24 21.8 21 23 22 16 16 9 8 5 11.5 6 8 14 27 30.1 25 38 32.5 26 38.1

8.1 13 7.6 6 11 7 9 10 9 8 7 6.5 6 4 5 7 4.7 4 6 5.6 5 3.5

Copyrights

Table

B.2.

ABC

News

Poll

on the stationing

of American

troops in

Lebanon.

Results f r o m A B C News Poll, 7 West 66th Street, N e w Y o r k , N Y 10023; Report 95 i n 1984 compiled w i t h the aid o f B . I . Page and R. Y . Shapiro and their colleagues at the National O p i n i o n Research Center. The question was: W o u l d y o u say the U.S. should send more troops to Lebanon, leave the number about the same, o r remove the troops that are there now? The responses were; Send more troops; Leave number the same; Remove troops there now; N o o p i n i o n .

Date

09/26/83 10/23/83 10/25/83 10/27/83 11/07/83 12/13/83 01 /03/84 01 /04/84 01/17/84

Send More

7 21 31 17 13 9 5 8 7

Leave Same

48 21 26 36 41 38 30 29 31

Remove

40 48 39 42 39 48 59 57 58

No Opinion

5 10 5 5 7 5 6 6 4

160

Copyrights

Results f r o m A B C N e w s P o l l c o m p i l e d w i t h the aid o f B. I . Page and R. Y. Shapiro and (heir colleagues at the National O p i n i o n Research Center. This question was asked to registered voters w h o identify themselves either as Democrats o r as independents w h o lean t o w a r d the Democrats: Imagine your state holds a Democratic p r i m a r y and these are the candidates: Reuben Askew, A l a n Cranston, John Glenn, Gary Hart, Ernest H o l l i n g s , Jesse Jackson, George VtcGovem, and Walter Mondale. Whether you are a Democrat o r not, f o r w h o m w o u l d y o u vote: A s k e w , C r a n s t o n , G l e n n , H a r t , H o l l i n g s , Jackson, M c G o v e r n , or Mondale? (slight variation o f w o r d i n g starting September 2 6 , 1983). T h e responses for M o n d a l e , G l e n n , and N o O p i n i o n were tabulated separately. A l l other opinions were pooled and included volunteered responses for other m i n o r candidates and for those w h o said they w o u l d not vote. T h i s last category was 1-2 percent i n all polls. Table BJ.

Date

06/19/83 08/01/83 09/26/83 11/07/83 12/13/83 01/15/84 01/17/84 02/15/84

ABC

News

Poll on the Democratic

Mondale

42 43 36 47 44 51

45 55

Glenn

primary.

Others

28 28 26 21 23 11 22 13

24 23 26 19 24 19 27 21

No Opinion

6 6

11 9 7 18

5 9

161

Copyrighte

Table B.4. ABC News Poll on the economic climate. Results f r o m A B C N e w s P o l l c o m p i l e d w i t h the aid o f B . I . Page and R. Y . Shapiro and their colleagues at the N a t i o n a l O p i n i o n Research Center. The question was: D o y o u t h i n k the nation's economy is: Getting better; Getting worse; Staying the same; N o o p i n i o n .

Date

03/06/81 05/20/8 09/20/81 10/18/81 11/22781 12/12/82 01/30/82 03/08/82 04/25/82 08/17/82 09/13/82 10/11/82 12/18/82 01/22/83 03/02/83 04/12/83 05/15/83 06/19/83 08/01/83 09/26/83 11/07/83 12/13/83 01/17/84

Better

9 14 12 17 11 12 17 13 21 17 21 21 20 18 39 37 43 36 50 44 44 46 49

Same

Worse

36 36 44 41 22 32 31 27 30 31 33 28 26 36 39 40 39 42 30 35 36 31 31

54 49 42 40 55 54 50 59 47 50 45 48 52 46 21 21 17 20 19 20 20 20 19

No Opinion

2 1 2 2 1 2 2 1 2 2 1 3 1 1 1 2 1 2 0 1 1 2 1

162

Copyrighte

163

Table B.5. NBC News Poll on the importance of unemployment versus inflation. Results from NBC News Poll compiled with the aid of B. I . Page and R. Y. Shapiro and their colleagues at the National Opinion Research Center. The question was: In your opinion which is the more important problem facing the country today--finding jobs for people who are unemployed or holding down inflation? The responses were: Finding jobs; Both equal; Holding down inflation; Not sure. Date 03/22/77 04/26/77 08/03/77 03/22/78 05/02/78 06/28/78 08/08/78 09/02/78 11/14/78 12/12/78 03/20/79 09/11/79 05/29/80 07/09/80 08/06/80 08/23/80

Unemployment 43 41 50 39 32 33 28 27 22 22 23 21 30 30 5 26

Equal 18 14 11 10 9 10 11 9 8 9 11 10 15 15 14 20

Inflation 37 43 36 49 56 55 59 61 69 68 64 67 52 53 48 53

Not Sure 2 2 3 2 3 2 2 3 1 1 2 2 3 2 3 1

Results f r o m (he p o l l i n g organizations indicated. A l l data f r o m the Roper Center for Public O p i n i o n Research and C B S News, 524 W . 57th St., New Y o r k , N Y 10019. The question wordings by poll number are: 1. (President Reagan has taken a number o f steps i n Central A m e r i c a to meet what he says is the mounting supply o f arms f r o m Russia and Cuba going to left-wing rebel forces i n E l Salvador and to the Sand in is ta government in Nicaragua.) Let me ask y o u i f you favor o r oppose arming and supporting the rebels i n N i c a r a g u a w h o are t r y i n g to o v e r t h r o w the Sandinista government in that country? Favor; Oppose; N o t sure. 2. D o y o u favor or oppose the U.S. a r m i n g and supporting the rebels i n Nicaragua w h o are t r y i n g to overthrow the Sandinista government i n that country? Favor; Oppose; N o t sure. 3. ( N o w let me read you some statements about President Reagan's handling o f foreign affairs. For each, tell me i f you agree or disagree.) (InterviewerRotate Question Order)...It is w r o n g for the C I A (Central Intelligence A g e n c y ) to help finance the anti-Sandinista forces i n Nicaragua? Disagree; Agree; N o t sure. 4 . D o you favor or oppose...the U.S. ( U n i t e d States) a r m i n g and supporting the rebels i n Nicaragua, w h o are t r y i n g to o v e r t h r o w the Sandinista government i n that country? Favor; Oppose; N o t sure. 5. Should the U n i t e d States be g i v i n g assistance to the g u e r r i l l a forces n o w opposing the M a r x i s t government i n Nicaragua? Yes; N o ; D o n ' t k n o w . 6. President Reagan recently asked Congress to authorize $100 m i l l i o n i n U.S. a i d to the rebels seeking to o v e r t h r o w the c o m m u n i s t g o v e r n m e n t i n Nicaragua, including $70 m i l l i o n for m i l i t a r y purposes and $30 m i l l i o n for non-military purposes, such as food and medical supplies. D o y o u think the Congress should o r should not authorize this new a i d package? S h o u l d authorize (includes 2 percent volunteering should authorize n o n - m i l i t a r y o n l y ) ; Shouldn't authorize; N o o p i n i o n . 7. The House o f Representatives has refused Reagan's request for 100 m i l l i o n dollars i n m i l i t a r y and other aid to the contra rebels i n Nicaragua. D o y o u approve or disapprove o f that action by the House? Disapprove; A p p r o v e ; Don't know/No o p i n i o n . 8. D o you favor or oppose the U.S. sending $100 m i l l i o n i n m i l i t a r y and n o n military aid to the Contra rebels i n Nicaragua? Favor; Oppose; N o t sure. 9. D o you think the U.S. government should g i v e $100 m i l l i o n i n m i l i t a r y and other a i d to the Contras t r y i n g to o v e r t h r o w the g o v e r n m e n t i n Nicaragua? Yes, should; N o , shouldn't; N o o p i n i o n . Table B.6.

Polls on the desirability

Poll Number and Source

1 2 3 4 5 6 7 8 9

HARRIS HARRIS HARRIS HARRIS GALLUP GALLUP ABC HARRIS CBS

of sending Contra

aid.

Percent Response Dale Favor

08/20/83 09/12/83 07/10/84 03/04/85 08/28/85 03/07/86 03/22/86 04/07/86 04/08/86

66 60 55 53 58 52 60 62 62

Oppose

23 24 32 36 29 37 35 33 25

Don't K n o w

11 16 13 11 13 11 4 5 13

Copyrights

165 Appendix C

Summaries of Text Analyses C.1 STRATEGY FOR CONTENT ANALYSIS BY SUCCESSIVE FILTRATIONS Text was first processed through a series of "filter" program runs to remove irrelevant material. Finally the remaining, fairly homogeneous text was scored for its support of each of the polled positions. The outline of these steps is provided in Chapter 3. This appendix summarizes the dictionaries and rules used for the six analyses in this book. All dispatches were given infon content scores corresponding to the positions for which poll data were available (data in Appendix B). To illustrate, a single dispatch is followed in detail through all the analytic steps for the defense spending example. C.2 TEXT ANALYSIS FOR DEFENSE SPENDING--INCLUDING DETAILED EXAMPLE 1. Filtration to select for dispatches on American defense spending The first step was a filtration to discard all dispatches not directly relevant to American defense spending. The entire text was marked for words referring to America (denoted by { A } ) , defense (denoted by { D } ) , and spending (denoted by {S}). Articles with all three word classes were retained for further analysis unless they also had the word "aid" or "fund," which led to the story being rejected. The following actual dispatch, dated February 19, 1983, was kept since it had many America, defense, and spending words and neither of the prohibited words "aid" or "fund": SECTION: Washington Dateline LENGTH: 576 words BYLINE: By MAUREEN SANTINI, Associated Press Writer DATELINE: WASHINGTON KEYWORD: { A } Reagan-{D}Defense

166

Appendix

C

President { A J R e a g a n , i n v o k i n g the menace o f A d o l f H i t l e r , asked { A } C o n g r e s s on Saturday to suppress the urge to reduce his " m i n i m a l " 1984 {DJmilitary {S}budget I n his weekly radio broadcast f r o m the W h i t e { A } H o u s e , the president said his $238.6 b i l l i o n { D j d e f e n s e {SJspending proposal for fiscal year 1984, w h i c h begins O c t . 1, was necessary "unless we're w i l l i n g t o g a m b l e w i t h o u r immediate security and pass on to future generations the legacy o f neglect we inherited." "That k i n d o f neglect w o u l d o n l y weaken peace and stability i n the w o r l d , both now and i n the years ahead." { A J R e a g a n s a i d , " N o w , I k n o w this is a h a r d t i m e to c a l l f o r increased { D } d e f e n s e {SJspending. I t isn't easy t o ask { A } A m e r i c a n families w h o are already m a k i n g sacrifices in the recession ..." " O n the other hand, it's always very easy a n d very t e m p t i n g p o l i t i c a l l y to come up w i t h arguments f o r neglecting { D j d e f e n s e {SJspending i n t i m e o f peace," the president said. " O n e o f the great tragedies o f this century was that i t was o n l y after the balance o f power was allowed to erode and a ruthless adversary, A d o l f H i t l e r , deliberately w e i g h e d the risks and decided to strike that the importance o f a strong {Djdefense was realized too late." Though {AJReagan called for an overall freeze on domestic {SJspending i n his 1984 {SJbudget, the {D}defense portion increased by 14 percent A n d that was after the president cut $8 b i l l i o n f r o m the Pentagon { S Jrequest before submitting i t to {AJCongress. { A J R e a g a n said he and his administration had "agonized" over the current { D j d e f e n s e { S J b u d g e t by t r i m m i n g { S j r e q u e s t s and c u t t i n g non-essential programs. " T h e { D j d e f e n s e {SJbudget we f i n a l l y presented is a m i n i m a l {SJbudget t o protect our country's v i t a l interests and meet o u r commitments," he said. The president said i t was "far better to prevent a crisis than to have to face i t unprepared at the last m o m e n t That's w h y w e have an o v e r r i d i n g m o r a l obligation to invest n o w , this year, i n this {SJbudget, i n restoring { A J America's strength to keep the peace and preserve our freedom." H e said the Soviet U n i o n outspends the { A } U n i t e d States o n ... "... fits and starts," he said, " w e w i l l never convince the Soviets that it's i n their interests t o behave w i t h restraint and negotiate genuine { D J a r m s reductions. W e w i l l also burden the { A J A m e r i c a n taxpayer t i m e and again w i t h the h i g h {SJcost o f crash rearmament" "Sooner o r later, the bills f a l l due." { A J S e n a t e M i n o r i t y Leader Robert C. B y r d o f W e s t V i r g i n i a gave the Democratic response to Reagan's comments and took issue w i t h the president's contention that c u t t i n g the administration { D j d e f e n s e {SJbudget {SJrequest w o u l d expose the country t o danger. " F o r example, we d o not need t w o new manned { D } b o m b e r s - one o f w h i c h w i l l be obsolete almost immediately after i t is b u i l t , " B y r d said, referring t o the B - l { D J b o m b e r under construction and the advanced Stealth plane expected to emerge f r o m development late in this decade. A r g u i n g that the national { D j d e f e n s e depends on a strong economy, B y r d stressed the need for greater... Ellipses (...) i n the text indicate that the remainder o f a sentence was n o t retrieved due to the text being further than fifty words f r o m one o f the seven search

Copyrighte

Summaries

of Text Analysis

167

words: D E F E N S E , M I L I T A R Y , A R M S , S P E N D ! , E X P E N D I T U R E , F U N D ! , or B U D G E T ! (see Appendix B ) . After this filtration, the total number o f characters o f text dropped f r o m 820,000 to 600,000. The average number o f words per dispatch increased f r o m 148 t o 199. T h e dispatch n u m b e r dropped more dramatically ( f r o m 692 to 377) than the number o f characters o f text because very little was retrieved f r o m irrelevant dispatches. The c o l l e c t i o n was stopped as soon as a story was seen t o be not pertinent d u r i n g the retrieval f r o m the Nexis data base. T h e increase i n average w o r d c o u n t per dispatch was a natural consequence o f discarding dispatches f r o m w h i c h very few words were collected.

2.

Filtration to select for paragraphs on defense spending

The second text analysis step selected only paragraphs directly discussing defense spending. The c o n d i t i o n was that a defense w o r d (denoted by { D } ) be close t o a spending w o r d (denoted by { S } ) . The paragraphs f r o m the dispatch given above were scored using this rule. T h e decision for each paragraph is g i v e n directly below the paragraph: President Reagan, i n v o k i n g the menace o f A d o l f H i t l e r , asked Congress o n Saturday to suppress the urge to {SJreduce his " m i n i m a l " 1984 { D J m i l i t a r y {SJbudget. A B O V E P A R A G R A P H W A S KEPT. I n his w e e k l y radio broadcast f r o m the W h i t e House, the president said his $238.6 b i l l i o n { D j d e f e n s e {SJspending proposal for fiscal year 1984, w h i c h begins O c t l , was necessary "unless we're w i l l i n g to g a m b l e w i t h o u r immediate security and pass o n to future generations the legacy o f neglect we inherited." A B O V E P A R A G R A P H W A S KEPT. "That k i n d o f neglect w o u l d o n l y weaken peace and stability i n the w o r l d , both n o w and i n the years ahead." ABOVE PARAGRAPH WAS DISCARDED. Reagan said, " N o w , I k n o w this is a hard t i m e t o c a l l f o r {SJincreased { D j d e f e n s e {SJspending. I t isn't easy to ask American families w h o are already making sacrifices i n the recession ..." A B O V E P A R A G R A P H W A S KEPT. " O n the other hand, it's always very easy and very t e m p t i n g p o l i t i c a l l y to come up w i t h arguments for neglecting { D j d e f e n s e { S J s p e n d i n g i n t i m e o f peace," the president said. A B O V E P A R A G R A P H W A S KEPT. "One o f the great tragedies o f this century was that i t was o n l y after the {SJbalance o f power was allowed to erode and a ruthless adversary, A d o l f Hitler, deliberately weighed the risks and decided to strike that the importance o f a strong {Djdefense was realized too late." ABOVE PARAGRAPH WAS DISCARDED. T h o u g h Reagan called for an overall freeze on domestic {SJspending i n his 1984 {SJbudget, the {Djdefense portion {SJincreased b y 14 {SJpercent. A n d that was after the p r e s i d e n t s J cut $8 b i l l i o n f r o m the {DJPentagon { S Jrequest before submitting i t to Congress. A B O V E P A R A G R A P H W A S KEPT. Reagan said he a n d his a d m i n i s t r a t i o n had " a g o n i z e d " o v e r the current { D j d e f e n s e {SJbudget by t r i m m i n g {Sjrequests and { S J c u t t i n g non-essential programs.

Copyrights

168

Appendix

C

A B O V E P A R A G R A P H W A S KEPT. "The { D j d e f e n s e {SJbudget we f i n a l l y presented is a m i n i m a l ( S ) b u d g e t to protect our country's v i t a l interests and meet our commitments," he said. A B O V E P A R A G R A P H W A S KEPT. The president said i t was far better to prevent a crisis than to have to face i t unprepared at the last moment. T h a t ' s w h y we have an o v e r r i d i n g m o r a l o b l i g a t i o n to invest n o w , this year i n this {SJbudget, i n restoring America's {SJstrength to keep the peace and preserve our freedom," ABOVE PARAGRAPH WAS DISCARDED. He said the Soviet U n i o n outspends the United States on ... A B O V E PARAGRAPH W A S DISCARDED. "... fits and starts," he said, " w e w i l l never convince the Soviets that it's i n their interests t o behave w i t h restraint and negotiate g e n u i n e { D } arms {S J reductions. W e w i l l also burden the American taxpayer time and again w i t h the high { S } cost o f crash rearmament." A B O V E P A R A G R A P H W A S KEPT. "Sooner or later, the bills f a l l due." ABOVE PARAGRAPH WAS DISCARDED. Senate M i n o r i t y Leader Robert C. B y r d o f West V i r g i n i a gave the Democratic response to Reagan's comments and took issue w i t h the president's contention that { S J c u t t i n g the administration { D j d e f e n s e {SJbudget {SJrequest w o u l d expose the country to danger. A B O V E P A R A G R A P H W A S KEPT. " F o r example, we do not need t w o new manned { D J b o m b e r s , one o f w h i c h w i l l be obsolete almost immediately after i t is b u i l t , " B y r d said, referring to the B - l {DJbomber under construction and the advanced Stealth {SJplane expected to emerge f r o m development late i n this decade. A B O V E PARAGRAPH WAS DISCARDED. A r g u i n g that the national { D j d e f e n s e depends o n a strong economy, B y r d stressed the need for greater... ABOVE PARAGRAPH WAS DISCARDED. This sample dispatch was chosen because it illustrates most o f the features o f the text analysis. I n consequence, this story was one o f the most complex found and was somewhat atypical i n c o n t a i n i n g a substantial amount o f i n f o r m a t i o n i n d i r e c t l y relevant to the topic o f defense spending. M o r e frequendy, the discarded text was o n a topic other than defense spending. I n press conferences, for instance, the shifts i n topic c o u l d be quite a b r u p t Nevertheless, the relevant thoughts i n the discarded text, even i n the above example, were almost always also f o u n d i n the retained t e x t For example, the first discarded paragraph was an expansion on the point i n the previous, retained paragraph rather than being a new idea. Also, the next to the last discarded paragraph o n l y illustrated the point i n the previous, retained paragraph. A l t h o u g h a s m a l l a m o u n t o f relevant i n f o r m a t i o n may have been lost by discarding the paragraphs w i t h pertinent information w h i c h was i n d i r e c t the gain was the immense s i m p l i f i c a t i o n o f the subsequent analysis, w i t h the total text f r o m a l l dispatches being reduced from 600.000 characters to 220,000.

3.

Numerical scoring for three positions o n defense spending

The paragraphs retained f r o m the second filtration described above were then scored for favoring more, same, o r less defense spending. Since the second filtration

Copyrights

Summaries

of Text A nalysis

16V

had already guaranteed that a defense w o r d was close to a spending w o r d , the scoring o n l y depended on a defense w o r d (denoted by { D } ) being close to modifiers i m p l y i n g these three positions. T h e modifiers fell into the three classes favoring more (denoted by { M J ) , same (denoted b y { S } ) , and less (denoted b y { L } ) - - w i t h a "less" w o r d close to a " m o r e " w o r d being equivalent t o a "same" w o r d and w i t h a less w o r d close t o another less w o r d also being equivalent t o a same w o r d . I n some combinations, w o r d order and p r o x i m i t y were also i m p o r t a n t I n addition, the prefix " n o n " (denoted by { n } ) preceding a defense word meant that the defense w o r d was not considered t o be relevant t o the military. A l l paragraphs had a total score o f 1.0 w i t h each cluster o f modifier words close t o a defense w o r d contributing t o the final score. I f a paragraph o n l y had one such cluster, the entire paragraph score o f 1.0 was assigned t o the appropriate position. W h e n there was more than one cluster, the score o f 1.0 was d i v i d e d i n t o equal fractions w i t h each cluster receiving one part. This scoring procedure is illustrated using the retained paragraphs o f the dispatch considered above. T h e score for each paragraph is given immediately f o l l o w i n g die paragraph. The scores were for whether the paragraph favored more, same, and/or less defense spending. Comments f o l l o w i n g the scores explain w h y the computer arrived at the decisions. President Reagan, i n v o k i n g the menace o f A d o l f H i t l e r , asked Congress o n Saturday t o {LJsuppress the urge t o { L } r e d u c e his " m i n i m a l " 1984 { D J m i l i t a r y budget. S C O R E F A V O R I N G : More=0.00 Same-1.00 Less=0.00 The "suppression" o f a " r e d u c t i o n " i m p l i e d f a v o r i n g the same level f o r the " m i l i t a r y " b u d g e t The three words were treated as a cluster because they were close to each other. I n his w e e k l y radio broadcast f r o m the W h i t e House, the president said his $238.6 b i l l i o n { D j d e f e n s e spending proposal for fiscal year 1984, w h i c h begins Oct. 1, was {MJnecessary unless we're w i l l i n g t o gamble w i t h o u r immediate security a n d pass o n t o future generations the legacy o f { M J n e g l e c t w e inherited." S C O R E F A V O R I N G : More=0.00 Same=0.00 Less-0.00 N o score here. "Necessary" and n e g l e c t " w h i c h i m p l i e d more spending, were Kx> far away f r o m "defense." Reagan said, " N o w , I k n o w this is a hard time t o c a l l f o r { M } i n c r e a s e d { D j d e f e n s e spending. I t { L J i s n ' t easy to ask American families w h o are already m a k i n g sacrifices i n the recession ..." S C O R E F A V O R I N G : M o r e - 1 . 0 0 Same=0.00 Less=0.00 The operative w o r d c o m b i n a t i o n was "increased" "defense." T h e w o r d " i s n ' t " only changed the sense o f words like "increased" i f i t preceded them. " O n the other hand, it's always very easy a n d very t e m p t i n g p o l i t i c a l l y t o come { M J u p w i t h arguments f o r { M J n e g l e c t i n g { D j d e f e n s e spending i n time o f peace," the president said. S C O R E F A V O R I N G : M o r e - 1 . 0 0 Same=0.00 Less=0.00 T h e scored c o m b i n a t i o n was " u p " ... " n e g l e c t i n g " "defense". " U p " and " n e g l e c t i n g " were scored together as meaning more s h o u l d be spent. T h e inclusion o f words like "neglect" and "inadequate" i n the dictionary d i d permit the public to reason and thereby take indirect information into a c c o u n t These words were included because they were usually found i n the context o f arguments that defense spending should have been increased i f it was neglected or inadequate. T h o u g h Reagan called f o r an overall {SJfreeze o n domestic spending i n his 1984 budget, the {Djdefense portion {MJincreased b y 14 percent A n d that was

Copyrightec

770 Appendix

C

after the president { L } c u t $8 b i l l i o n f r o m the { D } P e n t a g o n request before s u b m i t t i n g i t to Congress. S C O R E F A V O R I N G : More=0.50 Same=0.00 Less=0.50 T h i s paragraph was scored as m a k i n g t w o different statements, one f a v o r i n g more defense spending ("defense" "increased") and one f a v o r i n g less ( " c u t " "Pentagon"). Therefore, the paragraph score o f 1.0 was d i v i d e d i n t w o . "Freeze" was too far away f r o m "defense" to have a connotation for defense, as was consistent w i t h the actual meaning o f the paragraph. Reagan said he and his a d m i n i s t r a t i o n had " a g o n i z e d " o v e r the c u r r e n t { D j d e f e n s e budget by { L J t r i m m i n g requests and { L J c u t t i n g { n } non-essential programs. S C O R E F A V O R I N G : M o r e - 0 . 0 0 Same-1.00 Less-0.00 The reasonable score favoring unchanged military spending was serendipitous. T h i s score was due to " t r i m m i n g ' ' and " c u t t i n g " being equivalent to the concept o f same spending. This combination close t o "defense" gave the score favoring same spending. T h e " n o n " d i d not have a function here. I f the w o r d "defense" occurred i n place of the w o r d "essential," then the concept o f defense w o u l d have been n u l l i f i e d , indicating that the topic was not about defense. "The { D j d e f e n s e budget we f i n a l l y presented is a m i n i m a l budget to protect o u r country's vital interests and meet our commitments," he said. S C O R E F A V O R I N G : M o r e - 0 . 0 0 Same=0.00 Less-0.00 T h i s paragraph had no score since "defense" was close to no modifier words. I n fact, when this paragraph was read by itself, i t was consistent w i t h any o f the p o s i t i o n s . T h e actual i n f o r m a t i o n f a v o r i n g one p o s i t i o n o r another was elsewhere i n the text. "... fits and starts," he said, " w e w i l l never convince the Soviets that it's i n their interests to behave w i t h restraint and negotiate genuine arms {LJreductions. W e w i l l also burden the A m e r i c a n taxpayer time and again w i t h the { M J h i g h cost o f crash rearmament." S C O R E F A V O R I N G : M o r e - 0 . 0 0 Same=0.00 Less-0.00 T h i s paragraph also had n o score since there were no words d i r e c t l y c o n n o t i n g defense. I t c o u l d be argued that i t favored more defense spending i n d i r e c t l y . However, the statement is probably weaker than those above speaking d i r e c d y t o the issue. Senate M i n o r i t y Leader Robert C. B y r d o f West V i r g i n i a gave the Democratic response to Reagan's comments and took issue w i t h the president's contention that { L J c u t t i n g the administration { D j d e f e n s e budget request w o u l d expose the country to {LJdanger. S C O R E F A V O R I N G : M o r e - 0 . 0 0 Same-0.00 Less-1.00 The score o f favoring less spending came f r o m " c u t t i n g " ... "defense." "Danger" was too far f r o m "defense" to be scored. The score was p r o b a b l y correct, although a sounder basis for the conclusion w o u l d have included: " t o o k issue"..."cutting"..."defense"., "danger." The final score for this dispatch was 2.5 paragraphs favoring more, 2.0 favoring same, and 1.5 favoring less spending. T h i s sample dispatch was one o f the most complex retrieved. Six o f the nine paragraphs were scored for supporting one o f the three positions. For comparison, the average number o f relevant paragraphs was only 1.7 among all dispatches w i t h at least one paragraph w i t h a positive score. Since the average A P paragraph had approximately thirty words, the final scoring came f r o m approximately fifty words per story although approximately eighty words w e r e e x a m i n e d in e a c h s c o r e d d i s p a t c h .

Copyrights

Sunwiaries

of Text Analysis

171

The difference between the fifty and eighty words meant that 30-40 percent o f the paragraphs had n o score. T h i s was true for three o f the nine paragraphs i n the dispatch just examined. The scoring for defense spending came f r o m paragraphs representing 5-10 percent o f the words (400-900) i n the average dispatch. Only 20 percent o f the dispatches had fewer words and 10 percent had more. Being f a i r l y l o n g this dispatch also illustrated most o f the scoring features. I n fact, the problems were more severe f o r this text than for most others. The more typical dispatches had smaller numbers o f relevant paragraphs and usually made their points about defense spending quite directly before proceeding to other topics. There tended to be fewer crosscurrents to complicate the scoring. The most appropriate base for considering the scoring is the 377 dispatches retained after the first filtration step. The others were not about defense spending or were about n o n - A m e r i c a n forces. O f these 377, 72 percent were used i n the final scoring. t

4.

Numerical scoring for t w o positions on defense spending

As an alternative to the evaluation j u s t described, the text was also scored to favor o n l y t w o positions—more or less defense spending. T o do so, the concept o f same spending was eliminated. As a result, some modifier words like " m a i n t a i n " and "keep" were omitted f r o m the dictionary. These words were previously interpreted to favor the concept o f same spending. Other words, like "freeze" and "frozen." were moved f r o m the same spending class to the modifier class connoting less spending. N o w , a "less" w o r d preceding a " m o r e " w o r d was assigned t o favor less instead o f same spending (e.g., "cut...increase"). S i m i l a r l y , t w o nearby less w o r d s (e.g., "cut.areduction") were also assigned to favor more instead of same. Other dictionary changes included the deletion o f a few words favoring more ("bolster") and less C alternate," "weaken," " w i t h o u t " ) spending. T h e words "nuclear" and "arms" were added t o the list o f words referring to defense. Thus "nuclear arms reduction talks" was interpreted to support less defense spending w h i l e this phrase was simply ignored in the previous scoring. Using this alternate dictionary and its associated rules, the text scored i n the preceding section was rescored. Those paragraphs with changed final scores are listed below w i t h comments: President Reagan, i n v o k i n g the menace o f A d o l f H i t l e r , asked Congress o n Saturday t o {L}suppress the urge to { L } r e d u c e his " m i n i m a l " 1984 { D J m i l i t a r y budget. SCORE F A V O R I N G : More»1.00 Less-0.00 The "suppress" ... "reduce" was interpreted previously to favor same spending instead of more spending.

Reagan said he and his a d m i n i s t r a t i o n had " a g o n i z e d " o v e r the c u r r e n t { D j d e f e n s e budget by { L J t r i m m i n g requests and { L J c u t t i n g { n j non-essential programs. SCORE F A V O R I N G : M a r e - 1 . 0 0 Less-0.00 The " t r i m m i n g " ... " c u t t i n g " was misscored previously to favor same spending and was misscored this time to favor more spending. A g a i n , the w r o n g score was not entirely inconsistent w i t h the sense o f the paragraph. "... fits and starts," he said, " w e w i l l never convince the Soviets that it's i n their interests t o behave w i t h restraint and negotiate g e n u i n e { D J arms

Copyrighte

/ 72

Appendix

C

{LJreductions. We w i l l also burden the American taxpayer t i m e and again w i t h the { M j h i g h cost o f crash rearmament'' S C O R E F A V O R I N G : M o r e - 0 . 0 0 Less-1.00 Previously, this paragraph had a zero score since " a r m s " was not i n the dictionary i n order to o m i t reference t o arms reduction. Here, ''arms" " r e d u c t i o n " was interpreted to suggest that defense spending, likewise, should be diminished. W i t h these m o d i f i c a t i o n s , the new score was 4.5 paragraphs f a v o r i n g more spending and 2.5 favoring less. The text recoveries d u r i n g the scoring are presented in Table C . l .

5.

Text analysis for defense waste and fraud a.

Filtration

to remove

dispatches

not on American

defense

spending.

The

first step discarded dispatches i f they were not about American waste and fraud. The U.S. was usually not the focus when there was a mention o f a n o n - A m e r i c a n region i n the heading portion prepared by the A P and listed before the body o f a dispatch. T h e heading t y p i c a l l y contained most o f these designators: dateline, headline, k e y w o r d , and section (see example at the beginning o f Section C.2). Therefore, the filtration c o m m a n d s i m p l y removed stories w i t h one o f these designators f o l l o w e d closely by a w o r d referring to a foreign part o f the w o r l d . Dispatches referring to defense against waste and fraud for non-defense topics were also not retained f o r further study i f the stories mentioned other key words such as "hazardous" and " t o x i c " referring to non-military waste. b. Filtration to select paragraphs on defense waste and fraud. This filtration was accomplished by l o o k i n g for w o r d combinations referring to both the defense industry and to waste. S o m e c o m b i n a t i o n s w e r e s i m p l e , s u c h as "overcharge"..."weapons." O t h e r c o m b i n a t i o n s were m o r e c o m p l e x , such as "defense"..."contractor"..."cut corners." C. Numerical scoring for stories on defense waste and fraud. Any word combination suggesting defense waste such as those i n the preceding subsection led to the A P paragraph containing the combination t o be scored as favoring less defense spending. The recovery data for this waste and fraud analysis are given i n Table C.2.

C3

T E X T A N A L Y S I S FOR T R O O P S I N 1.

LEBANON

Filtration to select for paragraphs o n American troops i n Lebanon

The first filtration selected paragraphs containing words referring to America, troops, and Lebanon. A t this step, a mention o f policy or a synonym was considered to be equivalent to troops since p o l i c y often referred to troops. Paragraphs were discarded i f they had w o r d s r e f e r r i n g to n o n - A m e r i c a n troops (e.g., " A r a b , " " C h r i s t i a n , " " D r u s e , " " S y r i a n , " " I s r a e l i " ) , a non-Lebanon region o f the w o r l d (e.g., "Grenada"), or non-military activities (e.g„ "economy"). I n this analysis, paragraphs were considered to be about A m e r i c a or Lebanon after a previous m e n t i o n o f words indicating these geographic areas unless there was a w o r d (e.g., " C h r i s t i a n , " " S y r i a n , " "France") indicating non-American troops or a non-Lebanese location (e.g., "Israel"). Also, pronouns such as " t h e y " and " t h e m " were taken to refer to troops i f there was a mention o f troops in the previous paragraph.

Copyrights

Summaries of Text Analysis 2.

Filtration to remove entertainment

paragraphs o n m i l i t a r y

¡73

action and Christmas

This step removed a l l text o n actual combat and a l l paragraphs o n entertainer B o b Hope's Christmas visit to Lebanon.

3.

Numerical scoring for dispatches o n troops in Lebanon

The final scoring step used the major criterion that a w o r d referring to troops o r p o l i c y should be near m o d i f i e r words favoring more, same, o r less, although some words—such as "stay" and " w i t h d r a w " - - w e r e able by themselves t o favor keeping o r removing troops. Therefore, the paragraphs had i n f o n content scores favoring more, same, o r less troops. The recovery data for the analyses are given i n Tabic C.3.

C,4

T E X T A N A L Y S I S FOR T H E D E M O C R A T I C 1,

PRIMARY

Analysis using bandwagon words

a. Filtration for paragraphs about candidates. First, paragraphs were selected and kept o n l y i f they had the name o f at least one o f the Democratic candidates appearing i n the A B C News Poll o f Table b. Scoring using bandwagon words. T h e n the text was scored f o r being either favorable o r detrimental t o John Glenn, Walter Mandate, o r Others (Reuben As t e w , A l a n Cranston, Gary H a r t , Ernest H o l l i n g s , Jesse Jackson, o r George M c G o v e r n ) . These scores depended o n modifier words i m p l y i n g success or failure being close to a candidate name. T h e scores belonged to six positions, three favorable t o Mondale, Glenn, and Others, and three unfavorable to these candidates. The recovery data are i n Table C-4.

2,

Analysis using name count

The paragraphs i n the o r i g i n a l retrievals were scored w i t h o u t any further filtration steps. Every paragraph was stilt given a total score o f 1.0. I f o n l y one candidate was mentioned i n the paragraph, then the score i n favor o f that candidate was 1,0. I f several candidates were discussed, then that score o f 1.0 was shared among the candidates. This type o f scoring o b v i o u s l y d i d not generate any scores unfavorable to a candidate- Therefore, o n l y three types o f paragraph scores w e i e obtained, those mentioning Mandate, Glenn, and Others (recovery data in Table CA).

C.5

T E X T A N A L Y S I S FOR T H E E C O N O M I C I.

CLIMATE

Filtration to eliminate dispatches o n non-American economies

T h e f i r s t s t e p discarded dispatches i f they were not about the U n i t e d States. T h e procedure was very s i m i l a r t o the one described earlier w h i c h looked f o r n o n - U , 5 . words in the dispatch heading region.

MaTcpnan, samuLUCHHbw aerope*MI* npasoM

¡74

Appendix 2.

C

Filtration to select paragraphs discussing the economy

The next filtration step selected paragraphs w i t h at least one w o r d referring t o some aspect o f the economy. The reference d i d not have to be to the economy as a whole but c o u l d include components such as " a g r i c u l t u r e . " A l s o p e r m i t t e d were words like " r a l l y " describing economic performance.

3.

Numerical scoring

I n the f i n a l step, i n f o n c o n t e n t scores were assigned f r o m single w o r d s suggesting better, same, or worse i n the context o f economic conditions. Therefore, the dictionary had qualifiers like "best," "confusion," and " b a d " d i v i d e d i n t o classes f a v o r i n g better, same, and worse. A d d i t i o n a l dictionary words included those (e.g., " n o t " and " d i f f i c u l t " ) w h i c h could alter the sense o f the q u a l i f i e r words. Since the previous filtration had already guaranteed that each paragraph had to make reference to an aspect o f the economy, the score c o u l d be determined by single words suggesting better, same, o r worse. T h e recovery data for all steps are given i n Table C.5.

C.6

T E X T A N A L Y S I S FOR U N E M P L O Y M E N T V E R S U S 1.

INFLATION

Filtration to eliminate dispatches on non-American economies

T h i s step to remove dispatches o n foreign countries was l i k e the first n i t r a t i o n s for dispatches on defense waste and fraud and the economic climate.

2.

Numerical scoring

T h e scoring was f o r : u n e m p l o y m e n t more i m p o r t a n t , equal i m p o r t a n c e , or inflation more important. The m a i n c r i t e r i o n was that inflation, unemployment, or their synonyms should be close t o m o d i f i e r words indicating that the p r o b l e m was i m p o r t a n t . A s i g n i f i c a n t n u m b e r o f paragraphs spoke o f b o t h problems being i m p o r t a n t . I n r e c o g n i t i o n o f this fact, the score was for equal i m p o r t a n c e i f a m o d i f i e r w o r d made i n f l a t i o n i m p o r t a n t and i f an u n e m p l o y m e n t w o r d f o l l o w e d shortly after, as i n the phrase " w e must combat both i n f l a t i o n and u n e m p l o y m e n t ' * S i m i l a r l y , i f a paragraph had one w o r d cluster supporting Ihe importance o f each o f the t w o topics, the problems were considered to be equally crucial (recovery data are i n Table C.6).

C.7

T E X T A N A L Y S I S FOR C O N T R A A I D 1.

Filtration to select paragraphs o n Contra aid

Each paragraph retained i n this filtration step was required t o contain words i m p l y i n g Nicaragua, the U n i t e d States, and funding. For this filtration step i t d i d not matter i f the reference to Nicaragua was to the Contras opposing the government or to the government side. A mention o f Nicaragua meant that the next paragraph was also about Nicaragua unless there was discussion o f another C e n t r a l A m e r i c a n country like El Salvador or Honduras.

Copyrights

Summaries of Text Analysis 1.

175

Numerical scoring by Fan

The paragraphs Surviving the first filtration were scored by the author by l o o k i n g for m o d i f i e r words close to w o r d combinations discussing both die Contras and funding. Combinations o f modifier words were examined for whether they favored or opposed Contra aid. A n example o f a w o r d cluster favoring Contra a i d w o u l d be "approve"..."Contra",,,"aid," I f there were c o n d i t i o n a l w o r d s l i k e " i f " i n the paragraph, the paragraph was considered to favor both positions equally. Therefore, the final scores were for paragraphs either supporting or opposing Contra aid.

3,

Numerical Scoring by Simone French, Peter Miene, and Janet S w i m

The paragraphs scored by Fan were scored independenUy by these three graduate research assistants w o r k i n g as a team. Their scoring method was quite different f r o m that o f Fan, l o o k i n g at w o r d c o m b i n a t i o n s f a v o r i n g o r o p p o s i n g a i d w i t h o u t requiring that these words be close to words discussing C o n t r a a i d . T h e y also included many indirect pieces o f i n f o r m a t i o n that i m p l i e d a position o n Contra aid. For instance, the w o r d cluster "administration",./propaganda" by itself was scored as opposing Contra aid. This could safely be done because the paragraphs were already scored as being relevant to Contra aid by the initial filtration. The recovery data for both scoring methods are given i n Table C.7

Marepnan, 3awnmcHHbin a e r o p c s H M npnsoM

Table C.I. Summary of text analysis for defense spending. The upper portion gives the recoveries o f the text and paragraphs at different stages o f the text analysis. The Nexis search identified 9,314 dispatches, o f w h i c h 692 were retrieved at random. For the calculation, words are assumed to be approximately eight characters long. The lower p o r t i o n gives data for each position scored after the final step i n the upper p o r t i o n o f the table. T h e data f o r " a n y p o s i t i o n " refer to a l l p o s i t i o n s c o m b i n e d .

Step i n Analysis

Characters of Text No.

Dispatches

% Orig.

No.

8-Char. Words

% Orig.

per Dispatch

Nexis Retrieval

820,000

100

692

100

148

First F i l t e r

600,000

73

377

54

199

Second Filter

220,000

27

340

49

81

272 280

39 40

Scoring Runs: Scored to Favor More, Same, Less Scored to Favor M o r e and Less O n l y

Position Favoring

Average Paragraphs In

Total Dispatches

Dispatches W i t h at Least One Paragraph Favoring

W i t h A t Least One Paragraph Favoring T h i s Position

This

Position

Scored to Favor More, Same, Less: More 1.3 Same 1.1 Less 1.2 A n y position 1.7

177 66 132 272

Scored to Favor M o r e and Less O n l y : More 1.3 Less 1.3 A n y position 1.7

197 167 280

I

176

Copyrights

ud. Scoring o f all

Table C.2. Summary mentions of waste an<

S e e T a b l e C . 1 for

identified random.

Step i n

Dispatches

Analysis

No.

% Orig.

No.

8-Char. Words

% Orig.

per Dispatch

Nexis Retrieval

660,000

100

512

100

160

First F i l t e r

350,000

54

279

54

167

Second Filter

83,000

13

159

31

69

147

29

Scored to M e n t i o n Waste and Fraud

Position Favoring

Average Paragraphs I n Dispatches W i t h A t Least One Paragraph Favoring This Position

Less Due to Waste and Fraud

1.3

Total Dispatches W i t h A t Least One Paragraph Favoring T h i s Position

147

177

Copyrights

Table C J . Summary of text analysis for troops in Lebanon: scoring for more, same, and less troops. See Table C . l for explanation. T h e N e x i s search identified 1,517 dispatches, o f which 467 were retrieved at random.

Step i n Analysis

Characters of Text No. % Orig.

Disp; itches % Orig. No.

Nexis Retrieval

1,570,000

100

467

100

420

First F i l t e r

490,000

31

393

89

156

Second Filter

240,000

15

352

80

85

238

54

Scored to Favor M o r e , Same, and Less

Position Favoring

More Same Less A n y Position

Average Paragraphs I n Dispatches W i t h A t Least One Paragraph Favoring This Position

0.6 1.6 1.4 2.4

8-Char. Words per Dispatch

T o t a l Dispatches W i t h A t Least One Paragraph Favoring T h i s Position

36 197 172 238

178

Copyrighte

Table C.4. Summary of text analysis for Democratic primary: scoring for bandwagon words. See T a b l e d for explanation. T h e Nexis search identified 2,435

dispatches, o f w h i c h 425 were retrieved at random.

Step i n Analysis

Nexis Retrieval

Characters o f text No.

1,100,000

Bandwagon Analysis: First Filter 610,000 Scoring R u n

Dispatches

% Orig.

No.

% Orig.

per Dispatch

100

425

100

310

100

425 159

100 37

199

425

100

Name Count Analysis:

Position

8-Char. Words

Average Paragraphs In Dispatches W i t h A t Least One Paragraph Favoring This Position

Bandwagon Analysis: Pro Mondale Pro Glenn Pro Others Con Mondale C o n Glenn C o n Others A n y position

1.6 1.2 1.4 1.1 1.9 1.1 2.0

Name Count Analysis: M e n t i o n Mondale M e n t i o n Glenn M e n t i o n Others A n y position

2.6 2.0 4.2 5.6

T o t a l Dispatches W i t h A t Least One Paragraph Favoring This position

85 29 56 16 12 33 159

240 187 326 425

179

Copyrighted material

Table C.S. Summary of text analysis for economic climate: scoring for better, same, and worse. See Table C . l for explanation. The Nexis search identified 12,393

dispatches, of which 461 were retrieved at random.

Step i n

8-Char.-Words

.Analysis

Characters of text No % Orig.

Dispuitches % Orig. No.

Nexis Retrieval

730,000

100

461

100

197

First Filter

590,000

81

367

80

201

Second Filter

420,000

58

366

79

144

306

66

Scored to Favor Better, Same, and Worse

Position Favoring

Average Paragraphs I n Dispatches W i t h A t Least One Paragraph Favoring T h i s Position

per Dispatch

T o t a l Dispatches w i t h A t Least One Paragraph Favoring T h i s Position

Better

2.0

245

Same

0.7

10

Worse

1.8

222

A n y Position

3.0

306

180

Table C.6.

Summary of text analysis for unemployment

for unemployment

more important* equal

versus inflation:

imftortance, and inflation more

scoring important.

See Table C.l for explanation. The Nexis search identified 1,591 dispatches, of w h i c h 1 , 5 8 2 w e r e r e t r i e v e d at r a n d o m .

Step i n Analysis

Characters o f text No.

Dispatches

% Orig.

No.

8-Char. Words

% Orig.

per Dispatch

Nexis Retrieval

2,300,000

100

1582

100

177

First Filter

1,800,000

79

1183

75

189

695

44

Scored t o Favor Unemployment Important, Equal Importance, and Inflation Important

Position Favoring Importance Of

Average Paragraphs I n Dispatches W i t h A t Least One Paragraph Favoring This Position

Total Dispatches W i t h A t Least One Paragraph F a v o r i n g T h i s Position

Unemployment

1.4

281

Equal

1.1

243

Inflation

1.4

442

A n y Position

1.7

695

181

Table C.7. Summary of text analysis for Contra aid: scoring for infons favoring and opposing aid. See Table C . l for explanation. The N e x i s search i d e n t i f i e d 1,156 d i s p a t c h e s , o f w h i c h 969 w e r e r e t r i e v e d at r a n d o m .

Step i n Analysis

Characters o f text No.

% Orig.

Dispatches No.

8-Char. Words

% Orig.

per Dispatch

Nexis Retrieval

2,000.000

100

969

100

258

First F i l t e r

1,300,000

63

920

95

164

770 906

84 98

Scored to Favor and Oppose A i d : by Fan by S w i m et al.

Position

Average Paragraphs In Dispatches W i t h A t Least One Paragraph Favoring This Position

Total Dispatches W i t h A t Least One Paragraph Favoring T h i s Position

Fan Analysis: Favor A i d Oppose A i d Either Position

1.2 1.3 2.0

585 620 770

S w i m et al. Analysis: Favor A i d Oppose A i d Either Position

1.7 2.2 3.4

770 837 906

Copyright

Appendix D

Details of Actual Public Opinion Projections

The basic method for c o m p u t i n g public o p i n i o n f r o m A P infons was o u t l i n e d i n Appendix A (Section A . 10) and involved ( 1 ) constructing "skeleton persuasive force" functions G " f r o m the persistence constant, i n f o n content scores, and i n f o n emission times, ( 2 ) f o r m u l a t i n g a " p o p u l a t i o n conversion m o d e l " describing the o p i n i o n conversions due to persuasive messages, and ( 3 ) using functions G " , the population conversion model, the ' modified persuasibility constants," and an i n i t i a l set o f p o l l values t o compute public opinion. The calculations for the i n f o n content scores have already been presented i n A p p e n d i x C. T h i s appendix describes the calculations o f the persuasive forces d r i v i n g o p i n i o n changes and the subsequent projections o f public o p i n i o n .

D.l

C O M P U T A T I O N S O F PERSUASIVE FORCES

The equations f o r the calculations are those g i v e n i n A p p e n d i x A . T h e assumptions specific t o computations f o r A P dispatches are given i n Section A.10. F i r s t , a l l the infons f a v o r i n g a p o s i t i o n were p o o l e d a n d their dates o f transmission and content scores were used f o r c o m p u t i n g skeleton persuasive force f unctions G"j(t) using postulated values f o r the persistence constant and Equation t

A.27 o f A p p e n d i x A . Unless otherwise stated, this persistence constant was assigned a one day half-life f o r all infons.

D.2

POPULATION CONVERSION MODELS

These models, presented as figures i n Chapter S, specify the indices over w h i c h the s u m m a t i o n s are p e r f o r m e d i n E q u a t i o n A . 2 6 . E v e r y m o d e l has one subpopulation corresponding t o each o f the p o l l positions. T h e s u b p o p u l a t i o n names always begin w i t h " B " ( f o r "believers" i n a position). T h e other elements i n the models are the persuasive force functions ( o r the skeleton persuasive force f u n c t i o n s G " , since these t w o types o f functions d i f f e r o n l y by a constant o f proportionality). The models describe the opinion changes resulting f r o m the actions o f individual persuasive force functions.

IS4 D.3

Appendix

D

O P I N I O N PROJECTIONS

After f o r m u l a t i o n o f population conversion models, o p i n i o n computations were made f r o m Equation A . 2 6 . The computations began w i t h the t i m e and o p i n i o n percentages at the first measured poll point. Then calculations were made for opinion one At later using the o p i n i o n projection equations for a l l the subpopulations. The result was o p i n i o n predictions for a l l positions at the new time. These values where then used for further calculations after another At. The process was repeated u n t i l the time o f the last p o l l p o i n t A l o n g the way, estimates were obtained f o r a l l the intermediate poll points for a l l the poll positions. F o r these calculations, the At was usually chosen to be a few hours. T h i s At was k n o w n since it was chosen by the investigator. T h e 77/? ratio was also k n o w n since that was o n l y the reciprocal o f the fraction o f A P dispatches studied among the total identified. The unknowns required in projection equations are: 1. The persistence constant. F o r each o f the six examples, this constant was optimized by calculating the mean squared deviations ( M S D ) (see Appendix A. Section A.10) between the calculated opinion and opinion as measured in polls. T h e o p t i m u m persistence constant was the one g i v i n g the lowest M S D . Since a one day h a l f - l i f e was a g o o d consensus value for this constant (Chapter 4), this value was used for all plotted projections. 2. The modified persuasibility constant k'2. The same k*2 was assumed for all permitted transitions (see Equation A.14) and was obtained by m i n i m i z a t i o n o f the M S D . 3. R e f i n i n g weights for infons favoring different positions. Once the best c o m m o n k'2 was assigned, the projections were examined to see i f expected opinion for any one position was consistently too h i g h or too l o w . I f so. then a m i n i m u m M S D calculation was performed to see i f a refining weight w^yy different from 1.0 gave a lower M S D .

I f a significantly l o w e r M S D was observed, then that H>/,y" was used for the projections. W h e n e v e r any o f the constants discussed above were o p t i m i z e d , a l l other constants were retested to see i f they s t i l l gave the best value. I f not, a l l were r e o p t i m i z e d u n t i l the best c o m b i n a t i o n o f constants was o b t a i n e d . T h e o n l y exception was for Contra aid, where a one day persistence half-life was used since this was the best consensus constant for all issues. 4

Copyrights

References

Abelson, R. P., Kinder, D . R „ Peters, M . D . , and Fiske, S. T . (1982). A f f e c t i v e and semantic components i n political personal perception. Journal of Personality and Social Psychology. 42, 619-630. A l l e n , B . (1932). Journal

A stochastic interactive model f o r the diffusion o f i n f o r m a t i o n .

of Mathematical

Sociology,

Bartholomew, D . J . (1976). duration o f interest. Journal

8, 187-199.

C o n t i n u o u s time d i f f u s i o n models w i t h r a n d o m of Mathematical Sociology. 4 187-199. 9

Bartholomew, D . J . (1981). Mathematical Wiley. Bartholomew, D . J . (1982). Stochastic York: Wiley.

methods

in social science.

models for social processes

New York:

( 3 r d ed.). New

Beal, G . M . , and Rogers, E. M . (1960). The adoption of two farm practices in a central Iowa community (Special Report 26). Ames, I A : A g r i c u l t u r a l and Home Economics Experiment Station. Bender, P. M . ( 1 9 8 0 ) . M u l t i v a r i a t e analysis w i t h latent variables: modeling. Annual Reviews of Psychology. 31, 419-456. B e n d e r , P. M . , a n d S p e c k a r t , G ,

Psychological

Review,

(1979).

M o d e l s o f attitude-behavior

Causal

relations.

86 452-464. %

Berelson, B „ and Freedman, R. American, 210, 29-37.

(1964).

A study o f f e r t i l i t y c o n t r o l .

Scientific

Brams, S., a n d R i k e r , W . H . (1972). M o d e l s o f c o a l i t i o n f o r m a t i o n i n v o t i n g bodies. I n Herndon and Bernd (eds.), Mathematical applications to political science TV (pp. 79-124). Charlottesville: University o f V i r g i n i a Press.

Copyrighte

IS6

References

Campbell, B. (1979). 1he American

electorate:

Attitudes and action.

New York:

H o l t , Rinehart and W i n s t o n . Cavalli-Sforza, L . L., and Feldman, evolution;

A quantitative approach.

M.

W . (1981). Cultural transmission Princeton: Princeton University Press.

C a v a l l i - S f o r z a , L . L., F e l d m a n , M. W „ Chen, K . H „ a n d D o r n b u s c h , S. (1982). Theory and observation i n cultural transmission. Science. 2 7 5 , 2 7 . ( 1 9 7 5 ) . Political

Chaffee, S. H .

communication:

Enduring

issues for

and

M.

research.

Beverly H i l l s : Sage. Coleman, J . S. (1964). Introduction Press.

to Mathematical

Sociology.

Coleman, J . S., Katz, E., and Menzel, H . (1966). Medical study. N e w Y o r k : B o b b s - M e r r i l l . Comstock, G., Chaffee, S., Katzman, N., M c C o m b s , M., Television

Glencoe, I L : Free

innovation:

A

diffusion

and Roberts, D .

(1978).

N e w Y o r k : C o l u m b i a University Press.

and human behavior.

Conover, P. J . , and Feldman, S. (1986). E m o t i o n a l reactions t o the economy:

I'm

mad as hell and I ' m not going t o take i t anymore. American Journal of Political Science, 50,50-78. C o o k , F. L . , Tyler, T . R., Goetz, E. G., G o r d o n , M. T . , Protess D . , Leff, D . R.. and M o l o t c h , H . L . (1983). M e d i a and agenda setting: Effects o n the p u b l i c , interest group leaders, p o l i c y makers, and policy. Public Opinion Quarterly, 47, 16-35. Daley, D . J . , and K e n d a l l , D . G . Institute

Downs, A. Row.

of Mathematics

Stochastic r u m o r s . Journal and Its Applications, /, 42-53.

(1957). An economic

(1965).

theory

D u f f y , G „ and M a l l e r y , J . C. (1986, system for

natural

language

of democracy.

March).

modeling.

RELATUS:

of the

New York:

Harper and

An artificial

intelligence

Paper presented at the meeting o f the

International Studies Association. D u n p h y , D . C , B u l l a r d , C. G., and Crossing, E. E. M . (1974). Validation of the General Inquirer Harvard TV Dictionary. Paper presented at the Pisa, I t a l y Conference on Content Analysis. Dyer, M .

(1982).

I n - d e p t h understanding:

processing f o r narrative comprehension. Report,

a c o m p u t e r m o d e l o f integrated Yale

Computer

Science

Research

219.

E a g l y , A . H „ and H i m m e l f a r b , S. (1978). A t t i t u d e s and o p i n i o n s . I n M. Rosenzweig and L . W . Porter (eds.), Annual Review of Psychology, 29.

R.

Copyrights

References

187

E r b r i n g , L . , Goldenberg, E. N . , and M i l l e r , A . H . (1980). Front-page news and real-world cues: A new look at agenda-setting by the media. American Journal of Political Science, 24, 16-49.

(1982).

Everson, D . Public Y o r k : F r a n k l i n Watts.

opinion

and interest groups

in American

politics.

New

Fan, D . P. (1984). Mathematical models f o r the impact o f i n f o r m a t i o n o n society. Political Methodology, 10.479-494. Fan, D . P. (1985a). Ideodynamics: T h e kinetics o f the evolution o f ideas. Journal of Mathematical

11,1-24.

Sociology,

Fan, D . P. (1985b). Ideodynamic predictions for the evolution o f habits. Journal of Mathematical Sociology, 11, 2 6 5 - 2 8 1 . Fennel, R. D . , and Lesser, V . R. (1977, February). Parallelism i n A I p r o b l e m solving: A case study o f H E A R S A Y I I . IEEE Transactions on Computers, p . 98. Fishbein, M . , a n d Ajzen, I . (1975). Belief, attitude, intention and behavior: introduction to theory and research. Reading, M A : Addison-Wesley.

An

Funkhouser, G. R. (1973a). The issues of the sixties: An exploratory study in the dynamics o f public o p i n i o n . Public

Opinion

Quarterly,

37 62-75. %

Funkhouser, G . R. (1973b). Trends i n media coverage o f issues o f the sixties. Journalism Quarterly, 5 0 , 533-538. Funkhouser, G . R., a n d M c C o m b s , M . E . (1972). P r e d i c t i n g the d i f f u s i o n o f information t o mass audiences. Journal of Mathematical Sociology, 2, 121-130. Goldberg, A . S. (1966). Discerning a causal pattern among data o n voting behavior. American Political Science Review, 60, 913-922. Graber, D . A . (1984). Mass media and American DC: Congressional Quarterly. Granovetter, M . Sociology,

78,

(1973).

T h e strength o f weak lies.

( 2 n d ed.). Washington,

American

Journal

of

1360-1380.

Granovetter, M . (1978). Journal

politics

of Sociology,

Granovetter, M . (1980).

American

Threshold models o f collective behavior.

83, 1420-1443. The strength

of weak ties: A network

theory

revisited.

Paper presented at the meeting o f the International C o m m u n i c a t i o n Association, Acapulco. Gray, L . N „ and v o n Broembsen, M . H . (1974). O n simple stochastic d i f f u s i o n models. Journal of Mathematical Sociology, 3, 231-244. Haegerstrand, T . (1967). Innovation diffusion L o n d o n : University o f Chicago Press.

as a spatial process.

Chicago and

Copyrightec

188

References

H a m b l i n , R. L . Jacobsen, R. B., and M i l l e r , J . L . L . theory of social change. N e w Y o r k : W i l e y . f

H e w i t t , C. (1976). Viewing

control structures

(1973).

A

as patterns of passing

mathematical

messages

(AI

M e m o 410), Cambridge, M A : M I T A r t i f i c i a l Intelligence Laboratory. Hibbs, D . A . ( 1 9 7 9 ) . The mass p u b l i c and macroeconomic performance: The dynamics o f p u b l i c o p i n i o n t o w a r d u n e m p l o y m e n t and i n f l a t i o n . American Journal of Political Science, 23,705-731. H o v l a n d , C. I . (1959). Reconciling c o n f l i c t i n g results derived f r o m experimental and survey studies of attitude change. American Psychologist, 14, 8-17. Huba, G. J . , and Bentler, P. M . (1982). O n the usefulness o f latent variable causal m o d e l i n g i n testing theories o f naturally occurring events ( i n c l u d i n g adolescent drug use): A rejoinder to M a r t i n . Journal of Personality and Social Psychology, 43, 6 0 4 - 6 1 1 . Huba, G. J . , W i n g a r d , J . A., and Bentler, P. M . ( 1 9 8 1 ) . A comparison o f t w o latent variable causal models for adolescent drug use. Journal of Personality and Social Psychology. 40, 180-193. Iyengar, S., and K i n d e r , D . R. (1986). M o r e than meets the eye: T V p r i m i n g and p u b l i c evaluations o f the president Public Communication Behavior 1,135-171.

news, and

Iyengar, S „ K i n d e r , D . R., Peters, M . D., and Krosnick, J . A . (1984). The evening news and presidential evaluations. Journal of Personality and Social Psychology, 46, 778-787. Iyengar, S., Peters, M . D . and Kinder, D . R. (1982). Experimental demonstrations o f the " n o t - s o - m i n i m a r consequences o f television news programs. American Political Science Review, 76. 848-858. t

Karmeshu, and Pathria, R. K. (1980a). Stochastic evolution o f competing social groups. Journal of Mathematical Sociology, 7,47-58. Karmeshu, and Pathria, R. K . (1980b). Stochastic evolution o f a nonlinear model o f diffusion o f information. Journal of Mathematical Sociology, 7, 5 9 - 7 1 . Katz, E. (1957). The two-step flow o f c o m m u n i c a t i o n : A n up-to-date report on an hypothesis. Public Opinion Quarterly, 21, 61-78. Katz, I . , and Feldman, J . (1962). The debates i n the l i g h t o f research: A survey o f surveys. I n S. Krauss (ed.), The great debates. Bloomington: Indiana University Press. Katz, I . , and Lazarsfeld, P. F. (1965). people in the flow of communications.

Personal

K e l l y , E. F., and Stone, P. J . (1975). senses. Amsterdam: North Holland.

Computer

influence:

The part played

by

Glencoe, I L : Free Press. recognition

of English

word

Copyrights

References K e m e l l , S. (1978). E x p l a i n i n g presidential popularity. American Review,

72.

Political

189 Science

506-522.

K i n d e r , D . , and Sears, D .

(1985).

Public o p i n i o n and p o l i t i c a l action.

L i n d z e y and E. Aronson (eds.). Handbook

of social psychology

In G.

(3rd ed., v o l . 2,

p p . 659-741). Hillsdale, N J : Lawrence E r l b a u m . Klapper, J . T . (1960). Effects

of mass communication.

Glencoe, I L : Free Press.

Kraus, S., and Davis, D . (1976). The effects of mass communication behavior. University P a r k : Pennsylvania State University Press. Lane, R. E., and Sears, D . O. Prentice-Hall.

(1964).

Public

opinion.

on

political

Englewood Cliffs, NJ:

Lasswell, H . D., and Namenwirth, J . Z . (1968). The Las swell Value Dictionary vols.). N e w Haven: Yale University, ( m i m c o ) .

Lazarsfeld, P., Berelson, B., and Gaudet, H . Y o r k : D u e l l . Sloan, and Pierce.

(1944).

Lecomte, A., Léon, J . , and Marandin, J . - M . (1984). de description textuelle. Mots 9, 143-165.

The people's

choice.

(3

New

Analyse de discours: Stratégie

M a c K u e n , M . B . (1981). Social communication and the mass policy agenda. In M . B . M a c K u e n and S. L. Coombs, More than news; Media power in public affairs. Beverly H i l l s : Sage. M a c K u e n , M . B . (1983). Political drama, économie conditions, and the dynamics o f presidential popularity. American Journal of Political Science, 27,165-192. M a c K u e n , M . B . (1984). Exposure to information, belief integration, and individual responsiveness to agenda change. American Political Science Review, 78, 372¬ 391. Marcus, G. E. (1986). If there is a "Spiral of Silence," how might it work? Some speculations. Paper presented at the annual meeting o f the W o r l d Association for Public O p i n i o n Research/American Association for Public O p i n i o n Research, Saint Petersburg, F L . Markus, G . B. (1982). Political attitudes during an election year: A report on the 1 9 8 0 N E S P a n e l Study. American Political Science Review. 76. 538-560. M a r k u s , G . B . (1986). S t a b i l i t y and change i n p o l i t i c a l attitudes: recalled, and explained. Political Behavior, 5,21-45.

Observed,

M c C o m b s , M . C , and Shaw, N . (1972). The agenda setting function o f the mass media. Public Opinion Quarterly, 36,176-187. M c G u i r e , W . J . (1986). The m y t h o f mass media effectiveness: Savagings and Salvagings. I n G. Corns lock (ed.), Public Communication and Behavior.

Copyrights

190

References

M c l v c r , J . P. and Carmines, E. G. (1981). Sage. t

Unidimensional

scaling.

Beverly H i l l s :

M c T a v i s h , D . G „ and Pirro, E. B . (1985). Contextual content analysis. Paper presented at the meeting o f the Pacific Sociological Association, Seattle, Wash. M e n z e l , H . (1957). Public and private c o n f o r m i t y under different conditions o f acceptance i n the group. Journal of Abnormal and Social Psychology, 55, 398¬ 402. M u e l l e r , J . E. (1970a). C h o o s i n g Quarterly, 34, 395-402.

among

133 candidates.

Public

Opinion

Mueller, J . E. (1970b). Presidential popularity f r o m T r u m a n to Johnson. Political Science, 64, 18-34.

American

N a m e n w i r t h , J . Z., and Weber, R. P. (1984). The Lasswell Value Dictionary. I n J . Z. N a m e n w i r t h and R. P. Weber (eds.), Culture indicators research; A content analytic perspective. Unpublished manuscript N e u m a n , W . R. (1987). Parallel content analysis: O l d paradigms and new proposals. I n G . Comstock (ed.). Public communications and behavior. New Y o r k : Academic Press. Noel le-Neumann, E. (1973). Return to the concept o f powerful mass media. I n H . E g u c h i and K. Sata, (eds.), Studies in broadcasting. Tokyo: Nippon Hoso Kyokai. Noelle-Neumann, E. (1977). Turbulences i n the climate o f o p i n i o n : methodological applications o f the spiral o f silence theory. Public Opinion Quarterly, 41, 143¬ 158. Noelle-Neumann, E. (1984). The spiral of silence: Chicago: University o f Chicago Press.

Public

opinion-our

social

skin.

O s t r o m , C. W „ and S i m o n , D . M . (1985). Promise and performance: A dynamic model o f presidential p o p u l a r i t y . American Political Science Review. 79, 334¬ 357. Page, B . I . , and Shapiro, R. Y . (1982). Changes i n A m e r i c a n s ' p o l i c y 1935-1979. Public Opinion Quarterly, 42, 24-42.

preferences,

Page, B . I . , and Shapiro, R. Y . (1983a). Effects o f p u b l i c o p i n i o n on p o l i c y . American Political Science Review, 77, 175-190. Page, B . I . , and Shapiro, R. Y . ( 1 9 8 3 b ) . The mass media and changes in Americans' policy preferences: A preliminary analysis. Paper presented at the annual meeting o f the M i d w e s t Political Science Association, Chicago, 111. Page, B . I . , and Shapiro, R. Y . (1984). Presidents as o p i n i o n leaders: Some new evidence. Policy Studies Journal, 12.649-661.

Copyrights

References

191

Page, B. I ., Shapiro, R. Y., and Dempsey, G . R. (1985). The mass media do affect policy preferences. Paper presented at the annual meeting o f the A m e r i c a n Association for Public O p i n i o n Research, M c A f e e , N J . Page, B . I . , Shapiro, R. Y., and Dempsey, G . R. (1987). o p i n i o n . American Political Science Review, 8 7 , 23-43.

W h a t moves p u b l i c

Paletz, D . S., and Entman, R. M . (1981). Media power politics.

New York;

Free

Press.

Patterson, T . E., and M c C l u r e , R. D . (1976). The unseeing eye: The myth of television power in national elections. N e w Y o r k : G. P. Putnam. Robinson, M . J . ( 1 9 7 6 ) . Public affairs t e l e v i s i o n a n d the g r o w t h o f p o l i t i c a l malaise. American Political Science Review, 70,409-432. R o b i n s o n , M . J . , a n d Clancey, M . ( 1 9 8 4 ) . Quarterly, 7, 14. Rogers, E. M . (1983).

Diffusion

of innovations

Rogers, E. M . , and Shoemaker, F. F. ( 1 9 7 1 ) . N e w Y o r k : Free Press.

Teflon politics.

Public

Opinion

(3rd ed.). N e w Y o r k : Free Press. Communication

of

innovations.

Rosenberg, S. L . , McCafferty, B . P., and Harris, K . (1986). The image and the vote: T h e effect o f candidate presentation on voter preference. American Journal of Political Science, 30, 108-127. Ryan, B., and Gross, N . C. (1943). T h e diffusion o f h y b r i d seed corn i n t w o I o w a communities. Rural Sociology, 8, 15-24. Schank, R. C , G o l d m a n , N . , Rieger, C., and Riesbeck, C . ( 1 9 7 3 ) . M A R G I E : M e m o r y , analysis, response generation and inference i n English. Proceedings of tlte 1JCA1, Stanford University, p . 255. Sears, D . O , and Chaffee, S. H . (1979). Uses and effects o f the 1976 debates: A n overview o f empirical studies. I n S. Krauss (ed.), The great debates, 1976: Ford vs. Carter. B l o o m i n g t o n : Indiana University Press. Sears, D . O., and Freedman, J . L . (1967). Selective exposure t o i n f o r m a t i o n : A critical review. Public Opinion Quarterly, 3/, 194-213. Sharma, C. L „ Pathria, R. K., and Karmeshu. (1983). D i f f u s i o n o f i n f o r m a t i o n i n a social group. Journal of Mathematical Sociology, 9, 211-226. Shaw, D . L . , and M c C o m b s , M . E . (1977). The emergence issues: The agenda-setting function of the press. Publications. Stang, D . J . ( 1 9 7 4 ) . Methodological Psychological Bulletin, 81, 1014-1025.

factors

of American

political

S t . Paul, M N : W e s t

i n mere exposure

research.

CoDvriahted material

192

References

Stone

P. J . , D u n p h y , D . C , S m i t h , M . S., and O g i l v i e , D . M . ( 1 9 6 6 ) . general inquirer: A computer approach to content analysis. Cambridge: Press. f

Straffin, J . P. D., Jr. (1977). The bandwagon curve. American Journal Science. 27.695-709.

of

The MIT

Political

Thibadeau, R., Just, M . A., and Carpenter, P.A. (1982). A model o f the time course and content o f reading. Cognitive Science, 6, 157. Trenaman, J . , and M c Q u a i l , D . (1961). Television and the political image. L o n d o n : Methuen. Wagner, J. (1983). Media d o make a difference: T h e differential impact o f the mass media i n the 1976 presidential race. American Journal of Political Science, 27, 407-430. Weber, R. P. (1985). Basic content analysis.

Beverly H i l l s : Sage.

Weiss, W . (1969). Effects o f the mass media o f communication. I n G. Lindzey and E. Aronson (eds.), Handbook of social psychology (2nd ed., v o l . 5, pp. 77-195). Reading: Addison-Wesley. Zajonc, R. B . (1968). A t t i t u d i n a l effects o f mere exposure. Journal and Social

Psychology,

of

Personality

9, 1-27.

Zajonc, R. B . ( 1 9 8 0 ) . Feeling and t h i n k i n g : Preferences need no inferences. American Psychologist, 35, 151-175. Zielske, H . A., and H e n r y , W . A . (1980, A p r i l ) . Remembering and forgetting television ads. Journal of Advertising Research, 20, 7-13.

Copyrights

Author Index

Abelson, R. P., Kinder, D. R.. Peters, M . D . and Fiske, S. T., 14 Aj2en, I „ see Fishbein. A l l e n , B., 32 t

B a n h o l o m e w , D . J . , 16, 27, 32 Beal, G . M . , and Rogers, E. M . , 34 Bender, P. M . , 32, 33 see also Huba. and Speckart, G „ 32, 33 Berelson, B., and Freedman, R., 117 see also Lazarsfeld. Brams, S., and Riker, W . H „ 32, 5 2 , 139 v o n Broembsen, M . H . , see Gray. B u l l a r d , C. G., see D u n p h y . C a m p b e l l , B . , 31 Carmines, E. G., see M c l v e r . Carpenter, P. A., see Thibadeau. Cavalli-Sforza, L . L., and Feldman, M . W., 32, 33, 34 and Feldman, M . W., Chen, K. H . , and Dornbusch, S. M . , 2 0 , 32, 34 Chaffee, S. H . 11 see also Comstock. see also Sears. Chen, K . H . , see Cavalli-Sforza. Clancey, M . , see Robinson. Coleman, J . S., 3 2 and Katz, E.. and Menzel, H . , 32, 33, 133 t

Comstock, G.. Chaffee, S., Katzman, N „ M c C o m b s , M . , and Roberts, D . , 133 Conover, P. J . , and Feldman, S., 14 Cook, F. L . , T y l e r , T . R „ Goetz, E. G., G o r d o n , M . T . , Protess D . , L e f f , D . R., a n d M o l o t c h , H . L., 10, 32, 132 Crossing, E. E. M . , see D u n p h y . Daley, D . J., and Kendall, D . G., 32 Davis, D . , see Kraus. Dempsey, G . R „ see Page. D o r n b u s c h , S. M „ see C a v a l l i Sforza. Downs, A., 9, 18 Duffy, G., and Mallery, J . C , 45 Dunphy, D . C , B u l l a r d , C. G., and Crossing, E. E, M „ 4 5 see also Stone. Dyer, M . , 45 Eagly, A . H . , and H i m m e l f a r b , S., 1£ Entman, R. M . , see Paletz. E r b r i n g , L . , Goldenberg, E . N „ and M i l l e r , A . H . , 10, 35, 132 Everson, D., 39 Fan, D . P., 2 3 , 2 7 . 30, 34, 35, 139, 141 Feldman, J., see Katz. Feldman, M . W., see Cavalli-Sforza. Feldman, S., see Conover. Fennel, R. D., and Lesser, V . R., 45

3l

194

Author Index

Fishbein, M . , and Ajzen, L, 33 Fiske, S. T , see A b e i s o n . Freedman, R., see Berelson. see Sears. Funkhouser, G. R., 10, 29, L32 and M c C o m b s , M . E., 10 Gaudet, R , see Lazarsfeld. Goetz, E. G., see Cook. Goldberg, A. S., 32 Goldenberg, E. N., see Erbring. G o l d m a n . N . , see Schank. Gordon, M . T., see C o o t Graber, D . A., I I Granovetter, M . , 1 ^ 33 Gray, L . N., and von Broembsen, M . H . , 32 Gross, N . C , see Ryan. Haegerstrand, T., 32 H a m b l i n , R. L., Jacobsen, R. B., and M i l l e r , J . L . L „ 16, 2 7 , 139 Harris, K., see Rosenberg. Henry, W . A., see Zielske. H e w i t t , C , 45 Hibbs, D . A.. 27. 29, 130. 144. 146 H i m m e l f a r b , S., see Eagly. H o v l a n d , C. L 16 Huba, G. J . , and Bender, P. M . , 32, 33 and Wingard, J. A., and Bender, P. M . , 32, 33 Iyengar, S. and Kinder, D . R., 1 3 2 , 1 3 4 and Kinder, D. R., Peters, M . D., and Krosnick, J . A., 34, L32 and Peters, M . D., and Kinder, D . R., 10, 132, 134 t

Jacobsen, R. B., see H a m b l i n . Just, M . A., see Thibadeau. Karmeshu, and Pathria, R. K., 32 see also Sharma. Katz, E., 1 1 and Feldman, J., 132 and Lazarsfeld, P. F., 31 see also Coleman.

Katzman, N., see Comstock. K e l l y , E. F., and Stone, P. J . , 45 Kendall, D. G., see Daley. K e r n e l l , S., 13Q Kinder, D., and Sears, D., 5 2 see also Abeison. see also Iyengar. Klapper, J . T „ 1 1 , 31 Kraus, S., and Davis, D., 11 Krosnick, J . A., see Iyengar. Lane, R. E., and Sears, D . O., 134 Lassw e l l , H . D . , and N a m e n w i r t h , J . Z., 45 Lazarsfeld, P., Berelson, B., and Gaudet, i L 1 1 . 3 1 , 130, 131 see also Katz. Lecomte, A., Léon, J . , and M a r a n d i n , J.-M., 45 Leff, D., see Cook. Léon, J . , see Lecomte. Lesser, V . R., see Fennel. McCafferty, B . P., see Rosenberg. McCIure, R. D . , see Patterson. M c C o m b s , M . C , and Shaw, N „ 10 see also Comstoc k. M c C o m b s , M . E., see Funkhouser. see Shaw, D . L . McGuire, W . J., U M c l v e r , J . P., and Carmines, E. G., 32 M c Q u a i l . D., see Trenaman. M c T a v i s h , D . G., and P i n o , E. B „ 46

MacKuen, M. B., 10, 35, 120, L32

Mallery, J . C , see D u f f y . Marandin, J . - M . , see Lecomte. Marcus, G. E., 14 M a r k u s , G . B., 2 9 , 1 3 0 Menzel, H ^ 130 see also Coleman. M i l l e r , A . H ^ see E r b r i n g . M i l l e r J . L . L., see H a m b l i n . M o l o t c h , IL L., see C o o k . Mueller, J , E „ 52. 130. 132

Namenwirth, J. Z., and Weber, R. P., 4 5 see also Lasswell. Neuman, W . R., 35, 129 Noelle-Neumann, E „ 11. 131, 139

C

Author Index

O g i l v i e , D . M . , see Stone. Ostrom, C. W „ and S i m o n , D . M . , 27, 2 9 , 3 5 , 120, 130 Page, B . L and Shapiro, R. Y., 14, 15, 30, 35, 37. 38. 62, 123, 132 and Shapiro, R. Y., and Dempsey, G . R., 1 L 13, 15, 19, 30, 6 2 , 120, 123, 132 Paletz, D. S., and Entman, R. M . , 2 8 , 120 Pathria, R. K., see Karmeshu. see Sharma. Patterson, T . E., and M c C l u r e , R. D., 132 Peters, M . D . , see Abelson. see Iyengar. Pirro, E . B . , see M c T a v i s h . Protess, D . , see Cook. Rieger, C , see Schank. Riesbeck, C , see Schank. Riker, W . R , see Brams. Roberts, D., see Corns toe k. Robinson, M . J . 132 and Clancey, M . , 134 Rogers, E . M . , 1 1 , 22, 33, 132 and Shoemaker, F. F., 32, 33 see also Beat. Rosenberg, S. L., McCafferty, B . P., and Harris, K., 14 Ryan, B., and Gross, N . C , 117 f

Schank, R. C , G o l d m a n , N., Rieger, C , and Riesbeck, C , 45 Sears, D . 0. and Chaffee, S. R , 132 and Freedman, J. L., 31 see also Lane. Shapiro, R. Y., see Page. Sharma, C. L . , Pathria, R. K., and Karmeshu, 32, 34 Shaw, D . L., and M c C o m b s , M . E., 132 Shaw, N . , see M c C o m b s , M . C. Shoemaker, F. F., see Rogers. S i m o n , D . M . , see O s t r o m . S m i t h , M . S., see Stone. Speckart, G., see Bender. f

195

Straffin, Jr., P. D . , 5 2 , 139 Stang, D . J . , 5 2 Stone, P. J . , and D u n p h y , D . C , S m i t h , M . S., and O g i l v i c , D. M . , 45 see also K e l l y . Thibadeau, R., Just, M . A., and Carpenter, P. A., 4 5 Trenaman, J., and M c Q u a i l , D „ 131 Tyler, T . R., see C o o k . Wagner, J . , H , 132 Weber, R. P., 4 5 , 4 L 55 see also Namenwirth. Weiss, W., 31 W i n g a r d , J . A., see Huba. Zajonc, R. B., 52 Zielske, R A., and Henry, W . A., 130

Subject Index

A B C News. See O p i n i o n polls. Advertising, 4 , 3 8 Agenda setting b y media, 3^6, 28,

137

35.

Analogies: automobile, 16 colorblindness, 3D elephant, 4^5, 35 gorges, 133 m o v i n g pictures, 6=2 nuclear missile,

6^7, LL 12-13. 17,

140

M I R V e d , 7, 13, 12 tugboat, 137 A n t i b i o t i c s , 32-31 AP. See Associated Press. Associated Press. See Messages, mass media. A w a r e s , S u b p o p u l a t i o n o f . See Population.

Bigtie,5. 136-37 Birth and Death. See Ideodynamics. Cavalli-Sforza and Fcldman model, 34 Censorship, 3, 136 Cognitive dissonance, 4, 11, 12, 16 Coleman's physician studies, Content analysis: computer, 9,

32-33

121. 12S

7^

46-47. 55, 56,

artificial intelligence, 4 5

multipurpose computer programs: disambiguation, 45-47, 49, 55 M C C A , 46 General Inquirer, 4 5 unsuitability for opinion issues, 4 6 factor analysis, 55-56 method o f successive nitrations, 4M8,54^56 automation, 5 6 cheating, inability to, 56, 119 code words, 55 c u s t o m i z a t i o n b y issue, 4 5 , 55, 121 nitrations, 46-47, 122 disambiguation, 4 5 - 4 7 , 4 9 text homogeneity, 4 7 , 4 9 f l e x i b i l i t y , 54 generality, 54, 122 outline, 43 robustness, 122 scoring, 47-48. 121-23 specificity, 54, 55 speed, 47, S3 human, 120-21 Contras i n Nicaragua, a i d t o , 4. 12. 39 computer content analysis, 53-54 filtration, 53-54, 114 scoring: Fan, 53-54, 175 S w i m e t a l . . 53 54, 175 Nexis retrievals, 4 0 - 4 1 , 157-58

C

198

Subject

Index

opinion polls, 38-39, 53 opinion predictions, 54, 66-67 ldeodynamic parameters m o d i f i e d persuasibility constant, 62 persistence half-life, 66-67. 124,

12s

refining weights, 67 C r e d i b i l i t y . See Messages, i n f o n s , properties. C u m u l a t i v e effects o f information, 3¬ 4, 132-34 Data bases, electronic. See Messages, mass media. Death. See Ideodynamics. Defense spending, 4 , 7, 10, 12-15. 17. 20-22. 38 details o f calculation. See Ideodynamics, opinion predictions, computer content analysis, 49-50 nitrations, 4 8 , 165-72 scoring: three position, 49-50, 168-71 t w o position, 49-50, 171-72 waste a n d fraud, 50, 1 7 2

m i n i m a l effects o f the media, 132¬ 34 Nexis retrievals, 4 0 - 4 1 , 1 5 5 - 5 6 o p i n i o n polls, 38 opinion predictions, 57-62 1977-1984: persuasive force functions, 57¬ 58,12126 population conversion model, 2 5 . 58 ideodynamic parameters: mean squared deviation, 58, 152 modified persuasibility constant, 58-60, 125-26, 127 persistence half-life, 58-59 refining weights, 58-59 three position, 57, 60 t w o position, S7-S8. 60 model comparisons, 58, 152 1977-1986, 6 1 waste and fraud, 62 Deductive approach, 7 , 2 Democracy, 3, 120, 137, 139, 140

Democratic p r i m a r y , 4, 15, 1 6 , 20, 22,

135-36

computer content analysis, 4 8 , 52¬ 53 bandwagon analysis: filiations, 52, 173 scoring, 52, 173 name count analysis, 52-53, 127¬ 28,173 Nexis retrievals, 4 1 , 1 5 6 - 5 7 o p i n i o n polls, 38, 5 2 opinion predictions, 64-65 persuasive force functions, 64-65 bandwagon analysis, 64-65 modified persuasibility constant, 64-65 population conversion model, 64-65 social mechanism, 140 name count analysis, 64-65 inaccuracies, 65 population conversion m o d e l , 65 model comparisons, 68 Dictatorship, 6 Don't K n o w s , Subpopulation of. See Undecided Drug abuse, 33-34 Economic climate, 4, 39, 53, 65 computer content analysis, 53, 173¬ 74 Nexis retrievals, 4 0 , 157 o p i n i o n polls, 39 opinion predictions, 65 Elites, 130, 134ı 140 Empirical testability. See Ideodyanmics

Epidemiology, 16, 34 E q u i l i b r i u m , subpopulation, 2 9 Event-driven messages, 1 3 8 , 1 4 0 F i l t r a t i o n s , t e x t . See Content analysis, c o m p u t e r , m e t h o d o f successive nitrations. Focus g r o u p s . See Laboratory Studies. Foreign adventurism. See Ideodynamics, o p i n i o n predictions, refining weights. Habits, 23, 30, 34, 139-40

Sut'jct

Half-life, exponential, 4-5. See Messages, infons, properties, persistence constant Heterogeneity: geographic, 31 population. See Population. Ideas, w i t h i n issues. See Issues. Ideodynamics, and attitudes, 4 , 9 , 2 Q and behavior, 4, 6 , 2 1 birth, 22-23. 154 death, 22-23. 154 e m p i r i c a l t e s t a b i l i t y , 5, 6, 9-10» 19, 22, 27, 28, 29, 30, 37, 118-20, 123 and data availability, 10 conditions, 118-19 and model s i m p l i c i t y , 9-10, L8 and s i m p l i f y i n g approximations, 9^10,28 time intervals f o r calculations, 27-28. 29, 118, 139 time span for calculations, 5, 22, 154 o p i n i o n predictions, 4 , 5, 6, L 67¬ 68 for aw ares, 141, 145-46 dependence: o n i n f o n s , 7 , 126. See Messages. and persuasive force functions, 17-22. 24. 124¬ 25, 144-45, 150, 183 a d d i t i v i t y , 19, 5 7 , 120, 124. 127. 153 and message lags, 27, 2 9 ,

34

and i n f o n saturation, 17, 19. 124. 130. 146. 142 and opinion reinforcement, 11-12. 16, 17, 19, 2 2 , 57, 124, 131-32. 146-47 o n subpopulations, 4, 7, 10¬ 12, 14, 16-17. 19-21. 30-33. 124-25, 141-42. 143, 146¬ 47, 154 insensitivity: to infon source, 123 to starting o p i n i o n , 127 details o f method, f o r defense spending, 10, 12-15. 16, 17, 20-22

t Index

199

mean squared deviation, 58, 60, 68, 152 nonlinearity, 17, 2 8 . 3Q overview, 4, 14, 16-17, 57, 183 no o p i n i o n change, 117, 122 parameters, 2 7 , 6 L 62, 67-68. 127-28. 135-36 attentiveness c o n s t a n t , 2 2 , 148, 153 modified persuasibility constant, 21-22. 58-60. 64¬ 65, 67-68. 125-28, 150-52, 184 number of, 118 persuasibility constant, 2 0 - 2 1 , 125, 147 refining weights, 21-22. 58-59. 67, 124-28, 134-36, 151-52, 184 and foreign adventurism, 136 optimization, 59-60. 152 population c o n v e r s i o n models, 17-22, 124, 127, 148, 183-84 sequential conversion m o d e l , 20 direct conversion model, 63-64. 66, 124 quantitative nature, 133 root mean squared deviation, 58, 60. 64-65. 68. 152 through o p i n i o n change, 4 for unawares, 10, 18, 22, 141-42, 144-48. 153 Inflation. See U n e m p l o y m e n t versus inflation. Infons. See Messages, infons, coding as. InfoTrend, 4 5 , 4 2 , 54-56. 117, 121¬ 22 Innovations, 29-30. 32-34 "Innovators," 32-33 Interview studies, 5 Issues: d e f i n i t i o n of, 9 , 1 0 , 13, 141 and positions, 1 1 , 15, 16, 121 Laboratory studies, 5, 34, 133 "Laggards," 32, 33 Lags: in message duration. See Ideodynamics, opinion predictions, dependence on infons in persuasion onset, 5

r.

200

Subject

Index

L a w o f the 24-hour day, 5, 6, 18-19. 134. 137 and s u p e r f i c i a l i t y o f persuasion, 134 Lebanon, troops i n , 4, 38-39 computer content analysis: filiations, 5 0 - 5 1 , 172-73 scoring, 5 0 - 5 1 , 1 7 3 m i n i m a l effects o f the media, 132 Nexis retrievals, 41-42, 156 opinion polls, 3 8 - 3 9 opinion predictions, 62-64 persuasive force functions, 64-65 truck bombing infons, 62-64 population conversion model, 63¬ 65 ideixlynamic parameters: modified persuasibility constant, 63*65 persistence half-life, 6 3 - 6 5 refining weights, 62 model comparisons, 63 Logistic. See Messages, generation. M a c r o phenomena, 5 Malleability, of opinion. See Opinion. Mathematical models. 19. 20. 32. 35 Mean squared d e v i a t i o n . See Ideodynamics, opinion predictions. Media and opinion, 3 J Messages, ?, 1 L 12-16. 128 and affective appeals, 14 causing o p i n i o n conversion, 11-12. 17-18. 19-7?. causing o p i n i o n reinforcement, 11¬ 12, 16-17, 19, 22, 131-32 directionality, 1 4 , 3 1 generation, 5 ^ 8, 130. 137-40 and logistic equation, 16» 2h 29¬ 30, 3 2 , 3 4 J 5 and message effect, 1 L 29, 30-31 and population equilibrium, 29 and opinion, 4 infons, coding as, 7. 2 9 . 34, 142¬ 43 a s M I R V e d missiles, 1 3 , 1 3 definition of, 2 dimensions, 29, 143 directness, 13-14. 19, 2 2 , 143¬ 44 positions favored, 142-43. See Issues.

sender o r source, 1_L 13. 14. I S . 1 4 1 . 148 index number, L i 15, 143-44 indirect infons, 12L22 personal experience, 32, 117,

119-20, 139

properties, 2 9 audience size, 14, 16«

2L 124-25. 142. 144

18-19.

mass media messages, ?,

142. 149

15,

broadcast time, 16 memory constant, persistence constant, 16, 18, 2L 22, 66-67, 124. 127. 183-84 content, 14-15. 18, 19, 2 L 144. 151 v a l i d i t y or c r e d i b i l i t y , 3^4, 10,

16,145

135. 149. 152.

14, 15, 18. 21. 118-21, 123¬

22, 136-37, 142,

144

and cognitive appeals, 14 mass media, 129-30, 132-34.

149

142,

Associated Press, 5^7, 2 8 , 3 2 , 39-42. 132-33 news budget, 4 0 data bases, electronic: Nexis, 3L 40, 120, 155 r e t r i e v a l s w i t h i n 50 word limits, 4 L 155 news magazines, 28, 39 newspapers: local, 28, 137 national, 15* 28, 32, 120 Vanderbilt Television News Archives, 15, 28, 3?, 120 mixed, 30. 121. 133. 138 neutral, 30, 121 non-mass media, 14. 119 rumor, 4, truck b o m b i n g . See L e b a n o n , opinion predictions. Page et a l . coding, 14, 15, 19, 30,

120-21, 130*

119-20. 133, 137

35, 120. 123

reinforcement,

11-12. 16-17. 19,

3a 35, 129, 131-32

quality, 14-15 salience, 16. senders. See Messages, generation, saturation, 17. 19. 2 2 . 30 Structure, 15-16. 21

14-15.

48, 121

C

Sulyea

M i n i m a l effects o f the media, 3, 11¬ 12, H 2 - 3 4 M o d i f i e d persuasibility constant. See Ideodynamics, opinion predictions. M S D . See I d e o d y n a m i c s , mean squared deviation. Networks, social. 33-34 News magazines. See Messages, mass media. Newspapers. See Messages, mass media. Nexis. See Data bases, electronic. N o Opinions, Subpopulation of. See Undecided.

Index

201

and subpopulations, 10-11. 16. 17, 19-20. 29-33,35 and cognitive dissonance. 11-12. 16 and m i m i m a l effects o f the media, 11-12 and o p i n i o n reinforcement. See Messages, c a u s i n g o p i n i o n conversion and causing opinion reinforcement, and Ideodynamics, opinion predictions. o f unawares, 10, 18, 22 o f undecideds, 10. 22 Positions, w i t h i n issues. See Issues. Prejudice, 5

Opinion: forecasts. See Ideodynamics, opinion predictions impermanence, 32 leadership, 5, 3 L 129-30 m a l l e a b i l i t y , 5, 2 0 - 2 1 . 117, \2L 133. 135. 137 v o l a t i l i t y . 20 2 1 . 117, 133 polls, 3^8 Page et al. data, 37-38 Roper Center, 38-39. 156 p r e d i c t i o n s . See I d e o d y n a m i c s , opinion predictions, time trends, 4, 5* 6, L *L 5*L 60, 119,12122 Page, Shapiro and colleagues. See Messages, Page et al. Parsimony, o f ideodynamics, 22, 22 Pathways for social change, 5 Paucity o f parameters, 5, 6. 128 Persistence constant. See Messages, in t o n s , p r o p e r t i e s , persistence constant. Persuasive f o r c e f u n c t i o n s . See Ideodynamics, o p i n i o n predictions, dependence on in forts. Persuasibility constant. See Ideodynamics, o p i n i o n predictions, parameters. Physicians, 32-33 Polls. See O p i n i o n . Popularity, presidential, 29 Population, 10-11, 16-17. 18, 19 o f awares, UL 17-22 heterogeneity, 31-33 structure, 141-42

Quantitative analysis, 3, 4^5 Real time, 5, 2 7 , 34 Refining weights. See Ideodynamics, opinion predictions. Regressions, 6 1 Reinforcing messages. See Messages. Reputation, 4 , 5 R M S D . See I d e o d y n a m i c s , root mean squared deviation. Root mean squared d e v i a t i o n . See Ideodynamics, o p i n i o n predictions, parameters. R u m o r . See Messages, non-mass media. S a t u r a t i o n , by messages. See Messages. Sequential c o n v e r s i o n m o d e l . See Ideodynamics, p o p u I at i o n conversion models. S m o k i n g , 30-31 "Stiflers," 34 Strength-of-weak-ties, 12 Subpopulation. See Population. Superficiality o f persuasion. See Law o f 24-hour day. Survey studies, 130-31

Television, 28, 34 Terrorism, 5. 138 Text analysis. See Content analysis. T i m e trends, o f o p i n i o n . See O p i n i o n predictions.

Co

202

Subject

índex

T w o step i n f o r m a t i o n transfer. See Opinion leadership. Unawares. See Population. U n c o u p l i n g o f message impact and generation, 5-6 Undecided, population of, 1 0 , 2 2 Underground press, 4 U n e m p l o y m e n t versus i n f l a t i o n , 4 , 29, 39, 135 computer content analysis, 53, 174 Nexis retrievals, 4 0 - 4 1 , 1 5 6 o p i n i o n polls, 39 opinion predictions, 66 Vanderbilt Television News Archives. See Messages, mass media. Weed spray, 34

Copyrighted material

About the A u t h o r

D A V I D P. F A N is i n t h e D e p a r t m e n t o f Genetics a n d Cell Biology at t h e U n i v e r sity o f M i n n e s o t a .

CoDvriahted material