P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Archives of Sexual Behavior, Vol. 30, No. 2, 2001
Searching for Sexually Explicit Materials on the Internet: An Exploratory Study of College Students’ Behavior and Attitudes Patricia Goodson, Ph.D.,1,4 Deborah McCormick, Ph.D.,2 and Alexandra Evans, Ph.D.3
The convergence of sexuality messages with a computerized medium (specifically, the Internet) represents an unprecedented phenomenon with, as of yet, unknown outcomes. Despite the Internet’s widespread use, little is known about users’ behaviors and attitudes when searching for sexually explicit materials online. This study examined specific behaviors and outcome expectations and expectancies (or attitudes) of a sample of 506 undergraduate students at a public university in Texas. Data were collected through a self-administered questionnaire, designed and pretested by the authors. Main results revealed that most students in the sample were infrequent and relatively new users of the Internet. Forty-three percent (43.5%) of students had sometime accessed sexually explicit materials through the Internet, but the practice was not very common. Only 2.9% said they accessed these materials “frequently.” Male students were significantly more likely to have accessed the Internet for viewing sexually explicit materials and to claim curiosity about sex as their motivation for this behavior. Women were significantly more likely to have experienced sexual harassment while online. In terms of attitudes, this sample did not appear to value highly or exhibit strong beliefs about the potential outcomes associated with accessing the Internet for sexually explicit materials. Competing explanations as well as the limitations of this study are discussed. KEY WORDS: Internet; erotica; attitudes; behavior; undergraduates.
1 Texas
A&M University, Department of Health and Kinesiology, College Station, Texas. Arizona University, Department of Health Promotion and Exercise Science, Flagstaff, Arizona. 3 School of Public Health, The University of Texas, Houston, Texas. 4 To whom correspondence should be addressed at Dept. of Health and Kinesiology, Texas A&M University, 4243 TAMU, College Station, Texas 77843-4243; e-mail:
[email protected]. 2 Northern
101 C 2001 Plenum Publishing Corporation 0004-0002/01/0400-0101$19.50/0 °
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
102
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
INTRODUCTION Communication media have actively impacted the social construction of human sexuality worldwide. Television, radio, movies, magazines, and music (among others) have modeled attitudes and behavior, created new vocabulary and symbols, and promoted norms for personal interactions. An illustration of this impact is the manner in which North-American adolescents and college students have consistently pointed to the media as having similar or stronger influence than their families on the development of their sexual knowledge, attitudes, and behaviors (Ballard and Morris, 1998; Haffner and Kelly, 1987; Koch, 1998). Presently, the convergence of sexuality messages with a computerized network medium may well represent one of the most unique and unprecedented combination of these two potent forces (sex and the Internet). This association will undoubtedly yield unparalleled effects (both positive and negative) in terms of the social construction of sexuality, establishment and development of intimacy, sexual attitudes and behaviors, and communication patterns (sexual or otherwise). These effects are worthy of social scientific examination (Cooper, 1998). On the positive end of the spectrum, the Internet is already impacting the dissemination of sexual knowledge through informational web sites such as the Sexuality Information and Education Council of the United States (SIECUS) webpage (www.siecus.org), “Go Ask Alice!” (www.goaskalice.columbia.edu), the Germany-based Archiv f¨ur Sexualwissenschaft, (http://www.rki.de/GESUND/ ARCHIV/HOME.HTM), www.teenwire.com of the Planned Parenthood Federation of America, and the Kinsey Institute Sexuality Information Service for Students (KISISS; www.indiana.edu/∼kinsey). Dissemination of information is not, however, the only positive effect of the interplay between the Internet and sexuality. Other constructive outcomes include the potential for development and maintenance of relationships (Cooper, 1998); the development of deeper, more personal interactions, as users are better able to self-disclose and take greater interpersonal risks (Cooper and Sportolari, 1997; Parks and Floyd, 1999; Parks and Roberts, 1998); the use of the Internet for advocacy and networking, especially among gay, lesbian, bisexual, and transgender groups (Carey, 1996; Weinrich, 1997); the development of support groups, communities, and online sex therapy (Cart, 1997; Newman, 1997; Scheerhorn et al., 1995); the use of the Internet as a tool for data collection in research of sexuality-related topics (Binik et al., 1999; Ross et al., in press); and the potential use of the Internet for delivery of counseling, disease control and prevention services (Acevedo et al., 1998; DeGuzman and Ross, 1999; Henry, 1999; Roffman et al., 1997). Despite such a rich composite of benefits, the interaction between the Internet and human sexuality has also engendered strong concerns regarding issues of censorship (Portelli and Meade, 1998), addictive usage or disruption of personal routines and relationships (Cooper et al., 1999), psychological well-being (Kraut
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Searching for Sexually Explicit Materials Online: User Behavior and Attitudes
103
et al., 1998), and distancing of individuals from their offline communities (Cart, 1997). There is also concern that the problems traditionally associated with other types of media and their treatment of sexuality may be transferable to the Internet. Among these problems are the desensitization toward violence among viewers of violent sexual materials, the juxtaposition of consumption of pornography and violent or criminal sexual behavior, and the relationship between attitudes toward sexuality exhibited in the popular media and the socialization process of children and teenaged viewers (Allen et al., 1995a,b; Becker and Stein, 1991; Fisher and Barak, 1991; Harris, 1994; Koop, 1987; Kutchinsky, 1991; Scott, 1991; Zillmann and Bryant, 1988). While the number of users increases rapidly (56.7 million in 1997, 78.6 million in 1998, 92.0 million in 1999 in the US and Canada) “sex” is, by far, one of the most commonly searched topics on the Internet. According to searchterms.com the term sex was ranked number two among the top ten search terms on the Internet in December of 1999 (based on a sampling of at least a million searches); “mp3” (or compressed digital music files) was ranked number one while “porn” and “playboy” were ranked 12th and 14th, respectively (CommerceNet, 1999; Newburger, 1999; Searchterms.com, 1999). Despite such widespread use, little is known about Internet users regarding their behaviors and attitudes when searching for sexually explicit materials online (Barak and Fisher, 1997). The goal of this study is to contribute to this new corpus of knowledge by examining these factors within a sample of college students. College students were chosen as the target population because their age, education, interest in sexuality, computer literacy, and access to computers make them particularly attractive for Internet information, dissemination, communication, and commerce (Tamosaitis, 1995).
THEORETICAL FRAMEWORK AND PURPOSE Social Cognitive Theory was the theoretical framework for this study. This theory has been extensively used in studies of the diffusion of ideas, products, and values through mass communication, as well as for exploring the effects of consumption of sexually explicit materials upon violent sexual behavior (Allen et al., 1995b; Bandura, 1994). In contrast with earlier theories of media influence, Social Cognitive Theory (SCT) provides a nonsimplistic view of the relationship among media messages/symbols, human attitudes, behavior, and environment. The theory proposes that people are both products and producers of their environment: individuals’ behaviors, personal factors (such as cognition, affection, and biology) as well as environmental influences all interact bidirectionally, or reciprocally (Bandura, 1994). In the case of Internet influence, people are not passive consumers of the medium; rather, they control the potential effects of its messages through
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
104
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
mechanisms of self-regulation and self-reflection, that is, by “filtering” its symbols through personal beliefs and values (Bryant and Zillmann, 1994). Although the impact of the media upon attitudes and behavior is hypothesized to be mediated through mechanisms of self-regulation and self-reflection, this impact is, nonetheless, especially powerful because of the manner in which the content is modeled to consumers. Content modeling is usually carried out through mechanisms of emotional arousal and observational learning (Bandura, 1986; Harris, 1994). Considering that most people in their daily lives interact only with small portions of their physical and social environments, much of people’s understanding of reality is acquired vicariously: by observing others and learning from their experiences. Furthermore, these observations are not neutral in terms of affect: they may generate strong emotions which, in turn, influence the learning that occurs. According to Bandura, to a large extent “people act on their images of reality. The more people’s images of reality depend on the media’s symbolic environment, the greater is its social impact” (emphasis added; Bandura, 1994). Considering human beings’ capacity for self-regulation, self-reflection, and observational learning, as well as the power of media in shaping consumers’ views of reality, identifying users’ behavior and outcome expectations and expectancies (or attitudes) regarding the media is vital. Learning about these behaviors and attitudes is paramount to both understanding and possibly modifying (for health promotion purposes, for instance) the impact of the Internet upon the social construction of sexuality. Outcome expectations (or antecedent determinants of behavior, in SCT) are defined as a person’s beliefs concerning the results of a certain action (e.g., “Sexually explicit materials on the Internet help me improve my sexual relationships offline.”); expectancies (or incentives, in SCT) constitute the values placed on, or the importance given to those anticipated results (e.g., “It is very important to me to improve my sexual relationships offline.”; Bandura, 1986). Outcome expectations and expectancies are determinants of attitudes, that is, a person’s attitude toward an object may be construed as a linear combination of outcome expectations and expectancies toward the object being investigated. For the purposes of this paper the term attitude will be employed to refer collectively to the set of outcome expectations (beliefs about outcomes of certain behaviors) and expectancies (value placed on specific behaviors and their outcomes) exhibited by participants in this study (Bandura, 1986; Baranowski et al., 1997). Statistical analyses will be presented for outcome expectations and expectancies separately, however, as the detailed examination of each component is more informative and revealing than the combination of constructs. In summary, the purpose of our study was to examine the following questions: What are some of the specific behaviors reported by college students while searching the Internet for sexually explicit materials? What are their attitudes (outcome expectations and expectancies) while searching the Internet for sexually explicit materials? Do males and females differ in their expectations, expectancies, and
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Searching for Sexually Explicit Materials Online: User Behavior and Attitudes
105
behavior? In this study, the phrase “sexually explicit materials” was defined (for participants) as those materials “that either show clear pictures of, or talk/write about sexuality using sexual vocabulary.” The phrases “use of the Internet for viewing sexually explicit materials” and “use of the Internet for sexual entertainment” will be used interchangeably throughout the text. METHOD Sample A sample of 506 college students from a major public university in Texas responded to a paper-and-pencil, self-administered questionnaire. Participants were enrolled in upper-level undergraduate health classes during the Spring and Summer of 1998; they volunteered for the study and signed an informed consent, prior to participation. The study was approved by the university’s Institutional Review Board and students were ensured anonymity of responses. The sample consisted mainly of female students (61.9%) with a mean age of 25.21 years (SD = 5.85). One unique characteristic of this group was the large representation of Hispanic students. The sample was almost evenly split between Anglo (46.8% of Non-Hispanic/White) and Hispanic (41.6% of Mexican American/Hispanic/Latino). Sixty-three percent (63.6%) of students were single, whereas 21.1% were married. In terms of religious preference, half the sample (50.0%) was Catholic and 28.9% was Protestant; 42.4% of students with a religious preference considered themselves religiously moderate, 25.7% liberal, and nearly a fourth of the sample classified itself as conservative/fundamentalist. Measures The instrument used in this study was developed by the authors and tested for reliability and validity with the same sample described here. Description of the development of the instrument and its psychometric characteristics is presented elsewhere (Goodson et al., 2000). The questionnaire was designed to assess college students’ practices and attitudes when utilizing the Internet for three purposes: (a) obtaining information related to sexuality (for school, work-related projects, or personal information); (b) establishing and maintaining relationships (such as using e-mail or participating in chat-groups), and (c) sexual entertainment (viewing of sexually explicit materials). In addition to questions about e-mail and Internet use, the instrument contained items measuring practices, and outcome expectations and expectancies for the three functions just described. All outcome expectations and expectancy scales demonstrated appropriate internal consistency (Cronbach αs ranging from .76 to .95) and temporal stability over a 2-week period (Pearson rs ranged from .69 to
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
106
8:57
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
.78). The scales were also factor-analyzed; the resulting factor structure accounted for 68.8% of the variance. For the purposes of this paper, results relating only to the behavioral and attitudinal items in the third section of the questionnaire (use of the Internet for sexual entertainment) are presented. RESULTS E-Mail and Internet Use Seventy-four percent (74.4%) of respondents said they used e-mail and 90.1% used the Internet. Usage of both media was, however, not very frequent for the majority of the respondents: 40% of e-mail users said they used it less than once a week, and 73% of Internet users gave the same response. Twenty-nine percent of students who use e-mail said they did so on a daily basis; only 4% of users accessed the Internet daily. Male and female students did not differ significantly either in their e-mail and Internet use (χ 2 = 0.429, 1 df, p < .29 for e-mail use; χ 2 = 0.878, 1 df, p < .22 for Internet use) or regarding the frequency of use (χ 2 = 0.837, 2 df, p < .66 for e-mail; χ 2 = 0.539, 2 df, p < .76 for Internet). Although differences were not statistically significant, more male students declared they did not use email or Internet. (Table I). Similar to results in a study by Al-Khaldi and Al-Jabri (1998), academic rank was also not associated with e-mail or Internet usage. Use of e-mail and Internet was a recent phenomenon for most of our sample: 31.9% of e-mail users said they had begun using e-mail within the last 2–3 years; 26.6% had begun using it within the previous 6–12 months. Only 8.1% of students had used e-mail for more than 3 years. Regarding Internet use, 38.4% said they had begun using this medium within the last 2–3 years, whereas 34.3% had begun using it in the previous 6–12 months. Similar to e-mail users, only 8.0% said they had begun using the Internet over 3 years prior to the interview. Female and male students did not differ significantly in their history of e-mail or Internet use (χ 2 = 4.655, 4 df, p < .32 for e-mail; χ 2 = 5.659, 4 df, p < .23 for Internet). When asked from where they usually log on to e-mail or Internet, almost half of users (47.4% of e-mail users and 46.2% of Internet users) said they logged on from a campus computer. Men and women, once again, did not differ significantly in this category (χ 2 = 0.179, 3 df, p < .98 for e-mail log-on and χ 2 = .4.910, 3 df, p < .18 for Internet; Table I). Among other user characteristics examined, similar numbers of male and female students (64.2% and 70.2%, respectively) said they used e-mail to communicate with family and friends. Although 43.3% of Internet users said they used it to search for sex-related information, this answer was given more frequently by males (56.5%) than by females (35.2%), and the difference was statistically significant (χ 2 = 22.009, 1 df, p < .001). Twenty-eight percent of Internet and e-mail users (28.7%) reported having established new friendships over the Internet/e-mail; the distributions were similar for men and women (Table I).
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Searching for Sexually Explicit Materials Online: User Behavior and Attitudes
107
Table I. Percentage Distribution and Chi-Square Statistics for Male and Female Students According to Characteristics and Purposes of E-Mail and Internet Usage
Use of e-mail Uses Doesn’t use Frequency of use Daily 2–3 times a week
3 years Source of log-on Campus Home Work Friend/relative’s Use of Internet Uses Doesn’t use Frequency of use Daily 2–3 times a week 3 years Source of log-on Campus Home Work Friend/relative’s Purposes of Use Uses e-mail and Internet to communicate with family and friends. Uses Internet to search for sex-related information. Has made new friends over e-mail and Internet. a Data on gender ∗∗ p < .01.
Males (n = 191)
Females (n = 313)
Total sample (N = 504)a
72.8 27.2
75.4 24.6
30.4 28.3 41.3
χ2
df
p
74.4 25.6
.429
1
.290
27.5 32.6 39.8
28.6 31.0 40.4
.837
2
.658
11.0 28.3 35.3 5.8
15.3 25.5 29.9 9.5
13.7 26.6 31.9 8.1
4.655
4
.325
47.1 43.5 6.5 2.9
47.6 44.2 5.6 2.6
47.4 43.9 5.9 2.7
.179
3
.981
88.5 11.5
91.1 8.9
90.1 9.9
.878
1
.216
4.7 24.8 70.5
3.9 21.9 74.1
4.2 23.0 72.8
.539
2
.764
11.7 35.0 43.6 7.4
19.1 33.9 35.4 8.3
16.4 34.3 38.4 8.0
5.659
4
.226
50.9 41.7 3.7 3.7
43.5 43.5 4.7 8.3
46.2 42.8 4.3 6.6
4.910
3
.179
64.2
70.2
67.9
1.940
1
.169
56.5
35.2
43.3
22.009
1
.001∗∗
30.2
27.7
28.7
.335
1
.610
are missing for 2 subjects (original N = 506).
Use of Internet for Sexual Entertainment Although 56.4% of the students had never accessed sexually explicit materials on the Internet, 43.5% had done so. However, only a small percentage (2.9%)
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
108
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
of respondents reported accessing these materials “frequently”. Fourteen percent (14.0%) reported “sometimes,” and 26.6% of the students said they “rarely” accessed these materials. Males were significantly more likely to have accessed the materials either sometimes or frequently (Table II). Students who accessed explicit materials were similar to those who did not in terms of age, ethnicity, or marital status. They were also comparable regarding religious preference, religiosity, academic status, and having taken a sexuality course (analyses not shown). Curiosity about sex represented the most frequently acknowledged motivation for accessing sexually explicit web sites (29.6% of students chose the frequently or sometimes categories) as compared to doing it to become sexually aroused (19.0%) or to enhance their sex lives with offline partners (13.3%—Table II). More male students than female students reported accessing sex-explicit materials because they were curious about sex and wanted to become sexually aroused ( p < .01). No gender differences related to enhancing students’ sex life with their offline partners were found (Table II). Over a third of the respondents said they accessed sexually explicit materials on the Internet alone (35.4% had done it frequently or sometimes); 17.8% did it with their offline partner(s), whereas 14.6% searched the Internet within a group context. More students had been silent observers in sexually explicit chat-groups (25.8%) than had actively participated in discussions (12.8%). Five percent of respondents (5.1%) said they had sometimes posted objections to distasteful sexually explicit chat-group conversations, and 15.8% of students said they had felt sexually harassed during these interactions. Significantly more female students than male students reported this complaint (χ 2 = 11.99, 3 df, p = .007). Only 1.5% of respondents declared they had frequently or sometimes paid to register for sexually explicit websites. Most students (73.7%) had never used the Internet to window-shop or browse for sexual paraphernalia such as sex toys, videotapes, clothing, lingerie, condoms, and oils. Close to five percent (4.7%) had used the Internet to order or purchase sexual paraphernalia. Male and female students did not significantly differ in any of these behaviors (Table II). Students who accessed sexually explicit materials on the Internet were also asked about specific sexual behaviors while online. Among the behaviors investigated, masturbation was reported by 15.0% of students, and “cyber sex with an online partner” by 12.1%. Among those students who said they had had cyber sex with an online partner (N = 23), drinking and using drugs while having online sex was mentioned by 8 and 3 students, respectively (Table II). Although no significant differences were found between men and women in any of these behaviors, larger numbers of male students reported masturbating while online. Slightly more women, however, reported having had cyber sex with an online partner. Men were more likely to report using alcohol and drugs while having cyber sex online. However, these differences were not significant (Table II).
Accessed SEM Accessed sexually explicit materials on the Internet (either accidentally or intentionally). Reasons for accessing SEM Curiosity Sexual arousal To enhance sex life with offline partner(s). Behavior while accessing SEM Accesses alone Accesses with offline partner(s). Accesses with a group of people. Posts messages to sexually explicit chat-groups on the Internet. Observes sexually explicit chat-groups on the Internet (does not participate). Posts objections to sexually explicit chat-group conversations, when finds them distasteful. Subscribes to sexually explicit websites (pays for registering).
Sometimes
Rarely
Never
0.0 0.0 1.1
0.0
1.1
3.0
109 0.0 0.0
0.0
0.0
0.5
2.0
3.0
0.0
7.4
10.5
7.4
7.5
1.0
5.1
8.8
5.6
13.0
19.8 14.7
25.0 15.9 11.8
10.0
8.0
17.2
11.0
25.0
28.3 25.3
31.0 33.3 18.0
30.8
3.2
6.3
16.8
3.2
19.4
28.0 15.2
25.0 15.6 12.6
24.1
6.7
7.2
17.0
7.2
22.3
28.1 20.4
28.1 24.6 15.4
26.6
88.0
89.0
75.8
85.0
54.0
28.3 59.6
30.0 43.4 67.0
40.8
95.8
86.3
71.6
89.5
73.1
45.2 64.1
55.2 69.8 75.8
66.0
91.8
87.7
73.7
87.2
63.2
36.5 61.8
42.3 56.4 71.3
56.5
7.45 3 .059 5.52 3 .137
195
6.69 3 .082
2 .360
1.82 3 .610
5.26 2 .072
195 2.04
194
195
193 10.34 3 .016
192 191
196 14.55 3 .002∗∗ 195 15.41 3 .001∗∗ 195 6.38 3 .094
451 36.22 3 .000∗∗
8:57
0.0
7.1
4.0
18.0
15.1 15.2
17.7 13.5 8.4
14.0
Pb
January 25, 2001
0.5
0.0
1.6
24.2 14.1
32.0 18.2 15.0
8.9
df
PP016-291179
15.6 3.1
4.6 3.1 1.5
22.5
χ2
P2: FVI/GAY
11.8 5.4
2.1 1.0 3.2
7.0 5.1 0.0
2.9
Na
Archives of Sexual Behavior [asb]
19.2 1.0
1.1
5.9
Total Total Total Total Males Females sample Males Females sample Males Females sample Males Females sample
Frequently
Table II. Percentage Distribution of Behaviors, Perception of Sexual Harassment, and Preferences of Male and Female Students when Accessing Sexually Explicit Materials (SEM) on the Internet
P1: FZN/FNN/ZCE QC: FTK Style file version Nov. 19th, 1999
Rarely
Never
110 1.1 4.4
7.5 7.6
1.1
0.0
0.0
1.1
0.0
12.5
0.0
1.0
6.0
16.7
14.4 24.7
22.4
4.5
4.3
5.6
4.2
5.7
2.6
31.5
39.8
0.0
33.3
7.0
6.3
11.0
2.0
15.0
the criterion for significance.
32.6
30.1
0.0
7.1
10.5
5.3
3.2
2.1
13.5
22.2
22.2
7.1
14.3
12.6
8.5
6.5
1.1
6.5
26.9
31.1
4.5
21.7
9.7
7.4
8.8
1.6
10.9
28.3
22.6
75.0
44.4
92.0
90.6
80.0
93.9
67.0
56.7
62.2
92.9
78.6
75.8
85.1
90.3
96.8
80.6
42.3
42.1
86.4
65.2
84.1
87.9
85.0
95.3
73.6
2.05 3 .560
4.59 3 .204
1.42 3 .699
5.44 3 .142
4.25 3 .235
5.64 3 .130
182 15.56 3 .001∗∗
183 30.92 3 .000∗∗
22
23
195 11.99 3 .007∗∗
190
193
193
193
8:57
4.4
12.5
0.0
1.0
3.1
8.0
3.0
10.8
Pb
January 25, 2001
4.5
8.7
0.5
0.5
0.5
0.5
16.0
df
PP016-291179
0.0
0.0
1.0
2.1
χ2
P2: FVI/GAY
22.2
2.2
2.0
Na
Archives of Sexual Behavior [asb]
a Ns include only students who had ever accessed SEM online. b Due to the large number of comparisons, α = .01 was used as ∗∗ p < .01.
Uses the Internet to window-shop/browse/view sexual paraphernalia. Uses the Internet to order/ purchase sexual paraphernalia. Masturbates while viewing sexually explicit web sites Engages in cyber sex with an online partner. Perception of sexual harrassment Has felt sexually harassed during online conversations. Preferences Likes to drink alcoholic beverages while having cyber sex with an online partner(s). Likes to use stimulants (drugs) while having cyber sex with an online partner(s). Likes to view other types of sexually explicit materials such as adult magazines. Likes to view other types of sexually explicit materials such as adult videotapes.
Sometimes
Total Total Total Total Males Females sample Males Females sample Males Females sample Males Females sample
Frequently
Table II. (Continued )
P1: FZN/FNN/ZCE QC: FTK Style file version Nov. 19th, 1999
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Searching for Sexually Explicit Materials Online: User Behavior and Attitudes
111
Students were also asked about their regard for other types of sexually explicit materials such as adult magazines or adult videotapes. Twenty-six percent (26.8%) of students said they liked to view adult magazines either frequently or sometimes; 30.7% gave the same response concerning adult videotapes (Table II). Surprisingly, the association between using the Internet to access sexually explicit materials and appreciating adult magazines or videotapes or both was only moderately correlated (r = .49 for magazines and .50 for videos). All correlations, nevertheless, were significant at the .01 level of probability. Males and females differed significantly in their liking of adult magazines and videotapes. Men were more likely to say they liked viewing either magazines or videos frequently or sometimes (χ 2 = 30.92, 3 df, p = .0001 for magazines; χ 2 = 15.56, 3 df, p = .001 for videos). Outcome Expectations and Expectancies Six items in the questionnaire formed a scale to measure students’ outcome expectations (beliefs about specific outcomes) related to the use of the Internet for accessing sexually explicit materials. Items were scored on a Likert-type 5point scale, anchored by 1 (strongly disagree) and 5 (strongly agree). Scores on the scale ranged from 6 to 30, with higher scores indicating stronger beliefs in the outcomes experienced when using the Internet for sexual entertainment (e.g., “Sexually explicit materials on the Internet satisfy my curiosity about sex” and “Make me sexually aroused”). The sample’s mean score was 15.73 (SD = 5.76). This score places the average of the group below the theoretical midpoint of the scale, which was 18, indicating overall disagreement or weak beliefs concerning the specified outcomes (The Appendix presents the items for each of the scales). More than half of the sample (58.6%) scored below the theoretical midpoint of the scale. One-way Analysis of Variance was used to examine potential differences between mean scores on the expectation scale for males (M = 16.43, SD = 5.48) and females (M = 14.96, SD = 5.99). Results did not indicate any significant difference (F = 3.124, p = .079). Another set of six questions (Appendix) measured students’ expectancies or the personal importance they attributed to the outcomes specified in the expectation items. Items were also scored on a Likert-type scale with responses ranging from 1 (Not important at all) to 5 (Extremely important). Scores on the scale ranged from 6 to 30 for females, and from 6 to 25 for males (i.e., none of the male students rated all six items as “extremely important”). Higher scores on the expectancies scale reflected a higher degree of importance attributed to the expected outcomes of searching the Internet for sexually explicit materials. The mean score for the sample was 12.85 (SD = 5.60), also below the scale’s midpoint of 18. More than two-thirds of students (76.6%) had mean scores below the theoretical midpoint of the scale. On average, students appeared not to value highly the outcomes of searching the net for sexually explicit materials. Male (M = 13.19, SD = 5.26)
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
112
8:57
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
and female students (M = 12.49, SD = 5.96) were, once again, similar in this valuation (F = .744, p = .389). A moderate-to-high association between outcome expectations, expectancies, and use of Internet for sexually explicit materials was observed. For the combined sample, use and outcome expectations exhibited a correlation of .43 ( p < .01); expectancies was also significantly correlated with use (r = .41, p < .01); outcome expectations and expectancies were also significantly associated (r = .62, p < .01). DISCUSSION This study examined attitudes and behaviors of undergraduate students from a large, public university in the United States, concerning use of the Internet to access sexually explicit materials. In addition, the question of whether gender differences were present among these behaviors and attitudes was also considered. Primary findings indicated that the majority of the sample consisted of infrequent and relatively new users of the Internet. Almost half of the users logged on to the Internet from a campus computer lab. More than 40% had sometimes accessed sexually explicit materials through the Internet, but the practice was not very frequent. This study also uncovered that men were more likely to have accessed the Internet for sexually explicit materials and were more likely to claim curiosity about sex as their motivation for this behavior. Women, however, were more likely to have experienced sexual harassment while online. In terms of attitudes, this sample did not appear to value highly or have strong beliefs about the potential outcomes associated with accessing the Internet for sexually explicit materials. This study documented that 90% of students used the Internet, and 46.2% of Internet users logged on from a campus computer. No differences were found between male and female students in their Internet and e-mail use, in their frequency of use, or in their source for logging on. Similarly, data from the U.S. Census Bureau indicate that 62.3% of school-enrolled adults in the United States use a computer at school, and differences between men and women in terms of computer use are shrinking considerably (Newburger, 1999). Another study by Brenner in 1997 focusing on Internet interference with aspects of users’ daily lives found no significant differences between men and women in terms of time spent online (Brenner, 1997). Scherer’s examination of a sample of college students documented that 73% of the students used the Internet at least once a week, whereas our study found that 72.8% used it less than once a week (Scherer, 1997). Moreover, a recent study comparing Internet usage in two college campuses (one private and one public) verified that 93% of the sample had Internet access, and mean time spent online was 4.66 per week. Even though the study by Davis and colleagues found that men spent significantly more time online than women, their findings seem unique for a college sample (Davis et al., 1999).
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Searching for Sexually Explicit Materials Online: User Behavior and Attitudes
113
Our study also documented that men were significantly more likely than women to access sexually explicit materials on the Internet. Nevertheless, among our sample 10% of the women declared they had either frequently or sometimes accessed these materials (28.4% of men gave the same replies). Among the women in our sample who used the Internet for sexual entertainment, 15.5% said they either frequently or sometimes liked to view other types of sexually explicit materials such as adult magazines, and 21.1% enjoyed adult video tapes. In a study examining female college students and their reading of pornography, Walsh found that 25% of his undergraduate female sample were readers of pornography (Walsh, 1999). We could speculate that the smaller numbers found in our study are a function of our sample’s demographics: a larger percentage of Hispanics and Catholics (as compared to related studies). Although our finding that students who accessed the Internet for sexual purposes did not differ significantly in terms of religious preference, religiosity, or ethnicity appears to contradict this speculation, our results may be contingent upon a small sample size. Future studies with larger samples may be able to better determine if there is an association among ethnicity, religiosity, and accessing sexually explicit materials online. Some of the findings were contrary to our expectations, in view of current theories and data on behaviors and attitudes. We were surprised, for instance, with the absence of significant differences in frequency of masturbation while viewing sexually explicit materials online. Reported incidence of masturbation appears to remain one of the largest gender differences related to sexual behavior documented in the social scientific literature. A meta-analysis of gender differences in sexuality analyzed 117 studies and 239 independent samples. The analysis revealed that the largest difference reported across the years was related to incidence of masturbation, with men reporting significantly higher frequency than women (Oliver and Hyde, 1993). Could the absence of differences in our study reflect a change in behavior among women? Or does it represent underreporting by male students? Additional investigations, with larger samples, are needed to clarify this point. Another unexpected absence of significant differences was related to having had cyber sex with an online partner. Despite the lack of statistical difference, however, it is worth noting that women declared having performed this behavior frequently or sometimes in larger numbers than men. This finding appears to echo those from a recent study by Parks & Floyd of users of online newsgroups (Parks and Floyd, 1999). These authors found that significantly more women than men had started a personal relationship in a newsgroup with many of these relationships evolving into offline friendships and partnerships. Could cyber sex be perceived as a form of relationship that women are more willing to explore, given the anonymity and relative absence of physical harm provided by the Internet? In terms of students’ attitudes, the low outcome expectations and expectancies mean scores were somewhat surprising. We expected students to be enthusiastic about a new, easily accessible, anonymous, and inexpensive format for sexual
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
114
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
entertainment. Their not-so-positive attitudes could, however, be caused precisely by the novelty of the medium. As most students were relatively new users of the Internet, their scores on the outcome expectations and expectancies scales could very well reflect their uncertainty about, and unfamiliarity with, this new vehicle of information. Also surprising was the lack of statistically significant differences in men’s and women’s attitudes. Traditionally, females have had less favorable attitudes toward sexually explicit materials. A study of pornography as a source of sex information for college students (Li and Davey, 1996) found that 51.1% of the sample held negative attitudes toward pornography; males held more positive views than females. In our study, mean scores on the outcome expectation and expectancy scales reflected little enthusiasm for the outcomes of using the Internet for sexual entertainment. These results appear to coincide with those registered by Li and Davey (Li and Davey, 1996), but contradict the findings from a study of changes in attitudes toward sexually explicit materials in 10 states nationwide (Winick and Evans, 1994). In this latter investigation, more than half of the sample exhibited tolerant, positive views of these materials. Similar to our data, however, authors of the latter survey found no differences in attitudes between men and women. Our finding concerning the moderate-to-high association between outcome expectations, expectancies and use of the Internet for sexually explicit materials also appears to be corroborated by findings from a study by Al-Khaldi and colleagues carried out with Saudi Arabian college students (Al-Khaldi and Al-Jabri, 1998). When examining the relationship of attitudes to computer utilization (in general, not specifically for sex-related purposes), the researchers found that among a sample of 300 undergraduate business majors, attitudes toward computers exhibited a correlation of .42 with computer use ( p < .05). The correlations found between attitudes (outcome expectations and expectancies) and use of the Internet for sexual entertainment, in our study, were very similar: .43 and .42 for outcome expectations and expectancies, respectively. Despite the fact that our research questions did not specifically address concerns with Internet addiction in general, or with sex-related Internet addiction in particular, our data do not reveal behaviors or attitudes that signal concern. The study by Cooper and colleagues—examining pathological expressions of Internet use for sexual purposes within a sample of Internet users—found only 8% of respondents exhibiting sexually compulsive features (Cooper et al., 1999). In our study, only 2.9% of the students said they frequently accessed sexually explicit materials on the Internet. Of potential concern for college administrators is the use of campus computers for sexual purposes. In our study more students accessed sexually explicit materials from home than from the campus computer. About 1 in 10 campus computer users, however, logged on to sexually explicit websites from the university computers either sometimes or frequently. Among those who access the Internet
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Searching for Sexually Explicit Materials Online: User Behavior and Attitudes
115
from home, about 1 in every 4 students searched for sexually explicit materials (analyses not shown). Percentage-wise the numbers may be small and insignificant at this point. Nevertheless, Cooper’s original worry is worth considering: if these percentages are transformed into absolute numbers and generalized to the college student population nationwide, the data assume a more problematic dimension and may have implications for school administration and public policy. Anecdotal evidence indicates that many students resent having their mandatory computer access fees used to facilitate other students’ access to sexual entertainment. Our assessment of college students’ behaviors and attitudes was meant exclusively as an exploratory study. Because so little is known about users’ practices and beliefs, we did not wish to assume similarities between the Internet and other media in order to propose hypotheses. The descriptive nature of this study may well represent one of its major limitations, as many of the questions of why these findings occurred will require further research. On the other hand, we believe that our use of a theoretical framework to guide this exploration represents one of its major strengths. As has been noted throughout this study, little is known about Internet users when they use this medium for sexual entertainment purposes. The data presented here—albeit not generalizable to the U.S. college population—alert us to the need for better understanding the impact of the Internet upon users. Such understanding cannot be achieved, however, without an analysis of how users’ intrapersonal factors and environmental contexts mediate this impact. Even though the Internet appears to be, in itself, very powerful, psychosocial factors determine how users handle, process, and replicate the information available on the Internet. Understanding which mechanisms determine and shape users’ seeking of, interest in, attention to, and learning from Internet information is paramount to understanding the impact the Internet will have (is already having) on the social construction and dissemination of sexual knowledge. Now that we know a bit more about users’ characteristics, further research is needed to understand these mechanisms.
APPENDIX: OUTCOME EXPECTATIONS AND EXPECTANCIES QUESTIONNAIRE ITEMS Outcome expectations: Sexually explicit materials on the Internet . . . 1. 2. 3. 4. 5. 6.
Are a way of learning new sexual techniques. Help improve my sexual relationships offline. Are a way of fulfilling my sexual fantasies. Stimulate my sexual fantasies. Make me sexually aroused. Satisfy my curiosity about sex.
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
116
8:57
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
Expectancies: How important is it to you to use the Internet . . . 1. 2. 3. 4. 5. 6.
To learn new sexual techniques? To improve your sexual relationships offline? To fulfill your sexual fantasies? To stimulate your sexual fantasies? To become sexually aroused? To satisfy your curiosity about sex?
ACKNOWLEDGMENT This study was partially funded by a Faculty Small Grant from The University of Texas at San Antonio, awarded to the first author in 1998. REFERENCES Acevedo, E., Delgado, G., and Segil, E. (1998). INPPARES uses Internet to provide Peruvians with sexuality information and counseling. SIECUS Rep. 26: 14. Al-Khaldi, M. A., and. Al-Jabri, I. M. (1998). The relationship of attitudes to computer utilization: New evidence from a developing nation. Comput. Hum. Behav. 14: 23–42. Allen, M., D’Alessio, D., and Brezgel, K. (1995a). A meta-analysis summarizing the effects of pornography II: Aggression after exposure. Human Comm. Res. 22: 258–283. Allen, M., Emmers, T., Gebhardt, L., and Giery, M. A. (1995b). Exposure to pornography and acceptance of rape myths. J. Comm. 45: 5–26. Ballard, S. M., and Morris, M. L. (1998). Sources of sexuality information for university students. JSET 23: 278–287. Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory, Prentice Hall, Englewood Cliffs, NJ. Bandura, A. (1994). Social cognitive theory of mass communication. In Bryant, J., and Zillmann, D. (eds.), Media Effects: Advances in Theory and Research, Hillsdale, NJ, Lawrence Erlbaum Associates. Barak, A., and Fisher, W. A. (1997). Effects of interactive computer erotica on men’s attitudes and behavior toward women: An experimental study. Comput. Hum. Behav. 13: 353–369. Baranowski, T., Perry, C. L., and Parcel, G. S. (1997). How individuals, environments and health behavior interact. In Glanz, K., Lewis, F. M., and Rimer, B. K. (eds.), Health Behavior and Health Education: Theory, Research, and Practice, Jossey-Bass, San Francisco. Becker, J., and Stein, R. M. (1991). Is sexual erotica associated with sexual deviance in adolescent males? Int. J. Law Psychiatry 14: 85–95. Binik, Y. M., Mah, K., and Kiesler, S. (1999). Ethical issues in conducting sex research on the Internet. J. Sex Res. 36: 82–90. Brenner, V. (1997). Psychology of computer use: XLVII. Parameters of Internet use, abuse and addiction: The first 90 days of the Internet Usage Survey. Psychol. Rep. 80: 879–882. Bryant, J., and Zillmann, D., eds. (1994). Media Effects: Advances in Theory and Research, Lawrence Erlbaum Associates, Hillsdale, NJ. Carey, R. (1996). Betwixt and between: An organization’s relationship with online communications. SIECUS Rep. 25: 8–9. Cart, C. U. (1997). Online computer networks: Potential and challenges for community organizing and community building now and in the future. In Minkler, M. (ed.), Community Organizing & Community Building for Health, Rutgers University Press, New Brunswick.
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Searching for Sexually Explicit Materials Online: User Behavior and Attitudes
117
CommerceNet. (1999). CommerceNet and Nielsen Media Research issue results of Spring 1999 Internet Demographic Survey. Available: http://www.commercenet.com. Cooper, A. (1998). Sexuality and the Internet: Surfing into the new millennium. Cyberpsych. Behav. 1: 181–187. Cooper, A., Scherer, C. R., Boies, S. C., and Gordon, B. L. (1999). Sexuality on the Internet: From sexual exploration to pathological expression. Prof. Psycho. 30: 154–164. Cooper, A., and Sportolari, L. (1997). Romance in cyberspace: Understanding online attraction. JSET 22: 7–14. Davis, S. F., Smith, B. G., Rodrigue, K., and Pulvers, K. (1999). An examination of Internet usage on two college campuses. Coll. Stud. J. 3: 257–260. DeGuzman, M. A., and Ross, M. W. (1999). Assessing the application of HIV and AIDS related education and counseling on the Internet. Patient Educ. Couns. 36: 209–228. Fisher, W. A., and Barak, A. (1991). Pornography, erotica, and behavior: More questions than answers. Int. J. Law Psychiatry 14: 65–83. Goodson, P., McCormick, D., and Evans, A. (2000). Sex and the Internet: A survey instrument to assess college students’ behavior and attitudes. CyberPsych. Behav. 3(2): 129–149. Haffner, D. W., and Kelly, M. (1987). Adolescent Sexuality in the Media. SIECUS Rep. March/April: 9–12. Harris, R. J. (1994). The impact of sexually explicit media. In Bryant, J., and Zillmann, D. (eds.), Media Effects: Advances in Theory and Research, Lawrence Erlbaum Associates, Hillsdale, NJ. Henry, S. (1999, August 26). Health officials get web help; Gay site brings syphillis warning to California men at risk. The Washington Post, p. E08. Koch, P. B. (1998). Sexual knowledge and education. In Francouer, R. T., Koch, P. B., and Weis, D. L. (eds.), Sexuality in America: Understanding our sexual values and behavior. New York, Continuum. Koop, C. E. (1987). Report of the Surgeon General’s workshop on pornography and public health. Am. Psychol. 42: 944–945. Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T., and Scherlis, W. (1998). Internet paradox: A social technology that reduces social involvement and psychological well-being? Am. Psychol. 53: 1017–1031. Kutchinsky, B. (1991). Pornography and rape: Theory and practice? Evidence from crime data in four countries where pornography is easily available. Int. J. Law Psychiatry 14: 47–64. Li, Q., and Davey, M. R. (1996). Pornography as a source of sex information for college students in fraternities and sororities. J. Health Ed. 27: 165–169. Newburger, E. C. (1999). Computer use in the United States. October 1997. Current Population Reports, U.S. Census Bureau, Pages 1–11. Available: http://www.census.gov. Newman, B. (1997). The use of online services to encourage exploration of ego-dystonic sexual interests. JSET 22: 45–48. Oliver, M. B., and Hyde, J. S. (1993). Gender differences in sexuality: A meta-analysis. Psychol. Bull. 114: 29–51. Parks, M. R., and Floyd, K. (1999). Making friends in Cyberspace. J. Comput. Mediated Comm. 1: 1–12. Parks, M. R., and Roberts, L. D. (1998). ‘Making MOOsic’: The development of personal relationships online and a comparison to their offline counterparts. J. Social Pers. Rel. 15: 517–537. Portelli, C. J., and Meade, C. W. (1998). Censorship and the Internet: No easy answers. SIECUS Rep. 27: 4–8. Roffman, D. M., Shannon, D., and Dwyer, C. (1997). Adolescents, sexual health and the Internet: Possibilities, prospects, and challenges for educators. JSET 22: 49–55. Ross, M. W., Tikkanen, R., and Mansson, S. A. (2000). Differences between Internet samples and conventional samples of men who have sex with men: Implications for research and HIV interventions. Soc. Sci. Med. 47. Scheerhorn, D., Warisse, J., and McNeilis, K. S. (1995). Computer-based telecommunication among an illness-related community: Design, delivery, early use, and the functions of HIGHnet. Health Commun. 7: 301–325. Scherer, K. (1997). College life online: Healthy and unhealthy Internet use. J. Col. Stud. Devel. 38: 655–665.
P1: FZN/FNN/ZCE
P2: FVI/GAY
Archives of Sexual Behavior [asb]
118
QC: FTK PP016-291179
January 25, 2001
8:57
Style file version Nov. 19th, 1999
Goodson, McCormick, and Evans
Scott, J. E. (1991). What is obscene? Social science and the contemporary community standard test of obscenity. Int. J. Law Psychiatry 14: 29–45. Searchterms.com. (1999). The Top Ten. Available: http://www.searchterms.com/index/html. Tamosaitis, N. (1995). Net.Sex, Ziff-Davis Press, Emeryville, CA. Walsh, A. (1999). Life history theory and female readers of pornography. Pers. Indiv. Dif. 27: 779– 787. Weinrich, J. D. (1997). Strange bedfellows: Homosexuality, gay liberation, and the Internet. JSET 22: 58–66. Winick, C., and Evans, J. T. (1994). Is there a national standard with respect to attitudes toward sexually explicit media material? Arch. Sex. Behav. 23: 405–419. Zillmann, D., and Bryant, J. (1988). Effects of prolonged consumption of pornography on family values. J. Fam. Issues 9: 518–544.
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Archives of Sexual Behavior, Vol. 30, No. 2, 2001
Extra Relationship Involvement Among Women: Are They Different From Men? Sophie Banfield, M.Psych.,1 and Marita P. McCabe, Ph.D.1,2
Factors related to three types of Extra Relationship Involvement (ERI) in women were explored: emotional ERI, sexual ERI, and a combination of sexual and emotional ERIs . A model, based on a decision-making model of male ERI and the additional variable of commitment, was evaluated. The research involved two studies with female participants (N = 112, N = 44) who had been involved in heterosexual relationships of at least 6-month duration. The major finding was that women engage in emotional and combined ERIs but rarely enter into solely sexual ERIs. It was demonstrated that social norms, planning, relationship satisfaction, and commitment were influential in predicting emotional and combined ERI intentions. Past ERI behavior was a strong predictor of future emotional and combined ERI behavior, but planning also added to the prediction of combined ERI behavior. Women who had engaged in emotional ERIs and combined ERIs indicated romantic affect as the main reason for their ERI behavior. Overall, it was demonstrated that women’s intentions to engage in ERI were related to cognitive processes and relationship variables, and that ERI behavior, although generally habitual, was also predicted by cognitive processes. KEY WORDS: extrarelationship sex; female; relationship qualities; theory of planned behavior; nonmonogamy.
INTRODUCTION Intimate relationships involve activities and emotions that are generally expected to be restricted to the primary relationship (Roscoe et al., 1988; Weis and Felton, 1987). When these activities are perceived to overstep the boundaries of exclusivity and expectations of monogamy, they constitute an Extra Relationship 1 School
of Psychology, Deakin University, Burwood, Australia. whom correspondence should be addressed at School of Psychology, Deakin University, 221 Burwood Highway, Burwood 3125, Australia; e-mail: [email protected].
2 To
119 C 2001 Plenum Publishing Corporation 0004-0002/01/0400-0119$19.50/0 °
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
120
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
Involvement (ERI; Drake and McCabe, in press). ERI research has largely concentrated on sexual intercourse as the major violation of the primary relationship, to the exclusion of emotional involvement (Reiss et al., 1980; Thompson, 1983; Wiederman, 1997). However, the emotional components of ERI have also been demonstrated to be important, especially when considering female ERI (Glass and Wright, 1985; Spanier and Margolis, 1983). Thompson (1984) suggested that there are three types of ERIs: sexual ERI (flirting to sexual intercourse), emotional ERI (close friendship to being “in love”), and a combination of sexual and emotional ERI (combined ERI). The aim of the current study was to explain why women may become involved in emotional, sexual, and combined ERIs. Despite widespread community disapproval of ERI (Glenn and Weaver, 1979; Reiss et al., 1980; Singh et al., 1976; Weis and Felton, 1987), there has been consistent evidence demonstrating that ERI is common in relationships of all types (Drake and McCabe, in press; Maykovick, 1976; Reiss et al., 1980; Spanier and Margolis, 1983; Thompson, 1983). Men report a greater incidence of sexual ERI than do women. Glass and Wright (1985) found that 44% of men, compared with 24% of women, had at least one ERI experience involving sexual intercourse. Furthermore, in a review of 12 studies, Thompson (1983) found that approximately 45% of males and 30% of females had experienced sexual intercourse at least once with someone outside their primary relationship. However, when emotional, sexual, and combined ERI incident rates are considered together, gender differences are reduced. Thompson (1984) found that 42.2% of females and 45.8% of males had engaged in an ERI. This finding suggests that male ERIs are more likely to be sexual, and female ERIs are more likely to be emotional (Glass and Wright, 1985; Reiss et al., 1980). Considerable research has focused on factors that contribute to attitudes held toward ERI behavior (e.g., Bell et al., 1975; Gangestad and Thornhill, 1997; Kelley, 1978; Maykovick, 1976; Meyering and Epling-McWherter, 1986; Reiss et al., 1980; Saunders and Edwards, 1984). However, most of this research has been based primarily on sexual behavior (e.g., Drake and McCabe, in press; Edwards and Booth, 1976). Schafer and Keith (1981) believed that the lack of research on emotional ERI has limited the understanding of important aspects of intimate relationships, especially factors influencing female ERI. Drake and McCabe (in press) proposed a decision-making model of male ERI, based on Ajzen’s (Ajzen, 1991) Theory of Planned Behavior (TPB), to explain ERI intentions and behavior. Drake and McCabe (in press) extended TPB by expanding the construct of perceived behavioral control to include three components: selfefficacy, control, and planning. They also included five additional variables: past behavior, affect, relationship satisfaction, sexual satisfaction, and love styles. These authors confirmed that, for males, the more positive the attitudes and social norms held toward ERI, the stronger the intentions to engage in this behavior. Although
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
121
they failed to demonstrate that self-efficacy and control predicted intentions or behavior among males, these variables will be retained in the current study in order to evaluate the predictive value of the TPB among women. Likewise, planning was retained in the current study although Drake and McCabe (in press) found that planning contributed to ERI intentions, but not ERI behavior among males. Drake and McCabe (in press) also included three variables, in addition to those contained within TPB, that they predicted would influence ERI intentions and behavior: relationship quality (relationship satisfaction, sexual satisfaction, love styles), affect, and past behavior. Relationship satisfaction and sexual satisfaction have both been shown to influence ERI intentions and behaviors (Glass and Wright, 1985; Johnson, 1970; Roscoe et al., 1988). Drake and McCabe (in press) found that relationship satisfaction and sexual satisfaction failed to predict the ERI intentions and behavior of males beyond that explained by TPB. However, they failed to examine the factors that contribute to emotional ERI. Glass and Wright (1977) have suggested that lower levels of marital satisfaction expressed by women may be due to higher level of emotional involvement with their ERI partner. Furthermore, though these variables failed to add to the prediction of male sexual ERI, they may have a significant influence on female sexual ERI. Drake and McCabe utilized two constructs from Lee’s (Lee, 1976, 1988) “typology of love” as predictors of ERI intentions and behavior: Eros (romantic, passionate love) and Ludus (game-playing love). Support was found for the influence of love styles on intentions, but not on behavior. Saunders and Edwards (1984) also found that individuals who were able to dissociate love from sex, which is indicative of the Ludus love style, were more likely to have favorable ERI attitudes. However, there is limited research that has assessed the influence that love styles have on ERI behaviors. It has been postulated that high levels of affect may cause people to make decisions based on their emotions (sexual and emotional arousal) rather than logic and reasoning (Denes-Raj and Epstein, 1994; Epstein, 1994). Therefore, decisions to engage in ERI based on affect may facilitate intended behaviors, or allow for previously unintended behaviors to occur. Drake and McCabe (in press) have suggested that the experience of sexual desire may be so overwhelming that in the heat of the moment normal decision-making processes may be bypassed, thus allowing previously unintended and undesired behaviors to occur. Research suggests that women do not tend to become sexually involved in the absence of emotional commitment, whereas men find it easier to have sexual intercourse without emotional involvement (Townsend, 1995). Therefore, there may be a difference in the type of affect experienced by women for sexual, emotional, and combined ERIs, and this warrants investigation. Past behavior has been shown to be a good predictor of intentions (Buunk and Bakker, 1995; Gallois et al., 1994; Terry and O’Leary, 1995) and behavior (Gallois et al., 1994; Godin et al., 1992; Kashima et al., 1993; Terry and O’Leary, 1995).
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
122
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
Drake and McCabe (in press) found that, after accounting for attitudes, social norms, and the components of perceived behavioral control, past ERI contributed unique variance to the prediction of both ERI intentions and behavior. Buunk and Bakker (1995) also found that an individual’s willingness to become involved in sexual intercourse with someone other than their primary partner was predicted, among other things, by the number of times they had previously been involved in such behaviors. Thus, past behavior will be included in the current study. Drake and McCabe’s (Drake and McCabe, in press) model of male ERI needs to be extended to ensure that it is relevant for females. The additional variable of commitment was included in the current study because of its relevance to women (Buunk and Bakker, 1997; Sprecher, 1998) and because it was expected to have a substantial influence on the three types of ERI. A widely used theory of commitment, which was developed within the tradition of social-exchange theory (Kelley and Thibaut, 1978), is Rusbult’s investment model (Rusbult, 1980, 1983). According to the investment model, commitment is a subjective state that represents a long-term orientation, including feelings of psychological attachment to a partner and desire to maintain a relationship, through the good and bad times (Johnson and Rusbult, 1989; Rusbult, 1980, 1983; Rusbult and Buunk, 1993). The limited research that has utilized the investment model to predict ERI has demonstrated the value of commitment in the prediction of ERI intentions. Buunk and Bakker (1997) found, for both men and women, that the less committed an individual was to his/her relationship, the more likely he/she was to have intentions of engaging in sexual ERI. Buunk and Bakker (1997) believed that lack of commitment makes an individual less concerned about the feelings of the partner, and thus more inclined to follow through with sexual ERI intentions. Further, those with low commitment may be searching for new partners to replace the current one. Although Buunk and Bakker (1997) demonstrated the relationship of commitment with sexual intentions, its contribution to sexual ERI behavior, and emotional ERI intentions and behavior, is yet to be determined. The present study was developed to explore the factors contributing to a woman’s decision to enter into emotional, sexual, or combined emotional and sexual ERIs. It was predicted that the greater the social support a woman perceives for engaging in an ERI and the more positive her ERI attitudes, the stronger will be her intentions to engage in an ERI. Further, it was predicted that if a woman believes it would be easy for her to engage in an ERI and she has made specific plans for initiating an ERI relationship, her intentions of engaging in an ERI would be greater. It was further predicted that sexual and relationship dissatisfaction, lack of commitment to the primary relationship, lack of an Eros love style and presence of a Ludus love style, as well as past ERI behavior, would predict greater ERI intentions. In terms of sexual behavior, it was predicted that the greater a woman’s ERI intentions, the more control she has over the ERI situation and the more she makes plans to initiate an ERI relationship, the greater the likelihood of her engaging in
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
19:35
Extra Relationship Involvement Among Women
Style file version Nov. 19th, 1999
123
ERI behavior. It was also predicted that sexual and relationship dissatisfaction, lack of commitment to the primary relationship, lack of an Eros love style and presence of a Ludus love style, and past ERI behavior, would predict greater involvement in ERI behavior. Further, it was predicted that women would report that they became involved in emotional ERIs for emotional affective reasons rather than sexual and irresponsible affective reasons, in sexual ERIs for sexual and irresponsible affective reasons rather than emotional affective reasons, and in combined ERIs for emotional and sexual affective reasons rather than irresponsible affective reasons. STUDY 1 Method Study 1 was designed to determine the variables that best predict ERI intentions. Participants A sample of 112 females, recruited through advertisements placed in local and statewide newspapers, completed the ERI questionnaire. The mean age of participants was 35.6 years (SD = 11.4, Range = 19–72). Fifty-six percent of participants were married, 28% were in de facto relationships or living together, and 28% were in steady relationships. All participants had been involved in their current relationship for a period of at least 6 months. Materials The ERI questionnaire comprised the following scales: 1. Attitudes: Participants rated their attitude to becoming (1) emotionally involved, (2) sexually involved, and (3) both emotionally and sexually involved with someone other than their regular partner. For example “I think that becoming emotionally involved with someone other than my regular partner within the next six months is . . . ” Participants rated six attitudes, such as “bad–good,” “unfavourable–favourable” and “immoral– moral” (Ajzen, 1991), on 7-point scales for each type of ERI (i.e., −3: bad, 0: undecided, 3: good). All scores were averaged to provide an attitude to ERI score for each type of ERI. In the current study, Cronbach’s alphas for the three types of ERI attitudes were .95, .94, and .95 respectively. 2. Social Norms: Single items were used to measure the extent to which participants perceived important people in their lives thought they should engage in (1) an emotional involvement, (2) a sexual involvement, and
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
124
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
3.
4.
5.
6.
(3) a combination of both emotional and sexual involvement, with someone other than their regular partner. For example, “Most people who are important to me think that I should . . . ” Participants rated the level of acceptance for each type of ERI they would receive from others on a 7-point scale (−3: should not become involved, 0: undecided, 3: should become involved). Planning, Control, and Self-Efficacy: Single items, developed by Drake and McCabe (in press), were used to measure planning, control, and selfefficacy for the three types of ERI behavior. Participants indicated their level of agreement on a 7-point scale for the following statements: for planning, “I have planned to . . . ”; for control, “I have control over whether I . . . ”; and for self-efficacy, “It is easy for me to . . . ” (−3: strongly disagree, 3: strongly agree). Eros and Ludus Love Styles: The Eros and Ludus subscales of the Love Attitude Scale (Hendrick and Hendrick, 1986) were used. Each subscale has seven items, which are measured on a 5-point Likert scale (1: strongly disagree, 5: strongly agree). Examples of items are “My partner and I really understand each other” and “I believe that what my partner does not know will not hurt him” for Eros and Ludus respectively. Hendrick and Hendrick (1986, 1989) have demonstrated construct and discriminant validity and reported Cronbach’s alphas of .7 and above for these subscales. In the current study, Cronbach’s alphas were .84 for the Eros scale and .76 for the Ludus scale. Relationship and Sexual Satisfaction: Respondents completed McCabe’s (McCabe, 1994) Relationship and Sexual Satisfaction Questionnaires. The scales consist of nine and six items respectively and measure the frequency of behaviors and attitudes related to these constructs on a 6-point Likert scale (0: not at all, 5: always). Examples of items are “How often does physical affection lead to sex when you are in bed?” and “Do you take an active role during sex?” for relationship satisfaction and sexual satisfaction, respectively. McCabe (1994) reported Cronbach’s alphas of .85 and .78 respectively for the two measures. In the current study Cronbach’s alphas were .85 for relationship satisfaction and .86 for sexual satisfaction. Commitment: The commitment scale contained items that measured the extent to which participants felt committed to their relationship and to their partner (Buunk and Bakker, 1997). The scale consisted of three items that were measured on a 5-point Likert scale: “To what extent do you feel attached to your partner?” (1: not at all, 5: extremely), “Do you feel committed to maintaining your relationship with your partner?” (1: definitely not, 5: definitely), and “How likely is it that you will end you relationship in the near future?” (1: very unlikely, 5: very likely). Rusbult (1983) reported Cronbach’s alphas of .82 for the scale. In the current study the Cronbach’s alpha was .82.
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
125
7. Past ERI Behavior: Participants indicated ‘Yes’ or ‘No’ to whether they had engaged in five levels of (1) an emotional ERI, (2) a sexual ERI, and (3) a combination of both emotional and sexual ERIs with someone other than their regular partner. An example of the items is “Over the last six months of your relationship which of the following behaviors did you engage in with someone other than your regular partner?” Emotional involvement ranged from “casual friendship” to “deep love relationship,” and sexual involvement ranged from “kissing” to “sexual intercourse.” 8. Intentions to Engage in ERI: Participants indicated their intentions to engage in five levels of (1) an emotional ERI, (2) a sexual ERI, and (3) a combination of both emotional and sexual ERIs with someone other than their regular partner. Emotional involvement ranged from casual friendship to deep love relationship, and sexual involvement ranged from kissing to sexual intercourse. An example of the items is “Within the next 6 months, which of the following do you intend to engage in with someone other than your regular partner?” Participants rated each type of involvement on a 7-point Likert scale that ranged from 1 (certainly not) to 7 (certainly yes). 9. Definition of an ERI: Participants were asked to specify emotional and sexual activities that they believed constituted an ERI. Emotional involvement ranged from casual friendship to deep love relationship, and sexual involvement ranged from kissing to sexual intercourse. Based on these ratings, behaviors were selected to represent ERIs for the current study. A 75% agreement level was required for behaviors to be classified as an ERI activity. Love relationships (86%) and deep love relationships (91%) were considered to constitute an ERI, whereas casual friendships (2%), close friendships (7%), and deep friendships (26%) were excluded. Petting (91%), sexual intimacy without intercourse (93%), and sexual intercourse (95%) were considered to constitute an ERI, whereas kissing (64%) and hugging and caressing (74%) were excluded. Due to the similarity between petting and sexual intimacy without intercourse, they were incorporated into one variable (petting and sexual intimacy without intercourse) by obtaining an average of the responses to both variables.
Procedure Advertisements were placed in local and statewide newspapers to recruit participants. Individuals who responded to the advertisements were sent a questionnaire in the mail. Participants completed the questionnaire anonymously (within approximately 20 min) and returned it in a prepaid envelope that was provided. Participants were asked to answer questions the way they related to their current partner. Confidentiality and anonymity were assured by coding questionnaires and keeping consent forms and questionnaires separate.
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
126
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
Results Only 1.8% of women in the sample had engaged in a sexual ERI without any emotional involvement. Consequently, purely sexual ERIs were not analysed further, and only emotional and combined ERIs were analysed. Further, for combined ERIs, petting and sexual intimacy without intercourse were examined in the context of a love relationship, and sexual intercourse was examined within the context of a deep love relationship. Women’s intentions to definitely engage in ERI behaviors were 8.0% in a love relationship, 7.1% in a deep love relationship, 9.8% in a combined ERI involving petting and sexual intimacy without intercourse, and 11.6% in a combined ERI involving sexual intercourse. Structural equation analyses (using AMOS 3.61; Arbuckle, 1997) were used to test the hypothesized relations among the exogenous variables (attitudes, social norms, planning, self-efficacy, past behavior, sexual satisfaction, relationship satisfaction, the love styles Eros and Ludus, and commitment) and the endogenous variables of intentions to engage in a love relationship, a deep love relationship, petting and sexual intimacy without intercourse within a love relationship, and sexual intercourse within a deep love relationship. A chi-square test was used to test the adequacy of the proposed models in the prediction of ERI intentions. Other goodness-of-fit indices (Goodness-Of-Fit Index [GFI], Adjusted Goodness-Of-Fit Index [AGFI], Non-Normed Fit Index [NNFI], Tucker–Lewis Index [TLI], and Comparative Fit Index [CFI]; Bentler, 1980; Bollen, 1989) were also examined to determine the adequacy of the models. Ideally, before conducting a test of structural models, a measurement model for observed variables would be conducted. However, in this study, using multiple indicators of observed variables would generate more parameters to be estimated than data points. Therefore, scale scores for observed variables were used, with √ the measurement error variance (σ 2 × [1 − r ]) and regression weights (SD × r ) fixed accordingly. All scales produced good reliable measures (r > .7). For those observed variables measured by one item, measurement error variance was fixed at 1. Table I presents the goodness-of-fit indices for the hypothesized and adjusted models for women’s intentions to engage in the four types of ERI. The hypothesized model of women’s intentions to engage in a love relationship with someone other than their regular partner did not provide a good fit to the data, χ 2 (45) = 587.53, p < .05; RMSEA = .33, all goodness-of-fit indices <.9; see Table I. Inspection of the t-tests for regression weights values revealed that a number of exogenous variables (emotional ERI attitudes, emotional ERI self-efficacy, sexual satisfaction, Eros and Ludus) did not significantly contribute to the prediction of intentions. These variables were removed one by one from the model until all variables significantly contributed to the prediction of intentions (Model 2). The second model also did not provide a good fit to the data, χ 2 (10) = 131.02, p < .05; RMSEA = .33, all goodness-of-fit indices <.9; see Table I. Examination of the modification indices indicated that the second model
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Extra Relationship Involvement Among Women
127
Table I. The Goodness-Of-Fit Indices for the Different Types of ERI: Women’s Intentions to Engage in a Love Relationship, a Deep Love Relationship, Petting and Sexual Intimacy Without Intercourse Within a Love Relationship and Sexual Intercourse Within a Deep Love Relationship Model
χ2
df χ 2/df
p
SMC GFI AGFI NNFI TLI CFI RMSEA
Emotional ERI Love relationship 1 587.53 45 12.86 .00 .62 .39 .11 .19 2 131.02 10 13.10 .00 .55 .68 .32 .49 3 4.60 3 1.5 .20 .69 .99 .91 .98 Deep love relationship 1 557.93 45 12.40 .00 .58 .40 .12 .18 2 127.98 10 12.80 .00 .56 .68 .33 .49 3 6.07 4 1.52 .19 .67 .91 .98 .98 Combined ERI Petting and sexual intimacy without intercourse within a love relationship 1 718.60 55 13.07 .00 .66 .35 .08 .18 2 163.18 10 16.32 .00 .63 .61 .18 .49 3 2.20 2 1.10 .33 .76 .99 .93 .99 Sexual intercourse within a deep love relationship 1 690.63 55 12.56 .00 .66 .37 .11 .20 2 231.73 10 18.33 .00 .60 .54 .04 .42 3 6.39 5 1.28 .27 .77 .98 .92 .98
.01 .26 .97
.19 .50 .99
.33 .33 .07
.00 .24 .97
.18 .50 .99
.32 .33 .07
.02 .24 .99
.18 .50 .99
.33 .37 .03
.05 .14 .99
.21 .45 .99
.32 .45 .05
could be improved by adding pathways between a number of exogenous variables. Pathways were only included if deemed to be conceptually acceptable in the ERI literature. The final model, shown in Fig. 1 was shown to be a good fit to the data, χ 2 (3) = 4.60, p > .10. The ratio of chi-square/degrees of freedom was <2.0, and the RMSEA value and all the goodness-of-fit indices were in the acceptable ranges (see Table I). All pathways were significant after inclusion of pathways between exogenous variables except for the effect of relationship satisfaction on intentions. The model accounted for 69% of the variance in women’s intentions to engage in a love relationship with someone other than their regular partner (see Fig. 1). The hypothesized model of women’s intentions to engage in a deep love relationship with someone other than their regular partner did not provide a good fit to the data, χ 2 (45) = 557.93, p < .05; RMSEA = .32, all goodness-of-fit indices <.9; see Table I. Inspection of the t-tests for regression weights revealed that a number of exogenous variables (emotional ERI attitudes, emotional ERI selfefficacy, sexual satisfaction, Eros and Ludus) did not significantly contribute to the prediction of intentions. These variables were removed one by one from the model until all variables significantly contributed to the prediction of intentions (Model 2). The second model also did not provide a good fit to the data, χ 2 (10) = 127.98, p < .05; RMSEA = .33, all goodness-of-fit indices <.9; see Table I. Examination of the modification indices indicated that the second model could be improved by adding pathways between a number of exogenous variables. Pathways were only included if deemed to be conceptually acceptable in the ERI literature. The final model,
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
128
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
Fig. 1. Final structural equation model for a love relationship ERI (∗p < .05).
shown in Fig. 2, was shown to be a good fit to the data, χ 2 (4) = 6.07, p > .10. The ratio of chi-square/degrees of freedom was <2.0, and the RMSEA value and all the goodness-of-fit indices were in the acceptable ranges (see Table I). All pathways were significant and the model accounted for 67% of the variance in
Fig. 2. Final structural equation model for a deep love relationship ERI (∗p < .05).
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
129
women’s intentions to engage in a deep love relationship with someone other than their regular partner (see Fig. 2). The hypothesized model of women’s intentions to engage in petting and sexual intimacy without intercourse within a love relationship did not provide a good fit to the data, χ 2 (55) = 718.60, p < .05; RMSEA = .33, all goodness-of-fit indices <.9; see Table I. Inspection of the t-tests for regression weights revealed that a number of exogenous variables (combined ERI attitudes, combined ERI selfefficacy, sexual satisfaction, past love relationship within a combined ERI, Ludus and Eros) did not significantly contribute to the prediction of intentions. These variables were removed one by one from the model until all variables significantly contributed to the prediction of intentions (Model 2). The second model also did not provide a good fit to the data, χ 2 (10) = 163.18, p < .05; RMSEA = .37, all goodness-of-fit indices <.9; see Table I. Examination of the modification indices indicated that the second model could be improved by adding pathways between a number of exogenous variables. Pathways were only included if deemed to be conceptually acceptable in the ERI literature. The final model, shown in Fig. 3, was shown to be a good fit to the data, χ 2 (2) = 2.20, p > .10. The ratio of chi-square/degrees of freedom was <2.0, and the RMSEA value and all the goodness-of-fit indices were in the acceptable ranges (see Table I). All pathways were significant and the model accounted for 76% of the variance in women’s intentions to engage in petting and sexual intimacy without intercourse within a love relationship (see Fig. 3).
Fig. 3. Final structural equation model for petting and sexual intimacy without intercourse within as love relationship ERI (∗p < .05).
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
130
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
The hypothesized model of women’s intentions to engage in sexual intercourse within a deep love relationship with someone other than their regular partner did not provide a good fit to the data, χ 2 (55) = 690.63, p < .05; RMSEA = .32, all goodness-of-fit indices <.9; see Table I. Inspection of the t-test for regression weights revealed that a number of exogenous variables (combined ERI self-efficacy, relationship satisfaction, sexual satisfaction, past deep love relationship within a combined ERI, Eros and Ludus) did not significantly contribute to the prediction of intentions. These variables were removed one by one from the model until all variables significantly contributed to the prediction of intentions (Model 2). The second model also did not provide a good fit to the data, χ 2 (10) = 231.73, p < .05; RMSEA = .45, all goodness-of-fit indices <.9; see Table I. Examination of the modification indices indicated that the second model could be improved by adding pathways between a number of exogenous variables. Pathways were only included if deemed to be conceptually acceptable in the ERI literature. The final model, shown in Fig. 4, was shown to be a good fit to the data, χ 2 (5) = 6.39, p > .10. The ratio of chi-square/degrees of freedom was <2.0, and the RMSEA value and all the goodness-of-fit indices were in the acceptable ranges (see Table I). All pathways were significant after inclusion of pathways between exogenous variables except for the effect of combined ERI attitudes on intentions. The model accounted for 77% of the variance in women’s intentions to engage in sexual intercourse within a deep love relationship with someone other than their regular partner. Figure 4 shows the path coefficients for the final model.
Fig. 4. Final structural equation model for sexual intercourse within a deep love relationship ERI (∗ p < .05).
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
131
Discussion The results from Study 1 only provide partial support for TPB in the prediction of intentions with this variable only having an initially significant effect on women’s intentions to engage in an ERI involving sexual intercourse within a deep love relationship. However, once additional pathways were included between exogenous variables, the effect of attitudes on ERI intentions was nonsignificant. Self-efficacy failed to predict the four types of ERI intentions. However, social norms and planning were shown to have a significant influence on each of the four ERI intentions. Partial support was found for the utilization of the additional variables in the prediction of ERI intentions. The variables of sexual satisfaction, Eros and Ludus, which Drake and McCabe (in press) included within the model of ERI, were shown to have no influence on women’s intentions to engage in the four types of ERI. However, relationship satisfaction was shown to contribute positively to women’s intentions to engage in a love relationship, a deep love relationship, and petting and sexual intimacy without intercourse within a love relationship with someone other than their regular partner. Commitment, the variable included within the model of ERI in the current study, was shown to have a significant influence on women’s intentions to engage in the four types of ERI. Women’s intentions to engage in an ERI were shown to be positively influenced by social norms, planning, relationship satisfaction, and past behavior. Commitment was shown to have a negative effect on ERI intentions. Therefore, as a woman perceives greater social support for ERI, plans to become involved in an ERI, is more satisfied with her relationship and has engaged in ERI behavior in the past, her intentions to engage in each of the four types of ERI increases. However, as a woman’s commitment to her relationship increases, her intention to engage in an ERI decreases. All hypothesized relationships were confirmed, except for relationship satisfaction’s effect on ERI intentions, which was in the opposite direction to what was expected. There was only one difference across the four models of ERI, with intentions to engage in sexual intercourse within a deep love relationship being positively influenced by attitudes rather than relationship satisfaction. Thus, the more positive the attitudes toward combined ERI, the greater the intentions to engage in a combined ERI. Overall the four models were shown to account for a large amount of the variance in ERI intentions, with values ranging from 67% to 77%. STUDY 2 Method Study 2 was designed to prospectively determine the variables that best predict ERI behavior. Data for Study 2 were collected in two stages. A subsample (N = 90) of participants from Study 1 who indicated that they were prepared to participate in a follow-up questionnaire after a 6-month interval were the respondents for
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
132
Banfield and McCabe
Study 2. Sixty participants returned the Time 2 questionnaire, but only 44 could be matched with the Time 1 questionnaire. Participants The mean age of the subsample of participants (N = 44) was 36.48 (SD = 10.01, Range = 19–66), and of these, 50% were married, 22.7% were in de facto relationships or living together, and 27.3% were in steady relationships. Study 2 participants did not differ from Study 1 participants on demographic variables or the independent variables included in Study 1, F(19, 92) = 0.66. Materials The Time 2 questionnaire contained the following scales: 1. ERI Behavior Over Last 6 Months: Participants indicated ‘Yes’ or ‘No’ to whether they had engaged in 5 levels of (1) an emotional ERI, (2) a sexual ERI, and (3) a combination of both emotional and sexual ERIs with someone other than their regular partner. Emotional involvement ranged from casual friendship to deep love relationship and sexual involvement ranged from kissing to sexual intercourse. A continuum of depth of ERI involvement was developed from participants’ responses to the 6 levels of involvement for each type of ERI behavior: emotional involvement—0 (no involvement) to 5 (deep love relationship); sexual involvement: 0 (no involvement) to 5 (sexual intercourse). The behaviors examined in the current study were emotional ERI and sexual behavior within the context of a combined ERI. As none of the respondents had engaged in sexual ERI on its own, sexual ERIs were not examined. Participant’s placement along the continuum corresponded to the deepest level of involvement for each type of ERI. 2. Affect: A 20-item affect scale developed by Drake and McCabe (in press) was utilized to assess the reasons (romantic, sexual, and irresponsibility) for having each type of ERI and was measured on a 7-point Likert scale (1: not at all true, 7: extremely true). Only those respondents who had experienced an ERI were asked to complete this section, and only for those items that pertained to the type of ERI they had engaged in. Because of the small number of participants that completed the affect items, reliabilities could not be calculated. Procedure The participants completed the second questionnaire anonymously and returned it in a prepaid envelope that was provided. Anonymity was assured by
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
133
using codes chosen by participants. The Time 1 and Time 2 questionnaires were matched using these codes. Results In the 6 months since the completion of the Time 1 questionnaire, 4.5% of participants had engaged in a love relationship only, 6.8% had engaged in a deep love relationship only, 20.5% had engaged in petting and sexual intimacy without intercourse with some degree of emotional involvement, and 6.8% had engaged in sexual intercourse with some degree of emotional involvement, with someone other than their regular partner. These percentages were generally lower than that of the reports of past behavior from the Time 1 questionnaire, except for petting and sexual intimacy without intercourse with some level of emotional involvement. The percentage of women engaging in this type of ERI behavior more than doubled over the 6-month period since the completion of Time 1 questionnaire. None of the women in the subsample had engaged in a sexual ERI without some form of emotional involvement. Due to the small sample size, standard multiple regressions were conducted for the prediction of ERI behavior instead of utilizing structural equation modelling. The following results can only provide an indication of the prediction of ERI behavior and cannot be easily generalized to other populations. Additionally, as with ERI behavior, a continuum of depth of past ERI involvement (emotional ERI: 0 (no involvement) to 5 (deep love relationship); sexual ERI: 0 (no involvement) to 5 (sexual intercourse) was developed from participants’ responses to the six levels of involvement for past emotional and combined ERI behavior. Participants’ placement along the continuum corresponded to the deepest level of past involvement in both emotional and sexual behavior within the context of combined ERIs. A continuum of intentions to engage in emotional ERIs and sexual behavior within the context of combined ERIs was also developed. It involved summing the highest level of intentions to engage in each level of emotional ERI and sexual behavior within the context of a combined ERI (0: certainly no intentions, 30: certainly intend to). A standard multiple regression was conducted to determine the predictors of emotional ERI behavior. Table II shows the correlations between the variables, means, standard deviations, the unstandardized regression coefficients (B) and intercept, the standardized regression coefficients (β), the semipartial correlations (sr 2 ) R 2 , and adjusted R 2 . Affect was not included in the regression equation because data were only collected for participants who had engaged in an ERI. The results of the analysis confirmed that emotional ERI intentions, past emotional ERI behavior, planning, control, commitment, and relationship satisfaction significantly predicted a woman’s involvement in emotional ERI behavior, F(6, 37) = 6.80, p < .001. Past emotional ERI behavior (sr 2 = .08) contributed
1.95 1.52
.16 .67 2.27 1.88
−0.61 2.23
Control
−.12 −.56 −.30
Planning
12.93 2.82
.66
Commitment
29.27 7.73
Relationship satisfaction
β
.47a .39 .69
.05 .28 .40 .38∗ .05 .09 .06 .08 .01 .02 .00 −.01 −.13
B
.08
sr 2 (unique)
19:35
= .08; Shared variability = .39.
18.00 8.36
2.00 1.41
.50 −.08 −.56 −.20
Past emotional ERI
January 25, 2001
a Unique variability ∗ p < .05.
.69 .75 .03 −.51 −.30
Emotional intentions
.68 .67 .50 .04 −.26 −.17
Emotional ERI behavior
PP016-291180
Emotional ERI behavior Emotional intentions Past emotional ERI Planning Control Commitment Relationship satisfaction Intercepts Means Standard deviations R2 Adjusted R 2 R
Variables
Archives of Sexual Behavior [asb]
Table II. Standard Multiple Regression of Emotional ERI Behavior
P1: FYJ/FNV P2: FLF Style file version Nov. 19th, 1999
134 Banfield and McCabe
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
135
significant unique variance to the prediction of involvement in emotional ERI behavior. The six predictor variables in combination contributed a further .39 in shared variability. Altogether, 47% (39% adjusted) of the variability in an individual’s involvement in emotional ERI behavior was predicted by the six predictor variables. A standard multiple regression was conducted to determine the predictors of combined ERI. Table III shows the correlations between the variables, means, standard deviations, the unstandardized regression coefficients (B) and intercept, the standardized regression coefficients (β), the semipartial correlations (sr 2 ) R 2 , and adjusted R 2 . Affect was not included in the regression equation because data were only collected for participants who had engaged in an ERI. The results of the analysis confirmed that sexual intentions, emotional intentions, past emotional behavior, and past sexual behavior within the context of a combined ERI along with planning, control, commitment, and relationship satisfaction significantly predicted an individual’s involvement in sexual behavior within the context of a combined ERI, F(8, 35) = 11.73, p < .001. Past sexual behavior within the context of a combined ERI (sr 2 = .09) and planning (sr 2 = .03) contributed significant unique variance to the prediction of involvement in sexual behavior within the context of a combined ERI. The eight predictor variables in combination contributed a further .64 in shared variability. Altogether, 77% (71% adjusted) of the variability in an individual’s involvement in sexual behavior within the context of a combined ERI was predicted by the eight predictor variables. To test for differences in the reasons given by females for their involvement in emotional ERIs and sexual behavior within the context of combined ERIs, 2 oneway ANOVAs were conducted. Means and standard deviations, F values, degrees of freedom, and significance levels are presented in Table IV for the three affective reasons for the two types of ERI behavior. The ANOVAs revealed that women who engaged in emotional ERIs and sexual behavior within the context of combined ERIs attributed differentially affective reasons for their ERI behavior. Women who engaged in these two types of ERI behavior cited romantic reasons over sexual or irresponsibility reasons for their involvement in ERI behavior. Discussion The findings from Study 2 showed that past emotional ERI behavior and past sexual behavior within the context of a combined ERI were significant predictors of emotional ERI behavior and sexual behavior within the context of a combined ERI, respectively. It also indicated that planning has an influence on sexual behavior within the context of a combined ERI. Women cited romantic reasons as contributing most to their involvement in both emotional ERIs and sexual behavior within combined ERIs.
16.52 8.93
0.89 1.69
12.14 11.10
.66 −.75 −.92 −.07 −.66 −.39 0.80 1.62
−.93 −.59 −.01 −.49 −.37 1.00 1.92
.65 −.04 −.55 −.35 .33 .52 2.07 2.06
−.05 −.54 −.37 −0.98 1.95
12.93 2.82
.67 29.27 7.73
.77a .71 .88
.02 −.03 −.32 .82∗ .40∗ −.07 −.01 .04 1.09
B
.12 −.20 −.31 .93 .46 −.09 −.02 .18
β
.09 .03
sr 2 (unique)
January 25, 2001 19:35
= .12; Shared variability = .65.
.80 .45 .53 −.75 .01 −.56 −.31
.61 .75 .69 .81 −.73 −.04 −.47 −.21
Combined ERI Emotional Sexual Past Past Relationship behavior intentions intentions emotional ERI sexual ERI Planning Control Commitment satisfaction
PP016-291180
a Unique variability ∗ p < .01.
Combined ERI behavior Emotional intentions Sexual intentions Past emotional ERI Past sexual ERI Planning Control Commitment Relationship satisfaction Intercepts Means Standard deviations R2 Adjusted R 2 R
Variables
Archives of Sexual Behavior [asb]
Table III. Standard Multiple Regression of Combined ERI Behavior
P1: FYJ/FNV P2: FLF Style file version Nov. 19th, 1999
136 Banfield and McCabe
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Extra Relationship Involvement Among Women
137
Table IV. Analyses of Variance for Romantic, Sexual, and Irresponsibility Reasons for an Emotional ERI and Combined ERI Affective Reasons
Emotional ERI M SD Sexual behavior within the context of a combined ERI M SD ∗p
Romantic
Sexual
Irresponsibility
N
F value
(df )
17.10 9.24
7.32 5.28
3.89 2.71
19
22.21∗
(2, 54)
26.75 10.31
13.25 6.34
6.75 3.45
8
15.77∗
(2, 21)
< .001.
GENERAL DISCUSSION The current study was designed to develop a model for female emotional ERI, sexual ERI, and combined ERI intentions and behavior. The results of this study indicated that women engage in emotional ERIs and combined ERIs, but rarely enter into purely sexual ERIs. This finding is similar to other studies (Glass and Wright, 1985; Townsend, 1995). For example, Glass and Wright (1985) found that only 4% of female participants engaged in extramarital sexual intercourse in the absence of emotional involvement. The rarity of purely sexual ERIs suggests that there may be two alternative pathways to describe women’s ERIs. The first pathway involves a woman initially becoming emotionally involved with her ERI partner. As this emotional relationship develops, the pathway diverts, with the ERI either staying purely emotional or incorporating a sexual component. In this way, a woman may progress from a purely emotional ERI to a combined ERI. The second pathway involves a woman initially engaging in a sexual ERI and as a consequence of this involvement she develops an emotional bond with her ERI partner. Townsend (1995) found that 73% of women who had engaged in sexual intercourse with a person with whom they did not intend to be emotionally involved, reported that regardless of this intention, sexual relations made them feel emotionally vulnerable and concerned about how their sexual partner felt about them. From these findings, it is apparent that women find it difficult to separate sexual involvement from emotional attachment to their ERI partners. Thus, only models of emotional ERIs and combined ERIs were developed in the current study. The ERI models examined the contribution made by TPB, past ERI behavior, the person’s relationship, and commitment to the primary relationship to emotional ERI and combined ERI intentions and behavior. There was partial support for the utility of the TPB in the prediction of ERI intentions. The current study demonstrated that women’s ERI intentions were greater when they perceived positive support for ERI from others close to them
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
138
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
and when they had planned to engage in an ERI. However, women’s attitudes toward ERI were generally negative, and their belief in their ability to engage in ERI behavior was low regardless of their levels of ERI intentions. These findings are consistent with those of Buunk and Bakker (1995), who found that individuals are guided by the perceived approval of others when deciding to engage in an ERI. Further, Drake and McCabe (in press) found that planning predicted ERI intentions and self-efficacy failed to influence these intentions. Generally, the current findings indicate that women tend to consider others’ opinions when deciding to engage in ERI rather than being guided by their own beliefs. The negative ERI attitudes and low self-efficacy reported by women are likely to have been influenced by socialization processes and attitudes and beliefs that are sanctioned by the wider community. Planning may aid in imagining the act of ERI and thus may firm intentions. It was also found that the TPB did not fully explain the decision-making process involved in emotional and combined ERI behavior. ERI intentions and control failed to predict emotional ERI and combined ERI behavior. However, it was found that planning predicted sexual behavior in the context of a combined ERI but not emotional ERI behavior. Kashima et al. (1993) suggested that due to the dynamics of an intimate relationship involving two people “even if behavioral conditions are satisfied, people may change their minds because of last minute pressure from the partner or for other reasons” (p. 237). This may explain why women’s ERI intentions and perceived control over the ERI situation failed to predict ERI behavior, with the relationship between these variables being highly dependent on the willingness, cooperation, and availability of an ERI partner. However, the influence of planning on women’s involvement in sexual behavior within the context of a combined ERI indicates that women have considered possible problems that may hinder completion of the sexual behavior, and planned ways to deal with them to ensure the behavior will occur. The failure of planning to predict emotional ERI behavior indicates that emotional involvement with an ERI partner tends to be more habitual than sexual behavior within a combined ERI, which appears to be more deliberate and calculated. The TPB offers a framework from which to examine ERI intentions and behavior, but other variables that demonstrate the habitual nature of ERI were also examined. The habitual nature of ERI was illustrated by women’s previous involvement in emotional ERI and the strong association of combined ERI behavior with high levels of ERI intentions and behavior. Intentions to engage in an ERI in the future were generally similar to reports of past ERI behavior for the four types of ERI. Further, the number of women who engaged in each type of ERI during the study was generally slightly lower than the number who had engaged in these behaviors in the past. These results demonstrate that it would be of greater benefit to explore habitual factors contributing to ERI rather than relying on the examination of attitudes and control beliefs, which may be distorted by moral and social influences.
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
139
The relationship variables of Ludus, Eros, and sexual satisfaction were shown to not add to the prediction of women’s intentions to engage in ERIs. These variables were not entered into the prediction of ERI behavior. However, women’s ERI intentions were higher if they had low commitment to, and high satisfaction with, their primary relationship. This association between relationship satisfaction and ERI intentions was not in the hypothesized direction, and was inconsistent with Rusbult’s (Rusbult, 1980, 1983) Investment Model of commitment and the “deficit” model of ERI (Buunk, 1980). The results indicate that women are not displaying low relationship commitment and high ERI intentions because of “deficits” in their relationships (i.e., low relationship satisfaction) but instead are being pulled into contemplating ERI by other factors that may include a rich quality of alternative partners and minimal investment in the primary relationship (Maykovich, 1976). Interestingly, the relationship of commitment and relationship satisfaction with ERI intentions did not hold for ERI behaviors, which suggests that these variables are not influential in pulling women into engaging in an ERI. Women indicated that they engaged in emotional and combined ERIs because of the romantic affect they felt toward the ERI partner. Having alternative partners to choose from may lower commitment to the primary relationship and thus increase intentions, but for women to engage in an ERI there needs to be an emotional connection. Further research needs to examine the effect of availability of alternative partners and quality of these alternatives on ERI intentions and behavior. When comparing the results of the current study with Drake and McCabe’s (Drake and McCabe, in press) research on male ERI, it is apparent that males pursue more sexual ERIs and females pursue more emotional and combined ERIs. Women’s ERI intentions are influenced by their perceptions of what others think about their becoming involved in an ERI, whereas men’s intentions are influenced by their attitudes toward ERI behaviors. Drake and McCabe (in press) demonstrated that the game playing style of love influenced men’s ERI intentions, whereas relationship satisfaction and commitment were important predictors of women’s ERI intentions. Past ERI behavior was a strong predictor of future ERI behavior for both males and females. However, unlike Drake and McCabe’s (Drake and McCabe, in press) study, planning was shown to have an influence on combined sexual ERI behavior. This indicates that although emotional ERIs tend to be habitual behavior for women, sexual behavior within combined ERIs required a concerted planning effort for women to follow through with this behavior. Conversely, male involvement in sexual ERIs tended to be largely habitual, occurring in the absence of intentions and planning (Drake and McCabe, in press). Further, although men did not report that they engaged in ERI for affective reasons (Drake and McCabe, in press), women cited romantic affect as the strongest reason for becoming involved in both emotional and combined ERIs. This study is the first longitudinal study of women’s emotional ERI and combined ERI behaviors, and has provided a prospective investigation of why ERI behavior occurs. However, the prediction of behavior in the current study was
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
140
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
limited by the small sample size. The findings are also limited in that the data were gathered from respondents who were from an Anglo-Saxon background, from the middle socioeconomic class, who resided in a large urban city. Although the data can be compared to other studies, as they had similar respondents, this study along with the studies reviewed in the Introduction to this paper may have limited generalizability to women from other countries, other cultural groups, rural regions, or other socioeconomic classes. The ERI intentions and behaviors of women from these different groupings require further investigation. The role of children in the ERI intentions and behaviors of both women and men also needs to be explored. Future research needs to further explore the habitual nature of ERI intentions and behavior. An important area that may contribute to our understanding of ERI intention and behavior is the evolutionary interpretation of sexual behavior. Buss et al. (1999) have suggested that evolutionary selection pressures, namely differential parental investments, have produced a fundamental difference in the sexual psychology of males and females. Women have evolved to select sexual partners who provide good genetic material, emotional investment, aid, and protection to their potential offspring (Alcock, 1993; Buss et al., 1999; Townsend, 1995). Conversely, males have evolved to seek multiple partners, which confers the benefit of increasing their number of offspring (Alcock, 1993). Thus, Buss et al. (1999) believe that women are more likely to desire and engage in emotional ERIs, and it is more probable for them to become emotionally involved when engaging in a sexual liaison. This theory is able to account for the rarity of purely sexual ERIs in the female population and for the differences in the type of ERI relationships pursued by males and females. In conclusion, the current study found that women either engage in an emotional ERI or combined ERI but rarely in a purely sexual ERI. Further, the current study demonstrated that similar processes are involved in women’s emotional and combined ERI intentions and behaviors. It has been demonstrated that intentions to engage in ERI are related to cognitive processes and relationship variables and that ERI behavior, although generally habitual, requires more cognitive processes as the relationship becomes more sexual. Finally, women differ from men in the type of ERI they engage in and the factors that predict their intentions and behaviors.
REFERENCES Ajzen, I. (1991). The theory of planned behavior. Org. Beh. Hum. Dec. Proc. 50: 179–211. Alcock, J. (1993). Animal Behavior: An Evolutionary Approach, 5th Ed. Sinauer Associates, Massachusetts. Arbuckle, J. (1997). Amos 3.61, SmallWaters Corporation, Chicago. Bell, R., Turner, S., and Rosen, L. (1975). A multivariate analysis of extramarital coitus. J. Marr. Fam. 37: 375–644. Bentler, P. (1980). Multivariate analysis with latent variables: Causal modeling. Ann. Rev. Psychol. 31: 419–456. Bollen, K. (1989). Structural equations with latent variables. John Wiley and Sons, New York.
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
PP016-291180
January 25, 2001
Extra Relationship Involvement Among Women
19:35
Style file version Nov. 19th, 1999
141
Buss, D., Shackelford, T., Kirkpatrick, L., Choe, J., Lim, H., Hasegawa, M., Hasegawa, T., and Bennett, K. (1999). Jealousy and the nature of beliefs about infidelity: Tests of competing hypotheses about sex differences in the United States, Korea, and Japan. Pers. Rel. 6: 125–150. Buunk, B. (1980). Extramarital sex in the Netherlands: Motivations in social and marital context. Alter Lifes 3: 11–39. Buunk, B., and Bakker, A. (1995). Extradyadic sex: The role of descriptive and injunctive norms. J. Sex. Res. 32: 313–318. Buunk, B., and Bakker, A. (1997). Commitment to the relationship, extradyadic sex, and AIDS preventive behavior. J. Appl. Soc. Psychol. 27: 1241–1257. Denes-Raj, V., and Epstein, S. (1994). Conflict between intuitive and rational processing: When people behave against their better judgement. J. Pers. Soc. Psychol. 5: 819–829. Drake, C., and McCabe, M. (in press). Extrarelationship involvement among heterosexual males: An explanation based on the theory of planned behavior, relationship quality and past behavior. J. Appl. Soc. Psychol. Edwards, J., and Booth, A. (1976). Sexual behavior in and out of marriage: An assessment of correlates. J. Marr. Fam. 38: 73–81. Epstein, S. (1994). Integration of the cognitive and psychodynamic unconscious. Am. Psychol. 49: 709–724. Gallois, C., Terry, D., Timmins, P., Kashima, Y., and McCamish, M. (1994). Safe sexual intentions and behavior among heterosexual and homosexual men: Testing the theory of reasoned action. Psychol. Health 10: 1–10. Gangestad, S., and Thornhill, R. (1997). The evolutionary psychology of extrapair sex: The role of fluctuating asymmetry. Evol. Human Behav. 18: 69–88. Glass, S., and Wright, T. (1985). Sex differences in the types of extramarital involvement and marital satisfaction. Sex Roles 12: 1101–1119. Glenn, N., and Weaver, N. (1979). Attitudes towards premarital extramarital and homosexual relations in the U.S. in the 1970s. J. Sex. Res. 15: 108–119. Godin, G., Valois, P., Lepage, L., and Desharnais, R. (1992). Predictors of smoking behavior: An application of Ajzen’s theory of planned behavior. Brit. J. Add. 87: 1335–1343. Hendrick, C., and Hendrick, S. (1986). A theory and method of love. J. Per. Soc. Psychol. 50: 392–402. Hendrick, C., and Hendrick, S. (1989). Research on love: Does it measure up? J. Per. Soc. Psychol. 56: 784–794. Johnson, R. (1970). Some correlates of extramarital coitus. J. Marr. Fam. 32: 449–456. Johnson, D., and Rusbult, C. (1989). Resisting temptation: Devaluation of alternative partners as a means of maintaining commitment in close relationships. J. Pers. Soc. Psychol. 57: 976–980. Kashima, Y., Gallois, C., and McCamish, M. (1993). The theory of reasoned action and cooperative behavior: It takes two to use a condom. Brit. J. Soc. Psychol. 32: 227–239. Kelley, J. (1978). Sexual permissiveness: Evidence for a theory. J. Marr. Fam. 40: 455–468. Kelley, J., and Thibaut, J. (1978). Interpersonal Relations: A theory of Interdependence, Wiley, New York. Lee, J. (1976). The Colors of Love, New Press, Ontario. Lee, J. (1988). Love styles. In Sternberg, R., and Barnes, M. (eds.), The Psychology of Love, Yale University Press, New York, pp. 38–67. Maykovick, M. (1976). Attitude versus behavior in extramarital relations. J. Marr. Fam. 38: 693–699. McCabe, M. (1994). Childhood, adolescent and current factors associated with sexual dysfunction. Sex. Marr. Ther. 9: 267–276. Meyering, R., and Epling-McWherter, E. (1986). Decision making in extramarital relationships. Lifes: J. Chg. Patt. 8: 115–129. Reiss, I., Anderson, R., and Sponaugle, G. (1980). A multivariate model of the determinates of extramarital sexual permissiveness. J. Marr. Fam. 42: 395–411. Roscoe, B., Cavanaugh, L., and Kennedy, D. (1988). Dating infidelity: Behaviors, reasons and consequences. Adolescence 23: 35–43. Rusbult, C. (1980). Commitment and satisfaction in romantic associations: A test of the investment model. J. Exper. Soc. Psychol. 16: 172–186. Rusbult, C. (1983). A longitudinal test of the investment model: The development (and deterioration) of satisfaction and commitment in heterosexual involvements. J. Pers. Soc. Psychol. 43: 101– 117.
P1: FYJ/FNV
P2: FLF
Archives of Sexual Behavior [asb]
142
PP016-291180
January 25, 2001
19:35
Style file version Nov. 19th, 1999
Banfield and McCabe
Rusbult, C., and Buunk, B. (1993). Commitment processes in close relationships: An interdependence analysis. J. Soc. Pers. Rel. 10: 175–204. Saunders, J., and Edwards, J. (1984). Extramarital sexuality: A predictive model of permissive attitudes. J. Marr. Fam. 46: 825–834. Schafer, R., and Keith, P. (1981). Equity in marital roles across the family life cycle. J. Marr. Fam. 43: 359–367. Singh, B., Walton, B., and Williams, J. (1976). Extramarital sexual permissiveness: Conditions and contingencies. J. Marr. Fam. 38: 701–712. Spanier, G., and Margolis, R. (1983). Marital separation and extramarital sexual behavior. J. Sex. Res. 19: 23–48. Sprecher, -S. (1998). Social exchange theories and sexuality. J. Sex. Res. 35: 32–43. Terry, D., and O’Leary, J. (1995). The theory of planned behavior: The effects of perceived behavioral control and self efficacy. Brit. J. Soc. Psychol. 34: 199-220. Thompson, A. (1983). Extramarital sex: A review of the research literature. J. Sex. Res. 19: 1–22. Thompson, A. (1984). Emotional and sexual components of extramarital relations. J. Marr. Fam. 46: 35–42. Townsend, J. (1995). Sex without emotional involvement: An evolutionary interpretation of sex differences. Arch. Sex. Behav. 24: 173–205. Weis, D., and Felton, J. (1987). Marital exclusivity and the potential for future marital conflict. Soc. Work Jan. Feb., 45–49. Wiederman, M. (1997). Extramarital sex: Prevalence and correlates in a national survey. J. Sex. Res. 34: 167–174.
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
19:37
Style file version Nov. 19th, 1999
Archives of Sexual Behavior, Vol. 30, No. 2, 2001
A Framework For the Formation of Gay Male Identity: Processes Associated with Adult Attachment Style and Support From Family and Friends Yoel Elizur, Ph.D.,1,3 and Arlette Mintzer, Ph.D.2
We propose a novel conceptual framework for the study of gay male identity formation in relation to the person’s self, family, and social relations. The three basic processes of gay male identity are defined—self-definition, self-acceptance, and disclosure—and theoretically linked to attachment style and social support. The results, based on an Israeli sample of gay men (n = 121), indicated that self-acceptance and friends’ support predict secure attachment in close adult relationships, and that self-definition and support from family and friends predict disclosure. Supportive family attitudes toward same-sex orientation mediated the effect of general family support on disclosure. These results suggest that the independent assessment of identity processes provides a flexible alternative to stage models’ assumption of a single linear developmental process, that the formation of gay identity is associated with inner models of adult relationships, and that support of family and friends have a different role in the coming out process and in the formation of one’s adult attachment model. KEY WORDS: homosexuality; gay men; identity; attachment style; social support.
INTRODUCTION During the last three decades, a variety of developmental stage models of lesbian, gay, and bisexual identity formation have been formulated (Berzon, 1979; Cass, 1979; Troiden, 1989). The emphasis has been on individual development, 1 The
Hebrew University of Jerusalem, Jerusalem, Israel. Academic College, Beer Tuvia, Israel.
2 Achva
3 To whom correspondence should be addressed at Department of Psychology, The Hebrew University
of Jerusalem, Mount Scopus, Jerusalem, Israel; e-mail: [email protected]. 143 C 2001 Plenum Publishing Corporation 0004-0002/01/0400-0143$19.50/0 °
P1: FMN Archives of Sexual Behavior [asb]
144
PP016-291181
January 25, 2001
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
and the common assumption is that identity develops as individuals work through predictable conflicts and stresses that are related to their sexual orientation, and acquire an affirmative sense of self that enables them to accept and express their same-gender feelings. Notwithstanding their contributions, stage models have increasingly been challenged during the last few years for their emphasis on the unraveling of an inner psychobiological core and the claim for a universal sequence of development, which is defined differently by each stage model. Such an approach to theory and research minimizes the extent to which identity formation is a multifaceted and a protean process shaped by history, geography, and culture (Boxer and Cohler, 1989; Cox and Gallois, 1996; Eliason, 1996; Laird, 1993). Following this critique, our first purpose was to apply a multidimensional, process-oriented framework to the study of gay male identity formation. The development of this alternative approach began in a previous Israeli study during which we encountered the limitations of one such stage model and attempted to rectify it (Elizur and Ziv, manuscript submitted for publication). The results suggested that at this stage of knowledge there is a need for a more modest theoretical alternative that could identify processes or tasks of identity formation that appear to be most central and universal. The processes of working through these developmental tasks can then be studied without presupposing a rigid sequence that organizes their interrelationships. Different people in different contexts may achieve a relatively stable and integrated sense of identity through diverse developmental trajectories, and there is no one optimal endpoint against which they are to be evaluated. Subsequently, we defined three basic identity formation processes, which will be described in the first part of the Introduction. The present study was designed to enable a separate examination of each identity variable without making any of the theoretical assumptions that have been criticized in stage models (Baumeister, 1986; Jordan et al., 1992). The second purpose of this study was to broaden the understanding of identity formation in two main directions. The first direction was to create a link between the field of gay male identity formation and attachment theory. This is an innovative direction that has not been studied before, and hence, the second part of our Introduction will briefly present attachment theory together with the empirical and theoretical basis that supports this linkage. The second direction was to contextualize the study of identity by examining how various sources and types of social support are differentially associated with identity formation variables. A particular emphasis was given to family support variables because most developmental models of gay identity do not consider the family matrix. We sought to rectify this bias and distinguish between general family support and family acceptance of same-gender sexual orientation. In the last part of the Introduction we present the rationale for this approach and its implications regarding the links between gay male identity formation and various types of social support. The decision to focus on gay males was taken because of pragmatic reasons only. The primary reason was that we thought that it would be more feasible for
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
19:37
Gay Male Identity Formation Processes
Style file version Nov. 19th, 1999
145
us to obtain a relatively diverse sample of Israeli-born gay males, whereas the sampling of lesbian women would lead to an overly large representation of highly educated English-speaking immigrants. The other reason was that differences in identity formation between gay men and lesbians could make it necessary to run a separate analysis of the two genders (Jordan et al., 1992; Kitzinger and Wilkinson, 1995). Because resources limited our sample size, the inclusion of both genders would have compromised the statistical power of our analysis.
Gay Male Identity Development An examination of the three groups of respondents that were identified in our initial study revealed that each one was characterized by a major identity task (Elizur and Ziv, manuscript submitted for publication). These tasks were the most universal threads that also stood out in the different gay identity development models that we had reviewed. Consequently, three processes that are commonly accepted as fundamental for the integration of gay identity were delineated. Self-Definition The realization of same-gender feelings against the background of self and others’ expectations for the development of heterosexuality is a painful break from the “reality” that has been shared with the family and the community. The creation and consolidation of an alternative identity narrative requires the working through of denials, pressures to conform to family expectations and the majority culture, and fears of real and imagined consequences (Michaels, 1996). Inner and outer conflicts concerning sexual identity may well persist after an initial identity definition is made, and individuals may make substantial changes in their view about themselves. This is partly related to the undermining effects of social and often familial prejudices against gay identity. Because a substantial group of individuals have both same-gender and heterosexual feelings, these outside pressures may increase ambivalence and shifts in self-definition. Bisexuals may also experience identity pressures from members of the gay community (Fox, 1995). Self-Acceptance Prejudicial social practices against gay men make this population highly susceptible to traumatic experiences, loss of social status, and the internalization of heterosexism (Gonsiorek, 1995). Consequently, the development of acceptance helps to depathologize one’s sense of self and to consolidate a positive sexual identity. This is both an intrapsychic cognitive–emotional evolvement and an
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
146
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
interpersonal process (Fassinger and Miller, 1996). Contact with other gay men often helps to overcome the sense of isolation and stigmatization, providing selfaccepting role models, whereas sharing one’s identity with significant heterosexual others complements the growth of acceptance. Disclosure This is not a discreet and dramatic act, but a long-term process that encompasses both leaps of disclosure and continuous dialogues with others, during which the identity narrative is repeatedly reshaped and enriched with new meanings. Supportive relationships with heterosexual others help gay men to take emotional risks. Consequently, tensions related to incongruent experiences of self in gay and heterosexual circles are reduced, and one is better able to integrate conflicting feelings, cognitions, and behaviors (D’Augelli and Hershberger, 1993). In existential terms, the task of disclosure is to find the most appropriate way in one’s cultural and social context of “being in the world with others,” which is an essential foundation for authenticity (Yalom, 1980). According to this framework, the working through of each identity formation task is conceptualized as a process, even though one may experience discontinuous steps along the way. Unlike stage models, we do not categorize participants according to a preconceived linear sequence of developmental milestones, but assess their position on each of the three interrelated continuums. At the same time, there is a developmental logic that makes some processes more primary than others. One needs some degree of self-definition to begin the working through of self-acceptance, whereas progress in these two areas is expected to facilitate disclosure. Following this logic, it is hypothesized that as gay males subscribe to a more definite view of their sexual orientation and learn to accept themselves, they would be more likely to disclose themselves to family members and heterosexual friends. The identity formation scheme has interesting implications that can be studied once the parameters of gay identity are clearly differentiated. For example, it is possible to compare cross-cultural samples by analyzing the distributions of the different identity scores as a function of age, a comparison that is not possible when using global identity scales that are based on stage models. Our prediction would be that the time lags between self-definition and acceptance, and between acceptance and disclosure would vary in different cultures as a function of the prevailing attitudes toward the gay population. In societies that are more discriminatory, we would expect to find a greater time lag. This is consonant with research indicating that American gay men and lesbians have become more open at earlier ages about their sexual orientation, with a consequent shortening of the time lag between self-labeling and disclosure (D’Augelli and Hershberger, 1993; Herdt, 1992; Savin-Williams, 1998). This trend, which seems to characterize Western
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
Gay Male Identity Formation Processes
19:37
Style file version Nov. 19th, 1999
147
countries, is commonly assumed to be related to increased social acceptance as well as to empowering processes within the gay and lesbian communities. The greater prevalence of heterosexism in Israeli society (Lieblich and Friedman, 1985; Weishut, in press) can therefore explain the finding that Israeli gay men experience various developmental milestones (i.e., becoming aware of gay feelings, engaging in same-gender sexual behavior, and defining themselves as gays) at later ages in comparison with gay male Americans (Elizur and Ziv, manuscript submitted for publication). Similarly, we expect to find that the average age of first disclosure to parents would be higher in Israel than in the United States. Attachment Styles According to the evolutionary theory of Bowlby (1973), people from infancy onward develop “internal working models” of self in relation to significant others. These models are similar to generalized expectations that are based on actual attachment experiences and are subsequently applied to new situations and relationships. They have been empirically linked to variables that are also associated with the formation of gay male identity: adjustment, identity development, and functioning in close relationships. A meta-analysis of 33 studies that included clinical and nonclinical adolescents and adults from different countries showed an association between insecure attachment representations and psychopathology (van Ijzendoorn and Bakermans Kranenburg, 1996). Insecure attachment styles were found to be overrepresented in a variety of disorders, such as depression, eating disorders, conduct disorders, and borderline personality disorders (del Carmen and Huffman, 1996; Fonagy et al., 1996; West et al., 1993). Taken together, these findings support the proposal that gay male identity, which was found to be associated with mental health and self-esteem (Carlson and Steuer, 1985; Hammersmith and Weinberg, 1973; Savin-Williams, 1990), would also be associated with the formation of a secure attachment style. Above and beyond the empirical support for an association between these two sets of variables, we are suggesting a theoretical link between attachment theory and identity development. Our proposition is that the integration of gay male identity into these individuals’ overall matrix of identity contributes to the formation of a secure attachment style, operationalized as a more positive concept of self in current relationships. In other words, we expect that gay males would be better able to consolidate a secure attachment model as they progress toward the consolidation of their unique identity. Processes such as working through of inner confusions regarding their self-definition, validating and integrating disowned parts of their inner and outer experiences, becoming aware of and learning to externalize internalized heterosexist attitudes, increasing self-acceptance, and achieving a more open and congruent relationship with the social environment. This view is supported both by clinical work (Holmes, 1996) and by research (Benson et al., 1992),
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
148
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
which has demonstrated a general association between identity development and attachment. However, there have been very few studies of attachment styles in gay populations, and the relationship between attachment and gay male identity development has not been examined so far. To avoid confusion it should be mentioned that a person’s standing on attachment style has been studied both as a stable source of variance, a trait that predicts other variables (cf. Main, 1996), and as a changing source of variance, a product of unique person–situation interactions predicted by more stable personality or environmental qualities or both (cf. Kurdek, 1997). Both approaches are supported by longitudinal research findings that have indicated a continuity of attachment patterns, as well as unpredictability, partly derived from life events (Main, 1996). In the present research that is focused on developmental changes in the respondents’ sense of self, we were interested in the prediction of attachment style. We therefore followed the extension of attachment theory of Hazan and Shaver (1987) to the domain of close adult relationships. The assessment of attachment styles in the context of these relationships emphasizes the state component of the attachment variable (Kurdek, 1997). Moreover, changes in gay male identity would be expected to have a more proximal effect on current adult models of attachment than on narratives of early attachment experiences. The associations between attachment styles and qualities of intimate relationships that have been found in adult research (Collins and Read, 1990; Mikulincer and Erev, 1991; Mikulincer and Nachshon, 1991) support the hypothesis that identity formation processes would be positively associated with the consolidation of a secure adult attachment style. Social Support The traditional bias toward studying development as an individual voyage to maturity has been challenged by ecologically oriented developmental psychologists as early as two decades ago (Bronfenbrenner, 1979). In particular, systems theory, research, and clinical case studies have articulated the postulate that relationships with family members and the overall quality of the family system play a major part in the construction of the self and the world (Elizur and Minuchin, 1989; Reiss, 1981). Assuming that the consolidation of gay male identity is influenced by the interpersonal context and that the relationship between inner identity dynamics and transactional patterns may very well vary at different phases of identity formation, we decided to focus on the variable of social support. Studies of social support suggested that the characteristics of support providers should be examined to explicate specific mechanisms that operate in this global field (Elizur and Hirsh, 1999; Krahn, 1993; Schulz and Tompins, 1990). The most common differentiation is between family support and support from friends. Research with gay men and lesbians indicated that friends are more frequent providers of social
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
Gay Male Identity Formation Processes
January 25, 2001
19:37
Style file version Nov. 19th, 1999
149
support than the family (Kurdek, 1988). This finding is consonant with current knowledge about gay male identity development. Friends usually fill in the empty space that is created when gay youth hide their sexual orientation, and the difference between these two sources of support is maintained when they enter into cohabitating relationships (Bernstein, 1990; Herdt, 1992; Kurdek, 1988). At the same time, there are also indications that family reactions toward same-sex orientation affect identity formation, particularly the process of disclosure (Ben-Ari, 1995; Newman and Muzzonigro, 1993; Savin-Williams, 1989b; Strommen, 1989; Troiden, 1989; Wells and Kline, 1987). Because the association between gay identity formation and family support has not been adequately researched, we opted for a more detailed picture of this connection by measuring both general family support and family attitudes toward same-sex orientation. These two variables were found in a previous study to be positively associated with gay male identity formation (Elizur and Ziv, manuscript submitted for publication). Furthermore, the results suggested that the more specific variable of accepting family attitudes toward same-sex orientation played a more important role in the integration of gay male identity than did general family support. Subsequently, we wished to reconfirm the general hypothesis that both forms of family support would be positively related to gay male identity and to test a mediational model. The hypothesis was that the generic effect of general family support on identity formation would be mediated by the more specific effect of family acceptance of same-sex orientation. The conceptual framework of gay male identity formation, which in this study was applied in a planned way for the first time, enabled us to examine specific relationships between each of the three identity variables and the family variables. We could not only test for positive associations between family support and gay male identity but could also analyze the variations in these associations as a function of different identity formation processes. The literature review with respect to the development of gay male identity indicated that parents are not usually involved when their sons work through the process of self-definition, that parents’ attitudes toward same-sex orientation become relevant to self-acceptance if they are perceived as important components of a youth’s self-worth (Savin Williams, 1989a), and that these attitudes are highly important in the processing of disclosure (Ben-Ari, 1995; Savin-Williams, 1989b; Strommen, 1989; Troiden, 1989). Hence, it would be most appropriate to test the mediation hypothesis with respect to the prediction of disclosure. Another way to gain insight into the particular role of family support is to compare it with support received from friends. Following the finding of Kurdek (1988) that gay men report their friends to be more frequent providers of social support than their family members, we expected to find that friends’ support, but not family support, would be positively associated with adult attachment style. At the same time, on the basis of evidence presented earlier, both sources of support were expected to predict disclosure. Furthermore, according to our conceptual
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
150
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
framework of identity formation, the working through of the first two identity issues was also expected to facilitate the process of disclosure. Because self-definition and self-acceptance can be seen as individual level variables, whereas the different measures of social support represent the interpersonal level, we hypothesized that each set of variables would make an independent contribution to the prediction of both disclosure and adult secure attachment style (cf. Hayden et al., 1998). Using hierarchical regression analyses, these hypotheses would be tested by entering the two variables of identity formation in the first step and the two sources of social support in the second. Hypotheses The intention of this study was to propose a flexible process-oriented approach to gay male identity formation and to focus on the links between identity processes and two types of variables that represent the intrapersonal and interpersonal domains: adult attachment style and social support. On the basis of the literature review and the theoretical framework, seven major hypotheses were suggested: (a) that gay male identity variables would be positively related to secure adult attachment style; (b) that support from friends, but not from family, would be positively associated with adult attachment style; (c) that the two sets of variables—identity formation and friend support—would each make an independent contribution to the prediction of secure adult attachment style; (d) that friend support and the two family support variables would be positively related to disclosure; (e) that the effect of general family support on disclosure would be mediated by accepting family attitudes toward same-sex orientation; (f) that self-definition and self-acceptance would be positively related to disclosure; and (g) that the two sets of variables—identity formation and social support from friends and family— would each make an independent contribution to the prediction of disclosure. In addition, it was proposed that the age of first disclosure to parents would be higher in Israel in comparison with the United States. METHOD Participants Much effort was invested in reaching a more varied and representative sample than had been obtained in previous studies of Israeli gay men (Elizur and Ziv, manuscript submitted for publication; Ravitz, 1981). The participants were recruited at different types of gay meeting places, clubs, and associations, at various events that were organized by these associations, at HIV Testing Clinics, and through the friendship networks of four research assistants who live in different
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
Gay Male Identity Formation Processes
January 25, 2001
19:37
Style file version Nov. 19th, 1999
151
places in Israel. We also used newspaper advertisements because they reach a more representative sample than do sources within the gay community (Harry, 1986). According to the participants’ responses to a question concerning their recruitment to the study, 43% reported that they had received the questionnaire through a friend (a research assistant or another participant), 32% had found the questionnaire at a gay meeting place, 16% were recruited through newspaper advertisements, and 9% in a different way. The anonymous questionnaires were completed without any material incentive and returned by mail. The return rate was 30%, but this is an approximation based on the number of questionnaires received relative to those that had been distributed. The figure would be lower if we had information about the number of potential participants who had read the advertisements but did not respond, or friends who had been asked but refused to participate. The figure would be higher if we had information about the number of questionnaires that were left behind in gay meeting places and were not personally examined. Though the low estimated figure appears to threaten the external validity of the research, it needs to be considered in view of the inherent difficulties of engaging “hidden populations” in research. In fact, a similar return rate characterized previous research with gay males and lesbians (Kurdek, 1988). The sample consisted of 121 men, 95% of whom identified themselves as Jewish, whereas an additional 3% identified themselves as Jewish by birth and atheists by choice. There was one Christian and one Buddhist, and the latter was most probably Jewish by birth. Because there is no demographic data concerning the gay, bisexual, and lesbian population in Israel, we give in parenthesis the comparative figures of Jewish men (20+) in Israel for the same data gathering time period (Central Bureau of Statistics, 1996). Of our sample, 89% (51%) were born in Israel, 3% (30%) in Europe, North and South America, and 8% (19%) in North Africa and Asia. The average age was 32: 18% (14%) were 23–25 years old, 54% (22%) were 26–34 years old, 20% (22%) were 35–44 years old, and 8% (42%) were 45–72 years old. Of the participants, 2% (11%) studied up to 8 years, 34% (43%) studied for 9–12 years, 35% (21%) for 13–16 years, and 29% (24%) for 17 years or more. They reported the following monthly salaries: 1% had no income, 8% earned less than 1000 IS, 10% earned 1000–2000 IS, 22% earned 2000–3000 IS, 20% earned 3000–4500 IS, 20% earned 4500–6000 IS, and 21% earned over 6000 IS. At that time the average monthly salary per person was 2448 IS and 4544 IS per main provider. Of the participants, 77% (60%) lived in the central part of Israel, 19% (23%) in the northern part, 1% (14%) in the southern part, and 3% (2%) in settlements outside the green line. Of the sample, 85% described themselves as secular Jews, 9% as observant Jews, and 2% as Orthodox Jews. The comparison with the general population of Israeli Jewish men indicates that the participants were relatively younger, more educated, with higher income, and more likely to be Israeli born, secular, and living in either the central or northern parts of Israel.
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
19:37
152
Style file version Nov. 19th, 1999
Elizur and Mintzer
Procedure Most of our research instruments were tested and refined in a previous study of Israeli gay men (Elizur and Ziv, manuscript submitted for publication). For this study, our four research assistants completed the questionnaires. Based on their comments, some minor modifications were introduced, and it was decided that the questionnaires should not be filled in the presence of a researcher. They were to be returned using a self-addressed envelope, with postage paid by the researchers. A cover letter provided information about the general research objective, which was to learn about ways of life of gay men in Israel, including their personal, familial, and social conditions. They were asked to complete the questionnaire alone and at one sitting. There was space for comments and the feedback was positive, with many comments emphasizing the importance of researching the gay community in Israel. Measures A Biographical Questionnaire The respondents were asked to provide sociodemographic information regarding age, place of birth, residential area, education, income, and religion. In addition, they were also asked to indicate at what ages they had disclosed their sexual orientation to each of their parents, if such disclosures had occurred. Gay Male Identity Three instruments were used to assess different aspects of gay identity. The measure for self-definition was based on the classification originally proposed by Kinsey et al. (1948). They used a 7-point rating to assess how participants defined their same-gender and heterosexual responsiveness on a continuum that ranges from exclusive heterosexual reactivity to exclusive same-gender reactivity. This classification has been used by researchers to assess gay self-definition (Ravitz, 1981; Wells and Kline, 1987) and gay and bisexual male sex identity (SavinWilliams, 1995a). Expectable significant correlations, which were found between this classification and variables such as disclosure (Ravitz, 1981) and number of gay partners (Savin-Williams, 1995a), support the use of this measure. The scale had already been translated to Hebrew (Ravitz, 1981), and the question that was asked was, “How do you define yourself?” Only those who classified themselves in the highest three categories were included. These categories were predominantly gay, but more than incidentally heterosexual; predominantly gay, only incidentally heterosexual; and exclusively gay. Gay self-acceptance was assessed by a 13-item, 5-point questionnaire that was developed by Bell and Weinberg (1978). They found that scores of self-acceptance
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
19:37
Style file version Nov. 19th, 1999
Gay Male Identity Formation Processes
153
discriminated among different groups of gay men and lesbians, which indicates that the scale has discriminant validity. They also reported both very good testretest reliability over a 6-month period as well as convergent validity. The scale was translated into Hebrew and then retranslated into English by three judges. They compared the different Hebrew versions before they reached a consensus. Sample items are, namely, to what extent do you think same-gender orientation is normal like heterosexuality; to what extent do you regret being gay; to what extent do you accept yourself as gay. We created a unified response scale by asking the respondents to indicate their degree of agreement with each item, ranging from very much to not at all. Testing this Hebrew version, we found very good internal consistency reliability (α = .81). Disclosure was assessed by a 20-item questionnaire that was developed in Israel by Ravitz (1981). There are four dichotomous questions that ask about disclosure to each parent and to female and male heterosexual friends, and 16 items on a 5-point scale that ask about verbal and behavioral modes of disclosure with different people and in a variety of social situations. Verbal disclosure is assessed by questions that ask to what extent the respondent shares intimate information with a heterosexual friend, such as feelings of same-gender attraction toward particular people. Two items that illustrate the behavioral mode of disclosure ask to what extent the respondent has appeared in public/family meetings with an intimate partner. The response range for all disclosure items is from never to usually. The instrument has content validity, and its items encompass different aspects of the coming out process that have been described in the literature. Like Ravitz (1981), we found that the questionnaire has excellent internal consistency reliability (α = .91). The two questions on the scale that ask about disclosure to each parent were also analyzed separately as indicators of family knowledge of same-sex orientation. The three identity variables were positively interrelated (see Table I); an expectable finding that is consonant with our assumption that they reflect a common trajectory of gay male identity formation. At the same time the sizes of the correlation coefficients were not large, which again is consonant with the view that Table I. Means, Standard Deviations, and Correlation Coefficients for Observed Variables Variables Self-definition Self-acceptance Disclosure Secure attachment index Accepting family attitudes General family support Friends support Family knowledge ∗p
M 4.36 4.05 3.10 0.00 5.57 2.97 4.04 3.18
SD
1
2
3
4
1.10 — .34∗∗∗ .29∗∗∗ .19∗ 0.69 — .27∗∗ .33∗∗∗ 0.92 — .28∗∗∗ 0.91 — 1.78 0.84 0.65 1.83
< .05; ∗∗ p < .01; ∗∗∗ p < .001 (n = 121).
5
6
.09 −.07 .08 −.16 ∗∗∗ .36 .20∗ .13 .15 — .51∗∗∗ —
7
8
.00 .13 .41∗∗∗ .36∗∗∗ .44∗∗∗ .33∗∗∗ —
.34∗∗∗ .29∗∗ .65∗∗∗ .18∗ .20∗ .10 .04 —
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
154
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
the three measures reflect different aspects of this overall process and are not redundant. Attachment Styles Attachment style was assessed with two instruments that were based on the descriptions of Hazan and Shaver (1987) on how people typically feel in close relationships. Participants were presented with three descriptions of attachment styles and asked to choose the description that best described their own general feelings and cognitions in close adult relationships (e.g., Getting closer to others is easy for me; I feel uncomfortable getting close to others; I am often afraid that the other person will not want to stay with me). This instrument has been previously used and validated in attachment research (Collins and Read, 1990; Feeney and Noller, 1990). It was translated to Hebrew by Mikulincer et al. (1990) and consequently used in many studies in Israel (Mikulincer and Erev, 1991; Mikulincer and Nachshon, 1991). The different studies have repeatedly demonstrated that the typology has high predictive validity and construct validity. For our research we created a dummy variable of secure versus insecure attachment style. Because measures of attachment are not always consistently related (Holtzworth-Munrow et al., 1997), and because the typological instrument is basically a single-item measure, we decided to also use a 15-item, 7-point questionnaire (5 items per attachment style) that was constructed in Israel by separating the items of the descriptions of Hazan and Shaver (1987) (for more details, see Mikulincer et al., 1990). The respondents were asked how much each item characterized their feelings, wishes, and behaviors in close adult relationships. The scales were shown to have adequate levels of internal consistency reliability, and the classification of participants in these two ways is highly similar (Mikulincer and Erev, 1991; Mikulincer and Nachshon, 1991). We took out one item of the anxious-ambivalent subscale (i.e., I want an absolute relationship with certain people) because it was too general, not very clear, and did not fit with the other items that were concerned with the person’s actual experiences in relation to others. This item was not significantly correlated with other items of the subscale or with the dummy variable of security of attachment style. But even though this modified subscale achieved very good internal consistency, the internal consistencies of the other subscales were rather low in our study (α = .78, .58, and .45 for the ambivalent, avoidant, and secure attachment styles, respectively). Therefore, we used the 14 items together as one general measure of secure attachment style. This procedure has face validity because all 14 items have a shared meaning of experiencing security/insecurity in close relationships. Indeed, the three subscales were significantly correlated (r = .36, p < .001; r = −.22, p < .05; and r = −.45, p < .001, for the correlations between ambivalent and avoidant styles, between ambivalent and secure styles, and between avoidant and secure styles, respectively).
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
19:37
Style file version Nov. 19th, 1999
Gay Male Identity Formation Processes
155
The secure attachment style scale was found to have adequate internal consistency (α = .73) and high correlations with each of the subscales (r = −.71, p < .001; r = −.81, p < .001; and r = .74, p < .001, for the correlations with the ambivalent, avoidant, and secure attachment styles, respectively). An examination of the scale showed that it has a normal distribution with a skewness of 0 and kurtosis of −.62. The high correlation between the attachment scale and the secure attachment typology (r = .63, p < .001) attested to the concurrent validity of the scale. The attachment typology has been more frequently used in research, whereas the attachment scale assesses the respondents’ model of security in attachment with more sensitivity. Because the two assessment methods measure the same construct, we reduced redundancy and experiment-wise (Type I) error by forming a composite index of secure attachment. To create the index we averaged the Z scores of the two attachment measures, thereby creating a centered variable that was used throughout the data analysis. The secure attachment index was highly correlated with both the attachment typology (r = .90) and the attachment scale (r = .90). Social Support From Family and Friends Procidano and Heller’s scales of Perceived Social Support from Family and Perceived Social Support From Friends (Procidano and Heller, 1983) have been widely used in research, in Israel as well, to measure these two sources of social support. The two scales are composed of similarly phrased items that refer to the participant’s perception of support from either family or friends. The research by Procidano and Heller (1983) indicates that each of these 20 items, on 5-point scales, has excellent internal consistency reliability (.90 and .88 for the family and friends’ support scales, respectively) and construct validity. Cronbach’s alphas in our study were .93 for each of the social support scales. Family Supportive Attitudes Toward Same-Sex Orientation This scale was developed by Ross (1985) to measure actual and anticipated societal reactions to same-gender orientation. We used the part that pertains to the family. A list of seven family members (i.e., mother, father, brother, sister) was presented, and the respondents were asked to rate the actual or putative responses of each family member to his sexual orientation, using a 9-point response scale that ranges from acceptance to rejection. Ross (1985) tested this scale in two societies. Convergent validity was supported by the finding of a positive association between this questionnaire and another scale that measured perceived social attitudes toward the respondents’ same-gender orientation. A negative association with a measure of the respondents’ concealment of their sexual orientation provided further support of construct validity. The scale has very good internal consistency reliability: Cronbach alpha for the original full scale was .85, whereas we found an alpha of
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
156
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
.86 when using the shortened family scale. The expectable positive correlation (r = .51, p < .001) that was found between the shortened family scale and Procidano and Heller’s scale of perceived social support from family, supported the construct validity of the former scale. To simplify the differentiation between the two types of family support variables, we termed this scale as family acceptance of same-sex orientation, or family acceptance.
RESULTS Overall, the intercorrelations among variables, together with the regression and mediation analyses, supported all the research hypotheses, using one-tailed tests of significance. Prior to the presentation of these results, let us take a comparative look at the characteristics of our sample. Looking at the definition of the participants, we found that 4% considered themselves to be mainly gay but also significantly heterosexual, 23% were predominantly gay and only incidentally heterosexual, and 73% were exclusively gay. Sixty-three percent disclosed to their mothers at the average age of 22.4, whereas 51% disclosed to their fathers at the average age of 22.5. As expected, these figures are considerably lower than in the United States, where D’Augelli and Hershberger (1993) found that 75% of their lesbian, gay, and bisexual youth sample told a parent about their orientation at the age of 17. The difference between Israeli and American samples is also apparent when the comparison is made against the figures that were reported by Savin-Williams (1995b; 1998). With regard to the distribution of attachment style, 17% were characterized as avoidant, 23% as ambivalent, and 60% as having a secure style. A comparison with results from a predominantly heterosexual sample in the United States that found 24% avoidant, 20% ambivalent, and 56% secure (Hazan and Shaver, 1987), and an Israeli sample that found 28% avoidant, 15% ambivalent, and 57% secure (Mikulincer and Erev, 1991), produced nonsignificant chi-squares. The intercorrelations that are presented in Table I were used to test the first two hypotheses concerning attachment style. The first hypothesis was confirmed by the finding that the secure attachment index was positively correlated with each of the identity formation variables. There was also confirmation for the second hypothesis: perceived social support from friends was significantly associated with the attachment index, whereas general family support and family acceptance of same-sex orientation were not significantly related to secure attachment. We then conducted hierarchical regression analysis to predict the secure attachment index, assessing in the first step the relative contribution of the identity variables, and in the second step the additional contribution of friends and family support. The first step explained 15% of the variance of the secure attachment index ( p < .001), the second step explained an additional 8.4% of the variance, and
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
Gay Male Identity Formation Processes
January 25, 2001
19:37
Style file version Nov. 19th, 1999
157
the final equation explained 23.4% of the variance, R = .48, F(5, 113) = 6.59, p < .001. Both self-acceptance and disclosure were significant in the first step, whereas only self-acceptance and friend support were significant at p < .005 in the second step (standardized betas of .27 and .28 in the final equation, respectively). These results support the third hypothesis that both identity formation, the variable of self-acceptance in particular, and friend support have an independent effect on attachment style. Family support does not appear to make an independent contribution to the prediction of attachment style. The other major hypotheses concerning the prediction of disclosure were also supported. Testing the fourth hypothesis, we found that friend support and both forms of family support were significantly associated with disclosure. As expected, the family support variables were not significantly associated with the other two variables of identity formation. The differences between the correlations of the family support variables with disclosure, and the corresponding correlations with self-definition and self-acceptance, were significant using the Fisher-Z transformation (Z = 1.90–2.77, p < .05). We followed up this finding by testing the four conditions of mediational analysis (Baron and Kenny, 1986) to examine whether the association between general family support and disclosure was mediated by the more specific form of family acceptance of same-sex orientation. Looking at Table I we see that the first three conditions were met by the following significant correlations: family support with disclosure, family support with family acceptance, and family acceptance with disclosure. The forth condition is that when the effects of the mediator are controlled, the previously significant association between the independent and dependent variables would become nonsignificant or at least be significantly reduced (partial mediation). As predicted, the correlation between family support and disclosure was significantly reduced and became nonsignificant (partial r = .03) when controlling for family acceptance of same-sex orientation. In a regression equation, family support was nonsignificant whereas family acceptance made a significant contribution to the prediction of disclosure (unstandardized Bs of .03 and .17, respectively). Thus, evidence was found to support the fifth hypothesis relating to a mediational model. To check ourselves, we also tested an alternative mediational hypothesis that family support would have a direct effect on disclosure, which in turn would have an effect on family acceptance. This alternative hypothesis was disconfirmed: the correlation between family support and family acceptance was essentially unchanged when controlling for disclosure (r = .51 and .48, respectively). We then examined the associations of self-definition and self-acceptance with the disclosure scale and with family knowledge of same-sex orientation, which is part of the disclosure scale. In line with the sixth hypothesis, these two highly interrelated disclosure variables were found to be significantly correlated with selfdefinition and self-acceptance. The next step was to perform hierarchical regression in which we first entered the two identity variables and then friend and family
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
158
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
support (following the mediational analysis, only family acceptance of same-sex orientation was entered). Both blocks were significant ( p < .001), the first one explaining 11.5% of the variance of disclosure, the second explaining another 18.4% of the variance, and the final equation explaining 29.7% of the variance, R = .55, F(4, 114) = 12.02, p < .001. The two identity variables were significant in the first block, but only self-definition remained significant in the final equation (standardized β .22, p < .01). Both family and friend support were significant in the final equation (standardized β .19, p < .05 and standardized β .31, p < .001, respectively). These results, which are consonant with the seventh hypothesis, support our conceptual framework of identity formation. They indicate that when gay males subscribe to a more definite view of their sexual orientation and learn to accept themselves, they are more likely to disclose themselves to family members and heterosexual friends. It is also suggested that beyond the process of identity formation, both friend support and family attitudes toward same-sex orientation make an independent contribution to the prediction of disclosure.
DISCUSSION The application of a multidimensional process-oriented framework to the study of gay male identity formation and the subsequent links that were found between identity processes, secure attachment, and different types of social support provided fresh insights that will be discussed in two major parts. The first part will examine and elaborate on findings that are related to attachment. We shall apply a theoretical approach that distinguishes between two basic dimensions of attachment and draws on current knowledge of gay male identity formation to explain the regression effects of self-acceptance and friends’ support on attachment. The second part of the discussion will focus on the process of disclosure, especially on the most novel finding in this section: the effect of general family support and its mediation by supportive family attitudes toward same-sex orientation. The overall relationship between processes of gay male identity formation and attachment style is an original finding, the significance of which will first be considered in light of previous findings that linked attachment style, same gender orientation, and identity formation with mental health. Empirical evidence indicates that gay youth are at greater risk for major depression, generalized anxiety disorder, conduct disorder, substance abuse and dependence, multiple disorders, suicidal behaviors, sexual risk-taking, and poor general health maintenance than their heterosexual peers are (Fergusson et al., 1999; Lock and Steiner, 1999). These problems are related to the substantial developmental challenges faced by these youth, insufficient social support systems, and high risk of sexual orientation victimization (D’Augelli, 1998; Hershberger and D’Augelli, 1995). Although the comparative research of adults is more limited, research studies indicate that gay
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
Gay Male Identity Formation Processes
January 25, 2001
19:37
Style file version Nov. 19th, 1999
159
men’s experiences place them at increased risk for some mental health problems while protecting them from other mental health problems (Martin and Dean, 1993; Meyer, 1995; Ross, 1990; Rothblum, 1994). Overall, the evidence suggests a considerable decrease in the psychiatric vulnerability of adult gay men (Coyle, 1993; Frable et al., 1997). One of the contributing factor appears to be the consolidation of an affirmative sense of gay male identity, which has been shown to be related to self-esteem and to different indices of psychological adjustment (Carlson and Steuer, 1985; Elizur and Ziv, manuscript submitted for publication; Hammersmith and Weinberg, 1973; Savin-Williams, 1990; Schnitt and Kurdek, 1984). The present finding that links identity formation with attachment may be indicative of a potentially important risk/protective mechanism that explains the association between gay male identity and mental health. Attachment representations are not only linked to psychopathology (del Carmen and Huffman, 1996; Fonagy et al., 1996; van Ijzendoorn and Bakermans Kranenburg, 1996; West et al., 1993), but also predict relationship stability (Kirkpatrick and Davis, 1994) and relationship satisfaction (Davila et al., 1999). Following the epigenetic principle, our results suggest that the consolidation of gay male identity may have an effect on mental health that is mediated by the development of a secure adult attachment model and the formation of stable intimate relationships. This hypothesis, which needs to be tested by longitudinal studies, demonstrates how the linking of gay male identity formation with attachment theory can open up new and important directions for developmental research. The multidimensional study of gay male identity enabled us to proceed with a more finely tuned analysis of the relationship between identity formation and attachment style. The significant positive correlations between the three identity variables and attachment style can be explained by the common variance of the identity variables. According to our model this common variance is not just an artifact of method variance, but reflects a common trajectory of gay male identity formation that is related to attachment style. Moreover, the next step of the data analysis indicated that self-acceptance, rather than self-definition or disclosure, significantly predicted attachment in both steps of the hierarchical equation. Assuming that the results of the regression analysis reflect a theoretically meaningful effect (Asher, 1997), this interesting finding can be interpreted in light of Griffin and Bartholomew’s (Griffin and Bartholomew, 1994) distinction between two basic dimensions of attachment: positivity of self-concepts and positivity of interpersonal orientations. Applying their theoretical frame to gay males, whose identity is formed as they go through a challenging period during which they face heterosexist attitudes both inside and outside, the development of acceptance toward their own sexual orientation contributes to the consolidation of a positive self-concept. Hence, self-acceptance has a unique relationship with secure attachment that extends beyond the more general association between identity formation and attachment style.
P1: FMN Archives of Sexual Behavior [asb]
160
PP016-291181
January 25, 2001
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
The theoretical frame of Griffin and Bartholomew (1994) can also be applied to the finding that a social support variable was found to make an independent contribution to the prediction of attachment in the regression equation. According to this interpretation, social support is associated with a positive orientation toward the other, which is the second aspect of attachment style in intimate relationship. The finding that attachment style was predicted by friends’, rather than family support, was expected in light of the different role that is played by family and friends during the consolidation of male gay identity (Bernstein, 1990; Herdt, 1992). A parallel difference between these two sources of support was also found in research with gay and lesbian cohabitating couples. They were more frequently supported by friends and partners than by family members, and in fact, they often developed their relationship without the knowledge of their families (Kurdek, 1988). Apparently, the role of family support is different in heterosexual and samegender couples, and our findings suggest that a parallel difference may influence the formation of secure attachment styles in intimate relationships. According to our theoretical analysis of the attachment style construct, these differences are revealed when researchers focus on the state-like aspects of attachment style (i.e., current adult attachment style), and are concealed when attachment style is viewed as a trait (i.e., recollections of early interaction patterns). As we move on to the second part of the discussion, with respect to the prediction of disclosure, both family and friends’ support were found to make an independent contribution to the regression equation. Moreover, the support variables, which were entered in the second step of the hierarchical analysis, made a significant addition to the prediction of disclosure by the identity variables, which were entered in the first step. Following the conceptual framework of this study, each block in the regression analysis is interpreted as representing a different type of effect on disclosure. In other words, the findings support the theoretical proposition that both the inner development of the gay male, as well as the feeling of being supported by family and friends, facilitate disclosure. The findings concerning family support are especially interesting in light of the prevalent tendency of stage models to deemphasize the family context and to describe gay male identity formation in terms of inner development and relationships with same-gender peers. Although these models can explain the lack of association between family support and both self-definition and self-acceptance, the link between family support and disclosure strains these models to their limits. Evidently, the role that is played by family dynamics in the consolidation of samegender identity is much more complex and convoluted in comparison with the family’s influence on heterosexual identity development. Most families have no available rules, roles, or language to help them manage the disturbance that occurs when their child’s same-gender orientation is unveiled (Devine, 1984). Moreover, because the boundaries between families and their social context are not impermeable, family members’ coping with their own negative feelings is made all the
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
Gay Male Identity Formation Processes
January 25, 2001
19:37
Style file version Nov. 19th, 1999
161
more difficult by heterosexist social practices. Discriminatory attitudes against individuals with same-gender orientation are more prevalent in Israel than in other countries in Europe and North American (Lieblich and Friedman, 1985; Weishut, in press). One of the factors that is probably implicated in this situation is the orthodox Jewish view of same-gender sexuality as an abomination. This viewpoint is promoted by Israel’s religious parties and institutions, which, because of historical and political reasons, have a profound effect over family life. The other influential factor is the army and related social norms that emphasize expressions of masculine strength in the military realm and deprecate signs of emotional vulnerability (Elizur, 1996). The finding that ages of disclosure to parents occurred later in our Israeli sample in comparison with U.S.-based studies can be thus explained by Israel’s greater degree of heterosexism. Though this particular information was obtained retrospectively, research on autobiographical memory demonstrates that effective investigations of significant life points, particularly with respect to early adulthood, can be undertaken using current research methods (Conway and Rubin, 1993; Rubin, 1996; Rubin et al., 1998). However, this cross-cultural comparison should be interpreted cautiously. The later age of disclosure to parents in Israeli society could also be a cohort effect. The review of Savin-Williams (1998) studies of family disclosure by lesbian, gay, and bisexual youths, concludes by emphasizing the strong and consistent empirical support for a cohort effect. At the same time he also notes the existence of social and cultural effects. Growing up with prejudice in both their social and family context, gay males need to navigate their way between two perils, neither of which can be evaded without risking the other. Disclosure may bring about a massive disconfirmation of their feelings or family rejection or both, whereas nondisclosure increases isolation and alienation (Bozett and Sussman, 1989). Consequently they tend to hide their sexual orientation from their families, thereby cutting them off from an important source of support during a crucial phase in the development of their identity (Bernstein, 1990). When they begin to come out, they usually find it easier to talk with peers and selected family members so that their family support system is not fully available until later on in the process (Herdt, 1992; Savin-Williams, 1998). Yet, our findings suggest that the working through of this unique family process is associated with the consolidation of same-gender identity. In particular, the mediational analysis suggests that the general variable of family support has a distal effect on disclosure that is mediated by the more proximal effect of supportive family attitudes toward same-sex orientation. These findings are consonant with the view that supportive families are more likely to adopt supportive attitudes toward their sons’ sexual orientation, which in turn promote disclosure to family and heterosexual friends. They are also supported by qualitative research and clinical practice, which indicate that family values, relationships and behaviors toward same-sex orientation affect the coming out experience and are also changed by it (Ben-Ari, 1995; Holtzen et al., 1995; Newman and Muzzonigro,
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
19:37
162
Style file version Nov. 19th, 1999
Elizur and Mintzer
1993; Savin-Williams, 1989b; Strommen, 1989; Troiden, 1989; Wells and Kline, 1987). Limitations and Implications There are several limitations to the current research. First, the external validity of the findings is limited by the small sample size (n = 121) of Israeli gay men and by the low return rate of questionnaires (30%), which may reflect problems of self selection. Moreover, even though we recruited participants in different sites, using a variety of methods to reach people of diverse background, there is no claim that the sample is representative. Hence, the evaluation of the proposed identity formation framework should be reserved until further studies are done not only in Israel but also in other cultures and with lesbian and bisexual participants as well. All the same, one should keep in mind the real world difficulty of engaging “hidden populations” in research, given the social stigma and sense of shame associated with same-sex orientation (Savin-Williams, 1995a,b). Second, because of the cross-sectional nature of the data, there could be no testing of causal connections among variables. A longitudinal perspective is necessary to observe the unfolding of identity formation processes, and such a design would provide for a more rigorous testing of the mediational hypothesis. Third, there is a problem of common method variance that occurs when relying on self-report measures. Hence, the relationship between family variables and identity formation should be tested by subsequent research, using data from family sources. We also think that qualitative studies of such “high context” issues of identity development that involve cultural and familial factors can help to refine descriptive categories and expand the theoretical framework (Sells et al., 1995). Finally, we note the need for creating a more refined scale to measure self-definition of same-gender orientation. The scale of Kinsey et al. (1948) is too categorical for that purpose, and in the case of bisexual respondents it may confound between exclusivity and consolidation of the self-defining narrative. The assessment of sexual orientation in previous research has been inconsistent and frequently absent (cf. Chung and Katayama, 1996), whereas in regard to self-definition there is presently no established measure. Notwithstanding these limitations, this study supports the applicability of the proposed theoretical framework and the proposition that gay identity should be studied in relation to a multilevel matrix that is concerned with the person’s self, family, and social relations. In today’s complex societies, where multiple identities are inevitable, such an approach appears to provide a more flexible alternative to stage models’ elaboration of one, presumably universal, developmental sequence. The three developmental processes that were focused upon in this study—selfdefinition, self-acceptance, and disclosure—are based on previous studies of gay identity formation. However, we carefully defined these processes in more general
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
January 25, 2001
19:37
Gay Male Identity Formation Processes
Style file version Nov. 19th, 1999
163
terms, thereby supporting a broad conceptual approach that can be applied with some modification to studies of ethnic and racial identities (cf. Phinney, 1990; Phinney and Kohatsu, 1997), as well as be integrated with general theory and research on identity development. The theoretical and empirical links that were made between attachment models and identity development, especially with respect to the working through of self-acceptance, have some interesting implications that can be explored by future studies. Because persons with secure attachment models are more likely to maintain stable and satisfying intimate relationships, we would expect to find that the consolidation of gay male identity is related to the ability to be in such relationships. This type of hypothesis suggests a way for integrating future studies of gay men, lesbians, and bisexual women or men’s identities with the theoretical domains of attachment and close relationships. Social support provided by family and friends was related to both identity development and secure attachment. The particular value of friend support in gay and lesbian samples has been noted earlier, and our research supports this notion by showing that only the support provided by friends was predictive of attachment. On the other hand, the variable of family attitudes toward same-sex orientation was shown to contribute to the prediction of disclosure beyond the contributions of friend support and the other variables of identity formation. Family attitudes were also found to mediate the relationship between family support and disclosure. Theoretically, these results need to be interwoven with studies of disclosure in families (Savin-Williams, 1998), and with the finding that family support and self-acceptance in concert partly mediate the relationship between victimization and mental health (Hershberger and D’Augelli, 1995). Together, they suggest new possibilities for linking family system theory, the field of identity formation, and the study of risk factors in the development of lesbian, gay, and bisexual youths. ACKNOWLEDGMENT This study was supported by a grant from the Israel Foundations Trustees. The authors thank Bruce Oppenheimer and Daniel J. N. Weishut for their thoughtful reading of the manuscript and the helpful editorial suggestions. REFERENCES Asher, J. W. (1997). The role of measurement, some statistics, and some factor analysis in family psychology research. J. Fam. Psychol. 11: 351–360. Baron, R. M., and Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J. Pers. Soc. Psychol. 51: 1173–1182. Baumeister, R. F. (1986). Identity: Cultural Change and the Struggle For Self, Oxford University Press, Oxford.
P1: FMN Archives of Sexual Behavior [asb]
164
PP016-291181
January 25, 2001
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
Bell, A. P., and Weinberg, M. S. (1978). Homosexualities: A Study of Diversity Among Men and Women, Simon and Schuster, New York. Ben-Ari, A. (1995). The discovery that an offspring is a gay: Parents’, gay men’s, and lesbians’ perspectives. J. Homosex. 30: 89–112. Benson, M. J., Harris, P. B., and Rogers, C. S. (1992). Identity consequences of attachment to mothers and fathers among late adolescents. J. Res. Adolesc. 2: 187–204. Bernstein, B. E. (1990). Attitudes and issues of parents of gay and lesbians and implications for therapy. J. Gay Lesb. Psychother. 1: 37–53. Berzon, B. (1979). Developing a positive gay identity. In Berzon, B., and Leighton, R. (Eds.), Positively Gay, Celestial Arts, Millrose, CA, pp. 1–14. Bowlby, J. (1973). Attachment and Loss: Separation, Anxiety and Anger, Basic Books, New York. Bozett, F. W., and Sussman, M. B. (1989). Homosexuality and family relations: Views and research issues. Mar. Fam. Rev. 14: 1–8. Boxer, A. M., and Cohler, B. J. (1989). The life course of gay and lesbian youth: An immodest proposal for the study of lives. J. Homosex. 17: 315–355. Bronfenbrenner, U. (1979). The Ecology of Human Development, Harvard University Press, Cambridge, MA. Carlson, H. M., and Steuer, J. (1985). Age, sex role categorization and psychological health in American homosexual and heterosexual men and women. J. Soc. Psychol. 125: 203–211. Cass, V. C. (1979). Homosexual identity formation: A theoretical model. J. Homosex. 4: 219–235. Central Bureau of Statistics (1996). Israel’s Statistics Yearbook, 1996 (No. 47), Central Bureau of Statistics, Jerusalem, Israel. Chung, Y. B., and Katayama, M. (1996). Assessment of sexual orientation in lesbian/gay/bisexual studies. J. Homosex. 30: 49–62. Collins, N. L., and Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. J. Pers. Soc. Psychol. 58: 644–663. Conway, M. A., and Rubin, D. C. (1993). The structure of autobiographical memory. In Collins, A. F. (Ed.), Theories of Memory, Lawrence Erlbaum, Hove, England, pp. 103–137. Cox, S., and Gallois, C. (1996). Gay and lesbian identity development: A social identity perspective. J. Homosex. 30: 1–30. Coyle, A. (1993). A study of psychological well-being among gay men using the GHQ-30. Br. J. Clin. Psychol. 32: 218–220. D’Augelli, A. R. (1998). Developmental implications of victimization of lesbian, gay, and bisexual youths. In Herek, G. M. (Ed.), Stigma and sexual orientation: Understanding prejudice against lesbians, gay men, and bisexuals. Psychological perspectives on lesbian and gay issues, Vol. 4, Sage Publications, Inc., Thousand Oaks, CA, pp. 187–210. D’Augelli, A. R., and Hershberger, S. L. (1993). Lesbian, gay, and bisexual youth in community settings: Personal challenges and mental health problems. Am. J. Commun. Psychol. 21: 421– 449. Davila, J., Karney, B. R., and Bradbury, T. N. (1999). Attachment change processes in the early years of marriage. J. Pers. Soc. Psychol. 76: 783–802. del Carmen, R. D., and Huffman, L. (1996). Epilogue: Bridging the gap between research on attachment and psychopathology. J. Consult. Clin. Psychol. 64: 291–294. Devine, J. L. (1984). A systematic inspection of affectional preference orientation and the family of origin. J. Soc. Work Hum. Sex. 2: 9–17. Eliason, M. J. (1996). Identity formation for lesbian, bisexual and gay persons: Beyond a “minoritizing” view. J. Homosex. 30: 31–58. Elizur, Y. (1996). “Inside” consultation through self-differentiation: Stimulating organizational development in the IDF’s care of intractable, war-related, traumatic disorders. Human Systems 7: 189–210. Elizur, Y., and Hirsh, E. (1999). Psychosocial adjustment and mental health two months after coronary artery bypass surgery: A multisystemic analysis of patients’ resources. J. Behav. Med. 22: 157–177. Elizur, Y., and Minuchin, S. (1989). Institutionalizing Madness: Families, Therapy and Society, Basic Books, New York. Elizur, Y., and Ziv, M. (manuscript submitted for publication). Gay Male Identity, Family Support and Acceptance, and Mental Health: A Path Model, The Hebrew University of Jerusalem.
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
Gay Male Identity Formation Processes
January 25, 2001
19:37
Style file version Nov. 19th, 1999
165
Fassinger, R. E., and Miller, B. A. (1996). Validation of an inclusive model of sexual minority identity formation on a sample of gay men. J. Homosex. 32: 53–78. Feeney and Noller, (1990). Attachment style as a predictor of adult romantic relationships. J. Pers. Soc. Psychol. 58: 281–291. Fergusson, D. M., Horwood, L. J., and Beautrais, A.-L. (1999). Is sexual orientation related to mental health problems and suicidality in young people? Arch. Gen. Psychiat. 56: 876–880. Fonagy, P., Leigh, T., Steele, M., Steele, H., Kennedy, R., Mattoon, G., Target, M., and Gerber, A. (1996). The relation of attachment status, psychiatric classification, and response to psychotherapy. J. Consult. Clin. Psychol. 64: 22–31. Fox, R. C. (1995). Bisexual identities. In D’Augelli, A. R., and Patterson, C. J. (Eds.), Lesbian, Gay, and Bisexual Identities Over the Life Span: Psychological Perspectives, Oxford University Press, New York, pp. 48–86. Frable, D. E. S., Wortman, C., and Joseph, J. (1997). Predicting self-esteem, well-being, and distress in a cohort of gay men: The importance of cultural stigma, personal visibility, community networks, and positive identity. J. Person. 65: 599–624. Gonsiorek, J. C. (1995). Gay male identities: Concepts and issues. In D’Augelli, A. R., and Patterson, C. J. (Eds.), Lesbian, Gay, and Bisexual Identities Over the Life Span: Psychological Perspectives, Oxford University Press, New York, pp. 25–47. Griffin, D. W., and Bartholomew, K. (1994). Models of the self and other: Fundamental dimensions underlying measures of adult attachment. J. Pers. Soc. Psychol. 67: 430–445. Hammersmith, S. K., and Weinberg, M. S. (1973). Homosexual identity: Commitment, adjustment, and significant others. Sociometry 36: 56–79. Harry, J. (1986). Sampling gay man. J. Sex Res. 22: 21–34. Hayden, L. C., Schiller, M., Dickstein, S., Seifer, R., Sameroff, S., Miller, I., Keitner, G., and Rasmussen, S. (1998). Levels of family assessment: I. Family, marital, and parent-child interaction. J. Fam. Psychol. 12: 7–22. Hazan, C., and Shaver, P. R. (1987). Romantic love conceptualized as an attachment process. J. Pers. Soc. Psychol. 52: 511–524. Herdt, G. (1992). “Coming out” as a rite of passage: A Chicago study. In Herdt, G. (Ed.), Gay Culture in America: Essays From the Field, The Beacon Press, Boston, pp. 29–67. Hershberger, S. L., and D’Augelli, A. R. (1995). The impact of victimization on the mental health and suicidality of lesbians, gay and bisexual youth. Dev. Psychol. 31: 65–74. Holmes, J. (1996). Attachment, intimacy, autonomy: Using attachment theory in adult psychotherapy, Jason Aronson, New York. Holtzen, D. W., Kenny, M. E., and Mahalik, J. R. (1995). Contributions of parental attachment to gay or lesbian disclosure to parents and dysfunctional cognitive processes. J. Counsel. Psychol. 42: 350–355. Holtzworth-Munrow, A., Stuart, G. L., and Hutchinson, G. (1997). Violent versus nonviolent husbands: Differences in attachment patterns, dependency, and jealousy. J. Fam. Psychol. 11: 314– 331. Jordan, J., Kaplan, A. G., Miller, J. B., Stiver, I., and Surrey, J. (1992). Women’s Growth in Connection: Writings From the Stone Center, Guilford, New York. Kinsey, A. C., Pomeroy, W. B., and Martin, C. E. (1948). Sexual Behavior in the Human Male, Sanders, Philadelphia. Kirkpatrick, L. A., and Davis, K. E. (1994). Attachment style, gender, and relationship stability: A longitudinal analysis. J. Pers. Soc. Psychol. 66: 502–512. Kitzinger, C., and Wilkinson, S. (1995). Transitions from heterosexuality to lesbianism: The discursive production of lesbian identities. Dev. Psycol. 31: 95–104. Krahn, G. L. (1993). Conceptualizing social support in families of children with special health needs. Fam. Process 32: 235–248. Kurdek, L. A. (1988). Perceived social support in gays and lesbians in cohabitation relationships. J. Pers. Soc. Psychol. 54: 504–509. Kurdek, L. A. (1997). The link between facets of neuroticism and dimensions of relationship commitment: Evidence from gay, lesbian, and heterosexual couples. J. Fam. Psychol. 11: 109–124. Laird, J. (1993). Lesbian and gay families. In Walsh, F. (Ed.), Normal Family Processes, 2nd ed., Guilford, New York, pp. 283–328.
P1: FMN Archives of Sexual Behavior [asb]
166
PP016-291181
January 25, 2001
19:37
Style file version Nov. 19th, 1999
Elizur and Mintzer
Lieblich, A., and Friedman, G. (1985). Attitudes toward male and female homosexuality and sex-role stereotypes in Israeli and American students. Sex-Roles 12: 561–570. Lock, J., and Steiner, H. (1999). Gay, lesbian, and bisexual youth risks for emotional, physical, and social problems: Results from a community-based survey. J. Am. Acad. Child Adol. Psychiat. 38: 297–304. Main, M. (1996). Introduction to the special section on attachment and psychopathology: 2. Overview of the field of attachment. J. Consult. Clin. Psychol. 64: 237–243. Martin, J. L., and Dean, L. (1993). Effects of AIDS-related bereavement and HIV-related illness on psychological distress among gay men: A 7-year longitudinal study, 1985–1991. J. Consult. Clin. Psychol. 61: 94–103. Meyer, I. H. (1995). Minority stress and mental health in gay men. J. Health Soc. Beh. 36: 38–56. Michaels, S. (1996). The prevalence of homosexuality in the United States. In Cabaj, R. P., and Stein, T. S. (Eds.), Textbook of Homosexuality and Mental Health, American Psychiatric Press, Washington, DC, pp. 43–63. Mikulincer, M., and Erev, I. (1991). Attachment style and the structure of romantic love. Brit. J. Soc. Psychol. 30: 273–291. Mikulincer, M., Florian, V., and Tolmacz, R. (1990). Attachment styles and fear of personal death: A case study of affect regulation. J. Pers. Soc. Psychol. 58: 273–280. Mikulincer, M., and Nachshon, O. (1991). Attachment styles and patterns of self-disclosure. J. Pers. Soc. Psychol. 59: 321–331. Newman, B. S., and Muzzonigro, P. G. (1993). The effects of traditional family values on the coming out process of gay male adolescents. Adolescence 28: 213–226. Phinney, J. S. (1990). Stages of ethnic identity development in minority group. Psychol. Bull. 108: 499–514. Phinney, J. S., and Kohatsu, E. L. (1997). Ethnic and racial identity development and mental health. In Schulenberg, J., Maggs, J. L., and Hurrelmann, K. (Eds.), Health Risks and Developmental Transitions During Adolescence, Cambridge University Press, New York, pp. 420–443. Procidano, E. P., and Heller, K. (1983). Measures of perceived social support from friends and from family: Three validation studies. Am. J. Commun. Psychol. 11: 1–24. Ravitz, G. (1981). The link between disclosure of homosexual orientation and intimate relationship and cohabitation with homosexual partner. Unpublished master’s thesis, Tel Aviv University, Ramat Aviv, Israel. (Hebrew). Reiss, D. (1981). The Family’s Construction of Reality, Harvard University Press, Cambridge. Ross, M. W. (1985). Actual and anticipated societal reaction to homosexuality and adjustment in two societies. J. Sex Res. 21: 40–55. Ross, M. W. (1990). The relationship between life events and mental health in homosexual men. J. Clin. Psychol. 46: 402–411. Rothblum, E. D. (1994). “I only read about myself on bathroom walls”: The need for research on the mental health of lesbians and gay men. J. Consult. Clin. Psychol. 62: 213–220. Rubin, D. C. (Ed.) (1996). Remembering Our Past: Studies in Autobiographical Memory, Cambridge University Press, New York. Rubin, D. C., Rahhal, T. A., and Poon, L. W. (1998). Things learned in early adulthood are remembered best. Memory Cognition 26: 3–19. Savin-Williams, R. C. (1989a). Parental influences on the self-esteem of gay and lesbian youths: A reflected appraisals model. J. Homosex. 17: 93–109. Savin-Williams, R. C. (1989b). Coming out to parents and self-esteem among gay and lesbian youths. J. Homosex. 18: 1–35. Savin-Williams, R. C. (1990). Gay and Lesbian Youth: Expressions of Identity, Hemisphere, Washington, DC. Savin-Williams, R. C. (1995a). An exploratory study of pubertal maturation timing and self-esteem among gay and bisexual male youths. Dev. Psycol. 31: 56–64. Savin-Williams, R. C. (1995b). Lesbian, gay male, and bisexual adolescents. In D’Augelli, A. R., and Patterson, C. J. (Eds.), Lesbian, Gay, and Bisexual Identities Over the Life Span: Psychological Perspectives, Oxford University Press, New York, pp. 165–189. Savin-Williams, R. C. (1998). The disclosure to families of same-sex attractions by lesbian, gay, and bisexual youths. J. Res. Adolescence 8: 49–68.
P1: FMN Archives of Sexual Behavior [asb]
PP016-291181
Gay Male Identity Formation Processes
January 25, 2001
19:37
Style file version Nov. 19th, 1999
167
Schnitt, J. P., and Kurdek, L. A. (1984). Correlates of social anxiety in college students and homosexuals. J. Pers. Assess. 48: 403–409. Schulz, R., and Tompins, C. A. (1990). Life events and changes in social relationships: Examples, mechanisms, and measurement. J. Soc. Clin. Psychol. 9: 69–77. Sells, S. P., Smith, T. E., and Sprenkle, D. H. (1995). Integrating qualitative and quantitative research methods: A research model. Fam. Process 34: 199–218. Strommen, E. F. (1989). Hidden branches and growing pains: Homosexuality and the family tree. Mar. Fam. Rev. 14: 9–34. Troiden, R. R. (1989). The formation of homosexual identities. J. Homosex. 17: 43–73. van Ijzendoorn, M. H., and Bakermans Kranenburg, M. J. (1996). Attachment representations in mothers, fathers, adolescents, and clinical groups: A meta-analytic search for normative data. J. Consult. Clin. Psychol. 64: 8–21. Wells, J. W., and Kline, W. B. (1987). Self-disclosure of homosexual orientation. J. Soc. Psychol. 127: 191–197. Weishut, D. J. N. (in press). Attitudes toward homosexuality: A study on Israeli students. Isr. J. Psychiatry Relat. Sci. West, M., Rose, M. S., and Sheldon, A. (1993). Anxious attachment as a determinant of adult psychopathology. J. Nerv. Ment. Diseas. 181: 422–427. Yalom, I. D. (1980). Existential Psychotherapy, Basic Books, New York.
P1: FYJ/GBO
P2: FPX/FNV
Archives of Sexual Behavior [asb]
QC: FTK PP016-291182
February 5, 2001
11:33
Style file version Nov. 19th, 1999
Archives of Sexual Behavior, Vol. 30, No. 2, 2001
Which Behaviors Constitute “Having Sex” Among University Students in the UK? Marian Pitts, Ph.D.1,3 and Qazi Rahman, B.Sc.1,2
The aim of this study was to establish which behaviors were considered to constitute sexual relations and to compare a group of undergraduates in the UK with a group in the US. An opportunistic sample of 190 female and 124 male UK undergraduate university students was surveyed by questionnaire. The main outcome measure was percentage of responses to 11 different behaviors believed to constitute “having sex.” The majority of respondents regarded having sex as involving penile–vaginal and penile–anal intercourse. One-third of respondents regarded oral–genital contact as having sex, around 17% regarded touching genitals, whilst 6% regarded oral or other touching of breasts and nipples as constituting having sex. There were significant gender- and age-related differences in responses. These findings broadly support the findings of an earlier US study. It is clear that British students hold divergent opinions about which behaviors do and do not constitute having sex. The age-related trends merit further exploration. Any studies of sex-related behaviors need to specify precisely which are encompassed by the terms used. KEY WORDS: sexual behavior; terminology; gender differences, age differences.
INTRODUCTION A recent paper in Archives of Sexual Behavior reviewed and critiqued some unresolved issues in sexological research (McConaghy, 1999). Among the issues discussed were evaluations of the treatment of sexual offenders, the distribution of reported homosexual feelings in predominantly heterosexually identified individuals, and the validity of self-reports of sexual behavior. Another unresolved 1 Centre
for Health Psychology, Staffordshire University, Staffordshire, UK.
2 Present address: Department of Psychology, Institute of Psychiatry, De Crespigny Park, London, SE5
8AF, UK. whom correspondence should be addressed at Australian Research Centre in Sex, Health and Society, 215 Franklin PE, Melbourne, Victoria 3000, Australia; e-mail: [email protected].
3 To
169 C 2001 Plenum Publishing Corporation 0004-0002/01/0400-0169$19.50/0 °
P1: FYJ/GBO
P2: FPX/FNV
Archives of Sexual Behavior [asb]
QC: FTK PP016-291182
February 5, 2001
170
11:33
Style file version Nov. 19th, 1999
Pitts and Rahman
issue is the extent to which people differ with respect to interactions they regard as having “had sex.” The perceived normative patterns of sexual interactions held by various populations has a number of implications. These include relevance to clinical and public health matters (such as the reduction of risky sexual behavior and primary prevention of sexually transmitted disease); the elucidation of individual differences in sexuality (e.g., gender-related differences and sexual dysfunction); and relevance for research inquiry, particularly given the differing usage of terminology across studies and its subsequent operationalization as factorial or predictive variables in analyses (Bancroft, 1997; Seidman and Reider, 1994; Tiefer and Kring, 1995). A recent study has shed light on US conceptualizations of what does and does not constitute having had sex (Sanders and Reinisch, 1999). The study attracted much controversy because of editorial decisions surrounding it (Tanne, 1999); however, little direct comment has been made about the research findings themselves. The paper reported responses to a question included in a survey conducted in 1991, exploring sexual behaviors and attitudes of nearly 600 students at a Midwestern US university. Considerable variation was found in what was regarded as constituting having sex. Fifty-nine percent of the sample did not regard oral–genital contact as having sex; 19% responded similarly for penile–anal intercourse. Little is known about the conceptions of sex held by people in the UK. The National Survey of Sexual Attitudes and Lifestyle gathered information about sexual behaviors and experiences, but did not specifically address what respondents understood by the term “having sex” (Wellings et al., 1994). Many other studies have also either used an unspecified definition of sexual behavior or contextualized it as penile–vaginal intercourse (Cooper et al., 1998; Seal and Agostinelli, 1996). Differential frequencies of engagement in varying sexual practices are factors in the maintenance of sexual health (Blower and Boc, 1993); individuals’ own construals of what constitutes sex may be equally important. Cognitions and attitudes surrounding sex may modulate the intentional and motivational components that determine risky sexual behavior and condom use. Studies of such components often fail to explicate which behaviors constituted having had sex (e.g., Gold et al., 1992; Ratliff-Cairn et al., 1999; Rosenthal et al., 1996; Sutton et al., 1999). This small-scale study utilized the US questionnaire to gain preliminary data on conceptualizations of sex in the UK and investigated any differences associated with gender and age. METHOD Participants Data were collected from an opportunistic sample of undergraduates. Students were attending a middle-sized university in the central region of the UK and were
P1: FYJ/GBO
P2: FPX/FNV
Archives of Sexual Behavior [asb]
QC: FTK PP016-291182
February 5, 2001
11:33
Sex Constituting Behaviors in the UK
Style file version Nov. 19th, 1999
171
mainly studying Natural and Social Sciences. The majority of students at this university come from within 100 miles radius, which includes the West Midlands and North-west regions of England. Sample size was 314, with 124 males and 189 females (39.5% and 60.2%, respectively). Mean age was 22 years, with a standard deviation of 5.15; 86% of the sample was between the ages of 18 and 26 years. Ninety-four percent described themselves as white, and 97% identified themselves as heterosexual. They had strong beliefs in religion with the majority (77%) describing themselves as Christians, and 18% describing themselves as either atheist or agnostic. These demographics broadly parallel those of Sanders and Reinisch (1999), whose sample comprised 59% females, 92% white, and 96% self-identified heterosexuals. The mean age was 20.7 years. Materials and Procedure Students who were attending lectures on a wide variety of topics within Natural and Social Sciences were invited to complete the survey. Those who volunteered, the majority within each lecture group, completed the survey anonymously and returned it immediately to the researcher. They received no course credit or other incentive for participation. The questionnaire comprised 11 behavioral items relating to having had sex, which were arranged in random order. For each item, respondents were asked to answer Yes or No to the question: “Would you say you had sex with someone if the most intimate behavior you engaged in was . . . ?” The 11 behaviors were (a) a person had oral (mouth) contact with your breasts or nipples; (b) deep kissing (French or tongue kissing); (c) a person touched, fondled, or manually stimulated your genitals; (d) you touched, fondled, or manually stimulated a person’s genitals; (e) you had oral (mouth) contact with a person’s genitals; (f) penile–vaginal intercourse (penis in vagina); (g) a person touched, fondled, or manually stimulated your breasts or nipples; (h) penile–anal intercourse (penis in anus [rectum]); (i) you touched, fondled, or manually stimulated a person’s breasts or nipples; (j) a person had oral (mouth) contact with your genitals; and (k) you had oral (mouth) contact with a person’s breasts or nipples. Percentages (with 95% confidence intervals) were calculated for male and female respondents. Chi-square contingency tests and Pearson’s correlations were conducted for group comparisons as appropriate. RESULTS A hierarchy of behaviors emerged from the responses (see Table I). Almost all respondents (98.7%) agreed that penile–vaginal intercourse constituted having sex, whereas only 6.4% of the respondents regarded deep kissing as having sex. Of the sample, 7% or less considered contact with breasts or nipples as
2.1 (0.0–4.1) 4.7 (1.7–7.8) 4.2 (1.4–7.1) 4.2 (1.4–7.1) 4.7 (1.7–7.8) 15.3 (10.1–20.4) 15.8 (10.6–20.9) 32.1 (25.5–38.7) 32.6 (25.9–39.3) 82.1 (76.7–87.6) 98.9 (97.5–100)
Deep kissing Oral contact with breasts/nipples Person touches your breasts/nipples You touch other’s breasts/nipples Oral contact on other’s breasts/nipples You touch other’s genitals Person touches your genitals Oral contact with other’s genitals Oral contact with your genitals Penile–anal intercourse Penile–vaginal intercourse
12.9 (7.0–18.8) 9.7 (4.5–14.9) 8.9 (3.9–9.5) 10.5 (5.1–15.9) 11.3 (5.7–16.9) 21.8 (14.5–29.0) 22.6 (15.2–30.0) 34.7 (26.3–43.1) 35.5 (27.1–43.9) 70.2 (62.1–78.2) 98.4 (94.9–100)
UK men (N = 124) 6.4 (3.6–9.1) 6.7 (3.9–9.5) 6.1 (3.4–8.7) 6.7 (3.9–9.5) 7.3 (4.4–10.2) 17.8 (13.6–22.1) 18.5 (14.2–22.8) 33.3 (27.9–38.3) 33.8 (28.5–38.9) 77.9 (72.8–82.0) 98.7 (97.0–99.8)
UK overall (N = 314) 1.4 2.0 1.7 2.3 1.4 11.6 12.2 37.3 37.7 82.3 99.7
US women (N = 353)
2.9 4.5 5.7 4.1 6.1 17.1 19.2 43.7 43.9 79.1 99.2
US men (N = 244)
2.0 3.0 3.4 3.0 3.4 13.9 15.1 39.9 40.2 81.0 99.5
US overall (N = 598)
PP016-291182
Behaviors
Percentage indicating “had sex” (95% confidence intervals)
P2: FPX/FNV
UK women (N = 190)
Archives of Sexual Behavior [asb]
Table I. Percentages for Participants Including the Behavioral Items as “Having Sex”
P1: FYJ/GBO QC: FTK February 5, 2001 11:33 Style file version Nov. 19th, 1999
172 Pitts and Rahman
P1: FYJ/GBO
P2: FPX/FNV
Archives of Sexual Behavior [asb]
QC: FTK PP016-291182
Sex Constituting Behaviors in the UK
February 5, 2001
11:33
Style file version Nov. 19th, 1999
173
having sex and around 18% of the sample included manual contact with genitals as having sex. Oral contact with genitals was included by approximately one-third of respondents. Penile–anal intercourse was included by 78% of the respondents. Men and women broadly agreed in their classifications of behaviors; however, there were some disparities also. Chi-square tests showed significant differences between men and women in their responses to items of deep kissing, manual contact with genitalia, and penile–anal intercourse. Significantly, more men included deep kissing ( p < .001) and manual contact with genitals ( p = .037) as having sex, whereas more women included penile–anal intercourse ( p = .044). Out of the 11 behaviors, only penile–anal intercourse was rated as sex more frequently by women ( p = .012). Overall, the mean number of items positively included was 3.11; the median and mode were 2. For men, the mean was 3.3, and for women, it was 2.95; this difference was nonsignificant. There was a significant positive correlation between age and number of items included (r = .257, N = 312, p < .01). Age was collapsed into a dichotomous variable, and chi-square tests were carried out, considering those below 25 years in one group and those 25 and above in another group. Four of the eleven comparisons were significantly different: (a) a person touching your genitals (χ 2 = 14.684, df = 1, p < .001); (b) you touching a person’s genitals (χ 2 = 13.174, df = 1, p < .001); (c) you having oral contact with a person’s genitals (χ 2 = 22.92, df = 1, p < .001); and (d) a person having oral contact with your genitals (χ 2 = 19.18, df = 1, p < .001). In all cases, the younger age groups (<24 years) were significantly less likely to include these items as constituting having sex. A Guttman scalogram analysis was carried out to test whether the data formed a unidimensional scale. Guttman analysis for the whole data set revealed r = .9801, which indicates a strong degree of unidimensionality. Examining the data set for men revealed r = .974; for women the score was r = .988. Both are well above the recommended .9 level, but the women’s score indicates a greater uniformity of pattern. Dividing the data by age, those 20 years and below scored r = .988 and those of 21 and above scored r = .966, indicating that older respondents were less uniform in their responses. However, all scores indicate a strong element of unidimensionality in the data. Table I also shows comparisons with the US data. From inspection, the two data sets appear to be strikingly similar to each other; however, there are a few points to note. Working through the “hierarchy” of behaviors, it can be seen that overall UK respondents are higher in endorsements than US respondents for all behaviors up to oral contact with genitalia. At that point, there is a “cross-over” with more US respondents, including the remaining four behaviors in having sex. This pattern is replicated for both men and women respondents, with a particularly large difference of 8%–9% between US men and UK men on those items concerning oral contact with genitalia.
P1: FYJ/GBO
P2: FPX/FNV
Archives of Sexual Behavior [asb]
QC: FTK PP016-291182
February 5, 2001
174
11:33
Style file version Nov. 19th, 1999
Pitts and Rahman
DISCUSSION These data provide insights into what is regarded as constituting having sex. Among the behaviors studied there was a relatively clear consensus about the role of “penetration”; with penile–vaginal intercourse being almost universally included. One or two comments, however, indicate some degree of disagreement even in this area. Two female students wrote that orgasm, which had not been included in the questionnaire, was important in their definition of having sex. The area of least consensus concerns oral–genital contact. In line with the US study, we found that around one-third of respondents included this in their definition of sex. This may reflect Sanders and Reinisch’s notion of “technical virginity” (Sanders and Reinisch, 1999). That is, those who experience oral–genital contact in the absence of penetrative intercourse (penile–vaginal/penile–anal) do not consider oral–genital contact as having had sex. This attitudinal style may be prevalent in societal contexts where there are sanctions against oral–genital contact (as in some states in the US today) and/or the universal cultural insistence on the primacy of penetration in sex. The largest sex difference was for the item on “deep kissing”; our result of nearly 13% of men including this item is difficult to explain. We suspect that, for some men, this item was interpreted in the context of oral–genital contact, although “French or tongue kissing” was included in the definition offered in the questionnaire. Alternatively, inclusion of this item may reflect men’s greater positive orientation toward sexual behavior, or their greater “erotophilia” (Purnine and Carey, 1998). Indeed, men tend to incorporate more behavioral items in their construals of sex (Purnine et al., 1994). The greater male inclusion of manual contact with genitalia may also reflect male-typical incorporation of more sexrelated items or, more obviously, the greater incidence of manual masturbation in men as compared to women (Oliver and Hyde, 1993). Eighty two percent of women included penile–anal intercourse as having had sex, whereas men seemed somewhat ambivalent, with 30% not including it. The male response may reflect discomfort with the subject together with the misconception that anal intercourse is a predominantly a gay practice (Voeller, 1991). The correlation of number of behaviors included with age suggests that definitions of sex are changing, and that younger respondents include fewer kinds of sexual behavior in their construct. This interpretation is supported by other recent studies; Purnine and Carey (1998) reported that both older men and older women were less conventional than younger ones in their sexual behavioral preferences. Researchers need to be clear in their definitions if younger respondents are to report nonpenetrative sex in their self-reports of sexual activity. The scores for the Guttman scalogram analysis indicate that the questions form a single scale, although a number of the items are very close in their levels of response. There were no strikingly different patterns of responding by age or by gender that moderated the scalability. This study examined only a
P1: FYJ/GBO
P2: FPX/FNV
Archives of Sexual Behavior [asb]
QC: FTK PP016-291182
February 5, 2001
11:33
Sex Constituting Behaviors in the UK
Style file version Nov. 19th, 1999
175
convenience sample of white, university educated, mainly Christian respondents. Further research should seek a wider social economic and ethnic mix of respondents, with different educational levels. In addition, investigations into the construals of sex held by nonheterosexuals could prove interesting. A short report indicated that gay, lesbian, and bisexual respondents regard more activities (particularly manual contact with genitalia) as constituting sexual relations (Richters and Song, 1999). Comparisons with US results may indicate that opinions regarding sex differ between countries; this supports findings from two national surveys, which concluded that there were less liberal opinions in the US, but greater variability in sexual behaviors (Michael et al., 1998). It is clear from both the US study and this study that we cannot, as researchers, assume that all respondents share our construction of sexual relations, and that we need to be explicit and operationalize our terminology in future surveys. ACKNOWLEDGMENTS The authors would like to acknowledge the positive suggestions made by two anonymous reviewers of the manuscript. We would also wish to acknowledge the statistical advice offered by Dr. David Clark Carter, especially concerning the Guttman scalogram analyses. REFERENCES Bancroft, J. (1997). Researching Sexual Behavior, Indiana University Press, Indiana. Blower, S. M., and Boc, C. (1993). Sex acts, sex partners and sex budgets: Implications risk factor analysis and estimation of HIV transmission probabilities. J. Acquir. Immune Defic. Syndr. 6: 1347–1352. Cooper, M. L., Shapiro, C. M., and Powen, A. M. (1998). Motivations for sex and risky sexual behaviour among adolescents and young adults: A functional perspective. J. Pers. Soc. Psychol. 75: 1528– 1558. Gold, R. S., Karmiloff-Smith, A., Skinner, M. J., and Morton, J. (1992). Situational factors and thought processes associated with unprotected intercourse in heterosexual students. AIDS Care 4: 305– 323. McConaghy, N. (1999). Unresolved issues in scientific sexology. Arch. Sex Behav. 28: 285–318. Michael, R. T., Wadsworth, J., Feinleib, J., Johnson, A. M., Laumann, E. O., and Wellings, K. (1998). Private sexual behavior, public opinion, and public health policy related to sexually transmitted diseases: A US–British comparison. Am. J. Public Health 88: 749–754. Oliver, M. B., and Hyde, S. S. (1993). Gender differences in sexuality: A meta-analysis. Psych. Bull. 114(1): 29–51. Purnine, D. M., and Carey, M. P. (1998). Age and gender differences in sexual behavior preferences: A follow up report. J. Sex Marital Ther. 24: 93–102. Purnine, D. M., Carey, M. P., and Jorgensen, R. S. (1994). Gender differences regarding preferences for specific heterosexual practices. J. Sex Marital Ther. 20: 271–285. Ratliff-Cairn, J., Donald, K. M., and Dalton, J. (1999). Knowledge, beliefs, peer norms and past behaviours as correlates of risky sexual behavior among college students. Psychol. Health 14: 625–641.
P1: FYJ/GBO
P2: FPX/FNV
Archives of Sexual Behavior [asb]
176
QC: FTK PP016-291182
February 5, 2001
11:33
Style file version Nov. 19th, 1999
Pitts and Rahman
Richters, J., and Song, A. (1999). Australian university students agree with Clinton’s definition of sex. BMJ 318: 1011. Rosenthal, D. A., Smith, A. M. A., Reichler, H., and Moore, S. (1996). Changes in heterosexual university undergraduates HIV-related knowledge, attitudes and behavior. Genitourin Med. 72: 123–127. Sanders, S. A., and Reinisch, J. M. (1999). Would you say you had sex if . . . . ? JAMA 281: 275–277. Seal, D. W., and Agostinelli, G. (1996). College students’ perceptions of the prevalence of risky sexual behavior. AIDS Care 8: 453–466. Seidman, S. N., and Reider, R. O. (1994). A review of sexual behavior in the United States. Am. J. Psychiatr. 151: 330–341. Sutton, S., McVey, D., and Glanz, A. (1999). A comparative test of the theory of reasoned action and the theory of planned behavior in the prediction of condom use intentions in a national sample of English young people. Health Psychol. 18: 72–81. Tanne, J. H. (1999). JAMA’s editor fired over sex article. BMJ 318: 213. Tiefer, L., and Kring, B. (1995). Gender and the organization of sexual behavior. Psychiatr. Clin. North Am. 18: 25–37. Voeller, B. (1991). AIDS and heterosexual anal intercourse. Arch. Sex. Behav. 20: 233– 276. Wellings, K., Field, J., Johnson, A. M., and Wadsworth, J. (1994). Sexual behavior in Britain: The National Survey of Sexual Attitudes and Lifestyles, Penguin, Harmondsworth.
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
QC: FOM/FTK PP016-291183
February 12, 2001
14:6
Style file version Nov. 19th, 1999
Archives of Sexual Behavior, Vol. 30, No. 2, 2001
Prevalence of Sexual Dysfunctions: Results from a Decade of Research Jeffrey S. Simons, Ph.D.1 and Michael P. Carey, Ph.D.2,3
Ten years of research that has provided data regarding the prevalence of sexual dysfunctions is reviewed. A thorough review of the literature identified 52 studies published in the 10 years since an earlier review by Spector and Carey (Arch. Sex. Behav. 19(4): 389–408, 1990). Community samples indicate a current prevalence of 0%–3% for male orgasmic disorder, 0%–5% for erectile disorder, and 0%–3% for male hypoactive sexual desire disorder. Pooling current and 1-year figures provides community prevalence estimates of 7%–10% for female orgasmic disorder and 4%–5% for premature ejaculation. Stable community estimates of the current prevalence of other sexual dysfunctions remain unavailable. Prevalence estimates obtained from primary care and sexuality clinic samples are characteristically higher. Although a relatively large number of studies has been conducted since the earlier review, the lack of methodological rigor of many studies limits the confidence that can be placed in these findings. KEY WORDS: sexual dysfunction; prevalence; epidemiology; sexuality.
INTRODUCTION Sexual dysfunctions are believed to be among the more prevalent psychological disorders in the general population (Spector and Carey, 1990). The sales data and media attention associated with recent biomedical treatments (e.g., Viagra® ) corroborates the commonness of such dysfunctions. Despite their apparent prevalence, however, sexual disorders have typically not been included in large scale epidemiologic studies such as the Epidemiologic Catchment Area (ECA) Study 1 Department
of Psychology, 430 Huntington Hall, Syracuse University, Syracuse, New York 132442340. 2 Director, Center for Health and Behavior, Syracuse University, Syracuse, New York 13244-2340. 3 To whom correspondence should be addressed at Syracuse University, 430 Huntington Hall, Syracuse, New York 13244-2340; e-mail: [email protected]. 177 C 2001 Plenum Publishing Corporation 0004-0002/01/0400-0177$18.00/0 °
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
178
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
Simons and Carey
(Regier et al., 1990). Lacking such large-scale epidemiologic data, sexual health practitioners and scientists must attempt to integrate smaller studies to obtain population estimates regarding the prevalence of sexual disorders. To this end, the current paper provides an updated review and methodological critique of the published empirical studies that provide epidemiologic data as a resource for professionals who are investigating the etiology, assessment, or treatment of these disorders as well as for those involved in the allocation of resources for prevention, treatment, and research. Research on the sexual dysfunctions has increased dramatically since Spector and Carey’s review (Spector and Carey, 1990). During the last 10 years, there have been 52 empirical studies providing epidemiologic data on sexual dysfunctions (see Table I). This compares to 47 studies published in the previous 50 years (1940–89). In the current paper, we review these 52 studies published since the Spector and Carey’s review (Spector and Carey, 1990). To identify this research, we searched Medline and PsychInfo databases for articles that appeared between 1990 and 1999, using the following key words: (epidemiology OR incidence OR prevalence) AND (premature ejaculation OR impotence OR erectile dysfunction OR erectile disorder OR hypoactive sexual desire disorder OR sexual aversion disorder OR sexual arousal disorder OR orgasmic disorder OR inhibited female orgasm OR inhibited male orgasm OR dyspareunia OR vaginismus OR frigidity OR inhibited sexual desire OR anorgasmia OR ejaculatory dysfunction OR erectile incompetence OR aspermatism OR retarded ejaculation OR ejaculatory inhibition OR absence of ejaculation). Studies were excluded if they examined dysfunctions exclusively among surgical patients (e.g., radical prostatectomy), examined erectile dysfunction among diabetics (see Weinhardt and Carey, 1996), examined sexual dysfunction among patients with cancer (e.g., prostrate or ovarian cancer) or undergoing chemotherapy, or dysfunctions resulting from medication side effects (e.g., selective serotonin reuptake inhibitors). In addition to the database search, the reference sections of the included papers were reviewed for additional studies. We review the studies separately by gender, type of sample (i.e., clinical, community, comorbid disorders), and the phase of sexual response cycle (i.e., desire, arousal, orgasm) investigated. In reviewing studies of each dysfunction, we first review its prevalence in clinical and community samples without comorbid disorders. We then review prevalence rates in subpopulations when available. We have chosen to use the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV; American Psychiatric Association [APA], 1994) nomenclature because it provides a standard set of definitions with which to structure the review. However, only 8 of the 51 studies (16%) used the diagnostic criteria found in the DSM. More than a third of the studies did not provide any operational definition of the dysfunction being investigated. Thus, there exists the potential for great variation in the object of investigation within each of the sections that we have labeled according to the DSM-IV nomenclature. We discuss the potential
Sample NR; Current
Erectile dysfunction;∗ Premature ejaculation;∗ Lack of sexual desire;∗ Retarded ejaculation: Prevalence statistics not reported.
Prevalence
179
19:42
(Continued)
Strengths: Comparative study with a well-matched control group Limitations: Doesn’t provide prevalence statistics.
January 25, 2001
group significantly more symptomatic than control, p < .05. No differences between treated and nontreated groups.
Structured interview/ questionnaire (Schiavi et al., 1990)
Procedure; Assessment
PP016-291183
∗ Schizophrenia
N = 51 Age: Range NR (M: 33) Gender: Male Race/ethnicity: NR Health status: Healthy Recruitment: Military reserve
Comparison sample
P2: FLF/FOM/FJQ
Aizenberg N = 51 et al. (1995) Age: Range NR (M: 36) Gender: Male Race/ethnicity: NR Health status: Schizophrenia (on neuroleptic medication) Recruitment: Outpatient N = 20 Age: Range NR (M: 35) Gender: Male Race/ethnicity: NR Health status: Schizophrenia (no medication) Recruitment: Outpatient
Authors
Methodological strengths and limitations
Archives of Sexual Behavior [asb]
Criteria for dysfunction; Time frame
Table I. Studies of the Prevalence of Sexual Dysfunction (1990–99)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
N = 72 Age: NR Gender: Male Race/ethnicity: Egyptian Health status: NR Recruitment: Textile factory workers
N = 208 Age: NR Gender: Male Race/ethnicity: Egyptian Health status: NR Recruitment: Pesticide formulators DSM-III-R; Current
Painful intercourse: Total: 33/2011 = 2%; Age 55–64: 23/662 = 3%; Age 65–74: 8/697 = 1%; Age 75–84: 2/539 = 0.4%; Age >85: 0/113 = 0%
Erectile dysfunction disorder: Pesticide formulators: 11/208 = 5%; Textile workers: 2/72 = 3% Erectile dysfunction symptoms: Pesticide formulators: 56/208 = 27%;∗ Textile workers: 3/72 = 4%
Prevalence
January 25, 2001
180
Strengths: Representative population sample of older women Limitations: Large percentage (76%) of women were not sexually active; prevalence of sexual dysfunctions among those who are active are not reported
Strengths: Large comparative study; use of DSM-III-R criteria; use of within group analyses to demonstrate positive relationship between erectile dysfunction and length of exposure to pesticide Limitations: Did not adequately report test statistics; did not report comparison of additional groups included in the study
PP016-291183
Representative population Painful intersample of older British course in women; Structured the past interview 2 years
Random selection of pesticide exposed workers and matched controls; Participants were administered the GHQ (Goldberg & Hillier, 1979) and interviewed
Procedure; Assessment
Methodological strengths and limitations
P2: FLF/FOM/FJQ
Barlow et al. N = 2045 (1997) Age: 55–85+ (M: NR) Gender: Female Race/ethnicity: NR Health status: NR Recruitment: Community
Comparison sample
Sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Amr et al. (1997)
Authors
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK 19:42 Style file version Nov. 19th, 1999
N = 20 Age: 21–60+ (M: NR) Gender: 15% female Race/ethnicity: British Health status: NR Recruitment: Sex clinic Retrospective analysis NR; Current of records in sex and marital problems clinic; Referring physician and therapist diagnoses
Impotence: General practitioner dx, GP: Asian: 8/18 = 44%; Engl.: 9/17 = 53% Therapist dx, TH: Asian: 9/18 = 50%; Engl.: 8/17 = 47% Male desire disorder: (GP): Asian: 1/18 = 6%; Engl.: 1/17 = 6% (TH): Asian: 0/18 = 0%; Engl.: 2/17 = 12% Female desire disorder: (GP) + (TH): Asian: 0/3 = 0%; Engl.: 1/3 = 33% Premature ejaculation: (GP): Asian: 2/18 = 11%; Engl.: 1/17 = 6% (TH): Asian: 2/18 = 11%; Engl.: 0/17 = 0% Male orgasmic disorder: (GP): Asian: 1/18 = 6%; Engl.: 1/17 = 6% (TH): Asian: 0/18 = 0%; Engl.: 1/17 = 6% Female orgasmic disorder: (GP): Asian: 1/3 = 33%; Engl.: 0/3 = 33% (TH): Asian: 0/3 = 33%; Engl.: 0/3 = 33%
PP016-291183
(Continued)
Strengths: Data on an underrepresented population Limitations: Diagnostic criteria are not provided; small sample of patients in a sex and marital problems clinic
P2: FLF/FOM/FJQ
greater than control group ( p < .001).
N = 21 Age: 21–60+ (M: NR) Gender: 11% female Race/ethnicity: Asian Health status: NR Recruitment: Sex clinic
Archives of Sexual Behavior [asb]
∗ Significantly
Bhui et al. (1994)
P1: FOM/LZR QC: FOM/FTK January 25, 2001
181
19:42 Style file version Nov. 19th, 1999
N = 2616 Age: 40–60 (M: NR) Gender: Female Race/ethnicity: 46% Chinese, 38% Malay, 16% other Health status: NR Recruitment: Community, primary care clinics, Indonesian Menopause Society
Sample
Comparison sample Uniform questionnaire distributed in seven Asian countries
Procedure; Assessment NR; Current
Criteria for dysfunction; Time frame Dyspareunia Hong Kong (n = 427): Premenopause: 3%; Perimenopause: 10%; Postmenopause: 15%∗ Indonesia (n = 346): Premenopause: 9%; Perimenopause: 48%; Postmenopause: 47%∗ Korea (n = 500): Premenopause: 20%; Perimenopause: 10%; Postmenopause: 11%∗ Malaysia (n = 401): Premenopause: 4%; Perimenopause: 7%; Postmenopause: 10%∗ Philippines (n = 500): Premenopause: 7%; Perimenopause: 27%; Postmenopause: 13%∗ Singapore (n = 420): Premenopause: 5%; Perimenopause: 8%; Postmenopause: 8% Taiwan (n = 398): Premenopause: 11%; Perimenopause: 15%; Postmenopause: 10%
Prevalence
Strengths: Large multinational sample; use of a uniform questionnaire across sites Limitations: Subsample sizes of menopausal status by country not provided; time reference not reported
Methodological strengths and limitations
Archives of Sexual Behavior [asb]
Boulet et al. (1994)
Authors
Table I. (Continued)
P1: FOM/LZR P2: FLF/FOM/FJQ PP016-291183
QC: FOM/FTK January 25, 2001
182
19:42 Style file version Nov. 19th, 1999
N = 20 Age: 21–36 (M: 27) Gender: Female Race/ethnicity: 50% White, 50% Black Health status: HIV positive Recruitment: Referrals from U.S. Air Force (USAF) mandatory screening program
183
(Continued)
19:42
< .05 (χ 2 -test).
Strengths: Use of DSM-III-R criteria Limitations: Small nonrepresentative sample
Strengths: Use of DSM-III-R criteria; presentation of interrater reliability Limitations: Small nonrepresentative sample
January 25, 2001
∗p
Hypoactive sexual desire disorder: 10/32 = 31% Sexual aversion disorder: 0%
Hypoactive sexual desire disorder: 4/20 = 20%
PP016-291183
DSM-III-R; Current
DSM-III-R; Current
P2: FLF/FOM/FJQ
Standard assessment battery and semistructured interview given to referrals from USAF; Modified the Structured Clinical Interview for DSM-III-R to assess sexual dysfunctions
Standard assessment battery and semistructured interview given to referrals from USAF
Archives of Sexual Behavior [asb]
Brown & Rundell N = 43 (1993) Age: 21–36 (M: 29) Gender: Female Race/ethnicity: 54% White, 42% Black, 4% Hispanic Health status: HIV positive Recruitment: Referrals from U.S. Air Force (USAF) mandatory screening program
Brown & Rundel (1990)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
N = 36 Consecutive attendees at Age: Range NR (M: 38) clinic were given Gender: Male questionnaire battery Race/ethnicity: NR and semistructured Health Status: interview HIV negative and hemophilia Recruitment: Hemophilia clinic
N = 25 men Recruited gay men Age: Range NR (M: 33) from an STD clinic; Gender: Male Semistructured Race/ethnicity: NR interview Health status: HIV positive Recruitment: STD clinic
N = 24 men Age: Range NR (M: 36) Gender: Male Race/ethnicity: NR Health status: HIV positive Recruitment: STD clinic
Catalan et al. (1992b)
Procedure; Assessment
N = 37 Age: Range NR (M: 37) Gender: Male Race/ethnicity: NR Health status: HIV positive and hemophilia Recruitment: Hemophilia clinic
Comparison sample
184
Erectile dysfunction: HIV+ 6/16 = 38%; HIV− 5/23 = 22% Premature ejaculation: HIV+ 0/16 = 0%; HIV− 1/23 = 4% Delayed ejaculation:† HIV+ 6/16 = 38%; HIV− 2/23 = 9% Loss of interest: HIV+ 12/16 = 75%; HIV− 14/23 = 61%
Erectile dysfunction: HIV+ 3/20 = 15%; HIV− 2/24 = 8% Premature ejaculation: HIV+ 6/20 = 30%;∗ HIV− 1/24 = 4% Delayed ejaculation: HIV+ 4/20 = 20%;∗ HIV− 0/24 = 0% Loss of interest in sex: HIV+ 11/20 = 55%; HIV− 8/24 = 33%
Strengths: Controlled comparison study of HIV+ men Limitations: Small sample size; criteria for dysfunction not provided
Strengths: Comparative study with a well-matched control group Limitations: Criteria for dysfunction not reported
PP016-291183
NR; Current
NR; Current
Prevalence
Methodological strengths and limitations
P2: FLF/FOM/FJQ
Catalan et al. (1992a)
Sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Authors
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK January 25, 2001 19:42 Style file version Nov. 19th, 1999
Self-reported difficulty getting or maintaining an erection; Self-reported pain with intercourse; Current
Erectile dysfunction: Total: 114/283 = 40%; Married men: 92/240 = 38%;¶ Other marital categories: 22/43 = 51% Pain with intercourse: Married women: 21/164 = 13%; Married men: 3/214 = 1%
Decreased libido: 194/421 = 46%‡ Dyspareunia: 105/421 = 25%
185
(Continued)
19:42
† Significant
Strengths: Large stratified population sample of elderly Limitations: DSM criteria not used; correlates of erectile dysfunction based upon relatively small subgroups (i.e., 15–47 participants)
Strengths: Examined correlates Limitations: Self-selected sample; criteria for dysfunction not reported
January 25, 2001
= significantly greater, p < .05. difference ( p < .05). ‡ Correlated with age and dyspareunia. ¶ Prevalence positively associated with incontinence, interrupted urinary stream, heart attack, sedative use, caffeine nonuse; p’s < .05.
Interview
NR; Current
Strengths: Use of DSM-III diagnostic criteria; clinical sample Limitations: Generalizability restricted to PMS treatment seekers
PP016-291183
∗ HIV
N = 744 Age: 60–80 + (M: NR) Gender: 60% female Race/ethnicity: NR Health status: NR Recruitment: Probability sample of elderly population in Washtenaw County, Michigan
Diokno et al. (1990)
Volunteers at a menopausal center; Interview
Inhibited sexual desire: 9/43 = 21% Inhibited sexual excitement: 9/43 = 21% Functional dyspareunia: 5/43 = 12% Inhibited orgasm: 2/43 = 5%
P2: FLF/FOM/FJQ
N = 421 Age: Range NR (M: 50) Gender: Female Race/ethnicity: NR Health status: Healthy Recruitment: Volunteers at a menopausal center
Consecutive new patients DSM-III; at a PMS clinic; NIMH Lifetime Diagnostic Interview Schedule (Robins, 1981)
Archives of Sexual Behavior [asb]
Chiechi et al. (1997)
Chandraiah et al. N = 43 (1991) Age: 15–46 (M: NR) Gender: Female Race/ethnicity: 93% White, 7% Black Health status: NR Recruitment: PMS clinic
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
El-Rufaie et al. N = 36 (1997) Age: Range NR (M: 47) Gender: Male Race/ethnicity: Arab Health status: Healthy Recruitment: Primary health care center in the United Arab Emigrates
N = 39 Age: Range NR (M: 49) Gender: Male Race/ethnicity: Arab Health status: Hypertensives Recruitment: Primary health care center in the United Arab Emigrates
Comparison sample Semistructured interview NR; Current given to consecutive patients
Procedure; Assessment
Erectile weakness: Hypertensives (H): 12/39 = 31%; Healthy (C): 4/36 = 11% Impaired morning erection: (H): 17/39 = 44%; (C): 6/36 = 17% Complete erectile failure: (H): 2/39 = 5%; (C): 2/36 = 6% Impaired spontaneous erection: (H): 10/39 = 26%; (C): 4/36 = 11% Ejaculatory disturbances: (H): 7/39 = 18%; (C): 4/36 = 11% Reduced sexual interest: (H): 8/39 = 21%; (C): 4/36 = 11%
Prevalence
Strengths: Data on an underrepresented population Limitations: No diagnostic criteria used; not a representative sample Comments: This study was designed to examine relations between diabetes and erectile functioning. Thus, comparisons between the healthy and hypertensive group were not made
Methodological strengths and limitations
P2: FLF/FOM/FJQ
Sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Authors
Table I. (Continued)
P1: FOM/LZR PP016-291183
QC: FOM/FTK January 25, 2001
186
19:42 Style file version Nov. 19th, 1999
N = 197 Age: 29 Gender: Male Race/ethnicity: NR Health status: Healthy Recruitment: Community NR; 1-year
WOMEN: Sexual difficulties/ dissatisfaction: 46/218 = 21% Impaired interest: 35/218 = 16% Orgasmic difficulties: 15/218 = 7% Pain/dyspareunia: 13/218 = 6% MEN: Sexual difficulties/ dissatisfaction: 41/197 = 21% Impaired interest: 14/197 = 7%∗ Erectile dysfunction: 0/197 = 0% Premature ejaculation: 8/197 = 4% 10-year prevalence of sexual dysfunction/disturbance from age 20/21 to 29/30: Male: 74/164 = 45%; Female: 106/192 = 55% Chronicity (defined as reporting sexual dysfunction on at least 2 of 4 assessments): Male: 26/164 = 16%; Female: 59/192 = 31%†
PP016-291183 January 25, 2001
187
19:42
(Continued )
Strengths: Stratified sampling of a birth cohort; longitudinal design provides some information of course of sexual dysfunction; Examined numerous correlates of sexual dysfunction. Limitations: Diagnostic criteria not given; detailed assessment of sexual dysfunction not included until Time 4
P2: FLF/FOM/FJQ
disorder).
N = 218 Stratified sample based Age: 30 on SCL-90-R Gender: Female (Derogatis, 1977). Race/ethnicity: NR 2/3 were sampled Health status: Healthy from the 85th Recruitment: Community percentile and above. Longitudinal study of a birth cohort (1979, 81, 86, 88); Semistructured interview assessing dysfunctions in detail was only conducted at Time 4
Archives of Sexual Behavior [asb]
∗ Positively associated with depression, p < .05. † p < .001 (women report more chronic course of
Ernst et al. (1993)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Fass et al. (1998)
N = 508 Age: 20–88 (M: 48) Gender: 72% female Race/ethnicity: 78% White, 6% Black, 5% Hispanic, 3% Asian Health status: Irritable bowel syndrome (IBS)/nonulcer dyspepsia (NUD) Recruitment: Medical clinic
Procedure; Assessment
N = 247 Consecutive referrals to Any interference Age: 18–99 clinic were evaluated with sexual (M: 43) by questionnaire; functioning in Gender: 70% female Community sampling past 6 months Race/ethnicity: procedure not 66% White, specified 8% Black, 3% Hispanic, 12% Asian Health status: Healthy Recruitment: Community
Comparison sample
Decreased sexual drive: Control: 14/193 = 7%; IBS: 29/266 = 11%; NUD: 11/85 = 13%; IBS + NUD: 27/157 = 17%; IBS (non-patients): 3/41 = 7% Dyspareunia: Control: 5/193 = 3%; IBS: 10/266 = 4%; NUD: 1/85 = 1%; IBS + NUD: 9/157 = 6%; IBS (nonpatients): 3/41 = 7% Dyspareunia significantly more common in females (16% vs. 4%, p < .005). No significant differences across disease subtype. Prevalence of sexual dysfunction higher in patient group ( p < .001)
Prevalence
Strengths: Large clinic sample and control group Limitations: Poorly specified dysfunction criteria
Methodological strengths and limitations
P2: FLF/FOM/FJQ
Sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Authors
Table I. (Continued)
P1: FOM/LZR PP016-291183
QC: FOM/FTK January 25, 2001
188
19:42 Style file version Nov. 19th, 1999
Feldman et al. N = 1290 (1994) Age: 40–70 (M: 54) Gender: Male Race/ethnicity: 96% White, 2% Black, 2% other Health status: NR Recruitment: Random sample of noninstitutionalized men Questionnaire
Categories derived by statistical calibration of measure with self-reported dysfunction; Current
Archives of Sexual Behavior [asb]
P2: FLF/FOM/FJQ PP016-291183 January 25, 2001
189
19:42
(Continued )
Erectile dysfunction: Strengths: Overall: 671/1290 = 52% Large community Degree: sample. Examined Minimal: 219/1290 = 17%; association Moderate: 323/1290 = 25%; of multiple aspects Complete: 129/1290 = 10% of health with erectile dysfunction (Complete erectile dysfunction Limitations: tripled from 5% to 15% Categories derived between ages 40 and 70.) by statistical Disease states associated with calibration of probability of complete measure with a erectile dysfunction, age self-report rating adjusted ( p < .01): Comments: Treated heart disease Study found additional n = 90, 39%; significant Treated diabetes associations n = 52, 28%; between erectile Treated hypertension dysfunction and n = 200, 15%; other health indices Untreated ulcer n = 98, 18%; Untreated arthritis n = 228, 15%; Untreated allergy n = 261, 12%
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Sample
Comparison sample Impaired sexual desire defined as never/rarely experiencing sexual desire; endorsement of dysfunction occurring and perceiving it as a problem quite often–all the time; percentages (aside from sexual desire/ interest) based upon sexually active participants; 12 months
Impaired sexual desire:∗† (M): 44/1475 = 3%; (F): 187/1335 = 14% Decreased sexual interest:∗† (M): 88/1463 = 6%; (F): 188/1285 = 15% Erectile/lubricative disability:∗† (M): 39/1288 = 3%; (F): 89/1108 = 8% Orgasm disability: (F): 111/1110 = 10% Premature ejaculation: (M): 51/1281 = 4% Retarded ejaculation: (M): 13/1265 = 1% Vaginism: (F): 11/1102 = 1% Dyspareunia:∗† (M): 13/1331 = 1%; (F): 47/1167 = 4%
Prevalence
Strengths: Large representative population sample; provides operational definitions of dysfunction criteria; report percentage of individuals who perceive the reported dysfunction as a problem Limitations: Assessed group differences based upon symptom endorsement rather than symptom endorsement∗ perceived problem Comments: Study also examines relationships between sexual dysfunction, partnership status, health, and perceived problems
Methodological strengths and limitations
P2: FLF/FOM/FJQ
Random sample drawn from the Swedish Central Population Register; Structured interview
Procedure; Assessment
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Fugl-Meyer & N = 2810 Sjogren Age: 18–74 (M: NR) Fugl-Meyer Gender: 52% male (1999) Race/ethnicity: Swedish Health status: ≥77% healthy Recruitment: Community
Authors
Table I. (Continued)
P1: FOM/LZR PP016-291183
QC: FOM/FTK January 25, 2001
190
19:42 Style file version Nov. 19th, 1999
N = 54 Age: Range NR (M: 36) Gender: Female Race/ethnicity: 22% White, 46% Black, 27% Latina, 3% other Health status: HIV+ Recruitment: Community, HIV/AIDS health clinics
Goggin et al. (1998)
Strengths: Good size clinical sample; use of structured diagnostic interview; examined multiple demographic, psychological, and biological covariates Limitations: Convenience sample may have overrepresented low income consumers of community health services
Strengths: Clear definition of disorder; detailed examination of subtypes and course Limitations: Poor description of demographics
PP016-291183
Criteria HSDD: 21/54 = 39%‡ established by (Brown & Rundell, 1993); Current
Primary dyspareunia Spontaneous resolution: 35/313 = 11%; Resolution with therapy: 14/313 = 5% Secondary dyspareunia Spontaneous resolution: 12/313 = 4%; Resolution with therapy: 25/313 = 8% Persistent primary dyspareunia: 51/313 = 16% Persistent secondary dyspareunia: 54/313 = 17%
P2: FLF/FOM/FJQ
Structured Clinical Interview for DSM-III-R including a sexual disorders module (Brown & Rundell, 1993)
Anonymous Pain or questionnaire mailed discomfort in to 500 women the labial, involved in a study vaginal, or 15 years previously pelvic area while attending a during or after university intercourse; Current
Archives of Sexual Behavior [asb] January 25, 2001
191
19:42
Note: Group differences based on % endorsing dysfunction not associated problems. ∗ Significantly more prevalent in women. † Prevalence increased with age in women and men (excluding male dyspareunia). ‡ Positive relation to depressive symptoms, negative relation to life satisfaction, positive relation with perceived risk factor for HIV infection (i.e., sexual contact). (Continued)
N = 313 Age: Early 30s Gender: Female Race/ethnicity: NR Health status: NR Recruitment: Community
Glatt et al. (1990)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Comparison sample
NR; Current
Based upon GRISS norms; Current
Dyspareunia: Previous Pelvic Inflammatory Disease (PID): 14% No PID: 3%∗
Erectile dysfunction: 20/106 = 19% Premature ejaculation: 213/106 = 22% Vaginismus: 26/105 = 25% Anorgasmia: (F): 23/105 = 22%
Prevalence
Strengths: Large clinical sample Limitations: Demographics not provided; subsample sizes not clear; criteria for dysfunction not specified
Strengths: Use of a standardized questionnaire; large clinical sample Limitations: Relationship between GRISS scores and DSM disorders is unclear
PP016-291183
Questionnaire given to women referred for delivery or induced first-trimester abortion
Consecutive new patients were given Golombuk-Rust Inventory of Sexual Satisfaction (GRISS; Rust & Golombok, 1986)
Procedure; Assessment
Methodological strengths and limitations
P2: FLF/FOM/FJQ January 25, 2001
192
19:42
Heisterberg (1993) N = 1221 Age: NR Gender: Female Race/ethnicity: NR Health status: NR Recruitment: Gynecology clinic
N = 211 Age: (M: NR) Range: 16–78 (men) 16–49 (women) Gender: 50% female Race/ethnicity: Men: 75% White, 8% Black African, 6% Black Caribbean Women: 74% White, 11% Black African, 5% Black Caribbean Health status: Various STD Recruitment: STD clinic
Sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Goldmeier et al., (1997)
Authors
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
N = 400 Age: 40–60 (M: NR) Gender: Female Race/ethnicity: 13% Chinese, 70% Malays, 16% Indians Health status: NR Recruitment: Community and hospital
p < .01.
Dyspareunia: 248/543 = 46%† Pain for >1 year duration: 111/549 = 20% Decreased frequency of orgasm: 45/549 = 8% Decreased pleasure with orgasm: 26/549 = 5% Not orgasmic secondary to pain: 29/549 = 5% Less pleasure associated with sex: 11/549 = 2% Less interest in sex: 55/549 = 10%
Dyspareunia: 32/400 = 8% Ceased sexual activity for various reasons: 92/400 = 23%
January 25, 2001
193
19:42
(Continued)
Strengths: Comprehensive assessment Limitations: Criteria for dysfunction not specified
Strengths: Large sample of underrepresented groups Limitations: Sampling strategy and criteria for dysfunction not well defined
PP016-291183
∗ Odds ratio 3.87, 95% CI 2.35–6.37, no age relation. † Positively associated with lower income and Black ethnicity,
NR; Current
NR; Current
P2: FLF/FOM/FJQ
Consecutive patients were administered a questionnaire
Structured interview
Archives of Sexual Behavior [asb]
Jamieson & Steege N = 533 (1996) Age: 18–45 (M: 32) Gender: Female Race/ethnicity: 74% White, 26% Black Health status: NR Recruitment: Primary care/gynecology
Ismael (1994)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Strengths: Clear description of survey questions; relatively large sample covering diverse geographical areas Limitations: Not a representative sample; number of respondents is reported inconsistently (i.e., 1517 responded to questionnaire yet 1680 is used in calculating some percentages)
Strengths: Large clinical sample Limitations: Relied on female partner to report male sexual dysfunctions; criteria for dysfunction not given
PP016-291183
No sexual erections in past 12 months: 129/1680 = 8% Erections when stimulated in past few months among those who reported erections in past 12 months: Not at all: 74/1388 = 5%; Less than 1 time in 5: 98/1388 = 7%; Less than half the time: 101/1388 = 7%; About half the time: 166/1388 = 12%; More than half the time: 199/1388 = 14%; Almost always: 717/1388 = 52%
Dyspareunia: 25/200 = 13% Premature ejaculation: 16/200 = 8% Erectile dysfunction: 2/200 = 1% Anorgasmia: 81/200 = 41%
Prevalence
Methodological strengths and limitations
P2: FLF/FOM/FJQ
Questionnaire response 1 year; 3 months
Participants were recruited at free prostrate cancer screenings; Questionnaire
N = 1517 Age: Range NR (M: 60) Gender: Male Race/ethnicity: 70% White, 25% Black, 3% Hispanic, 1% Arabic Health status: NR Recruitment: Community
Jonler et al. (1995)
Procedure; Assessment Consecutive new NR; Current patients were assessed with a structured interview. When the women reported a dysfunction in her male partner, the partner was then interviewed to confirm this
Comparison sample
N = 200 Age: NR Gender: Female Race/ethnicity: Indian Health status: NR Recruitment: Infertility clinic
Sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Jindal & Dhall (1990)
Authors
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK January 25, 2001
194
19:42 Style file version Nov. 19th, 1999
N = 3159 Age: 18–59 (M: NR) Gender: 55% female Race/ethnicity: NR Health status: NR Recruitment: Community
195
Inhibited sexual desire: (M): 17/428 = 4%; (F): 67/417 = 16% Inhibited sexual excitement: (M): 4/428 = 0.9%; (F): 25/417 = 6% Inhibited orgasm: (M): 3/428 = 0.7%; (F): 15/417 = 4% Functional dyspareunia: (M): 1/428 = 0.2%; (F): 13/417 = 3% (All dysfunctions more prevalent in women, p < .01.)
Lack interest in sex: (M): 213/1346 = 16%; (F): 535/1622 = 33% Unable to achieve orgasm: (M): 108/1346 = 8%; (F): 389/1622 = 24% Pain during sex: (M): 40/1346 = 3%; (F): 227/1622 = 14% Sex not pleasurable: (M): 108/1346 = 8%; (F): 341/1622 = 21% Climax too early: (M): 390/1346 = 29%; (F): 162/1622 = 10% Erectile dysfunction: (M): 135/1346 = 10% Difficulty lubricating: (F): 305/1346 = 19%
January 25, 2001 19:42
(Continued )
Strengths: Large representative sample; structured interview using DSM-III criteria Limitations: Sexual dysfunction criteria have evolved significantly since the DSM-III
Strengths: Large representative sample. Limitations: Dysfunctions were assessed by a dichotomous response to a problem area
PP016-291183
DSM-III; Lifetime
Endorsement of problem occurring in the last year
P2: FLF/FOM/FJQ
Random selection of one half of the 1931 Icelandic birth cohort; Participants interviewed using the DIS-IIIA (fully structured interview)
National probability sample living in household throughout the United States
Archives of Sexual Behavior [asb]
Lindal & Stefansson N = 845 (1993) Age: 55–57 (M: NR) Gender: 49% female Race/ethnicity: Nordic Health status: NR Recruitment: Random selection of population cohort
Laumann et al. (1999)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
N = 4462 Age: 31–49 (M: 38) Gender: Male Race/ethnicity: 82% White, 12% Black, 4% Hispanic, 2% other Health status: NR Recruitment: Random selection of Vietnam-era U.S. Army veterans
N = 62 Age: 18–73 (M: 38) Gender: Male Race/ethnicity: NR Health status: NR Recruitment: 33% undergraduate class; 35% sex dysfunction clinic; 33% general medical practice
Metz & Seifert (1990)
Sample
Mannino et al. (1994)
Authors
Comparison sample
January 25, 2001
196
Strengths: The assessment instrument and problem list were clearly described Limitations: Not a representative sample; the question asked respondents to indicate “sexual concerns which you have had or experienced yourself.” This makes the criteria for dysfunctions quite vague
PP016-291183
Inhibited sexual desire: 24/61 = 39% Difficulty getting an erection: 16/61 = 26% Difficulty maintaining an erection: 18/61 = 29% Premature ejaculation: 40/61 = 65% Inhibited orgasm: 6/61 = 10% Dyspareunia: 5/61 = 8%
Strengths: Large sample; defined criteria for dysfunction; controlled for multiple confounds Limitations: Examination of current smokers found no relation between erectile dysfunction and number of cigarettes per day or number of years smoked; lack of a dose-response relation casts some doubt on pharmacological explanation
P2: FLF/FOM/FJQ
Endorsement of lifetime problem or concern
Persistent Erectile dysfunction difficulty Smokers: getting an 74/2008 = 4%; erection in Non smokers: last year 26/1162 = 2%; Former smokers: 25/1292 = 2% (Adjusted odds ratio for association with current smoking was 1.5, p < .05. Adjusting for other erectile dysfunction risk factors.)
Prevalence
Methodological strengths and limitations
Archives of Sexual Behavior [asb]
Questionnaire
Questionnaire
Criteria for dysfunction; Procedure; Assessment Time frame
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK 19:42 Style file version Nov. 19th, 1999
197
19:42
(Continued )
Strengths: Clear definition of erectile dysfunction; use of a clinical sample demonstrated high prevalence of disorder in a medical outpatient population; conducted informative group comparisons Limitations: Demographic characteristics of sample not well documented
January 25, 2001
.01 (MW-U test).
Erectile dysfunction: Full sample: 77/227 = 34%; Diabetes mellitus: 22/38 = 58%;† Nondiabetic: 55/189 = 29%† Hypertensive: 35/80 = 44% Medicated 33/54 = 61%;† Unmedicated 2/26 = 8%†
PP016-291183
∗ HIV+ group significantly higher, p < † Significant differences ( p < .0001).
Consecutive male Insufficient outpatients were erection to interviewed, given a prevent physical exam, and coitus in the chart was at least reviewed 25% of attempts; Current
Insufficient vaginal response: Strengths: HIV+: 12/30 = 40%; Use of a well-matched HIV−: 6/30 = 20%∗ control group; well Vaginismus: defined criteria for HIV+: 7/29 = 24%; dysfunctions HIV−: 7/33 = 21% Limitations: The study examined additional dysfunctions but did not report specific percentages
P2: FLF/FOM/FJQ
N = 227 Age: 21–84 (M: NR) Gender: Male Race/ethnicity: NR Health status: n = 80 hypertension n = 38 diabetes mellitus n = 109 other Recruitment: Medical outpatient clinic
N = 37 Participants were Dysfunction Age: Range NR (M: 37) assessed with a on 50% or Gender: Female semistructured more of Race/ethnicity: interview. All sexual 19% White, interviews were occasions; 57% Black, checked for Current 24% Hispanic completeness and Health status: HIV− audiotape review injection drug users was used to Recruitment: Media monitor interviewer advertisement/clinical performance
Archives of Sexual Behavior [asb]
Modebe (1990)
Meyer-Bahlburg N = 38 et al. (1993) Age: Range NR (M: 38) Gender: Female Race/ethnicity: 16% White, 74% Black, 10% Hispanic Health status: HIV+ injection drug users Recruitment: Media advertisement/clinical
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Strengths: Study demonstrates significant differences among participants who do or do not volunteer for intrusive sex studies; use of a standard assessment instrument; relatively large clinical sample Limitations: Study assessed multiple dysfunctions but only reported data for the one statistical difference across groups
PP016-291183
Premature ejaculation: Volunteers: 18/74 = 24%;∗ Nonvolunteers: 2/108 = 2%
Dyspareunia: Strengths: (UC): 15/40 = 38%; Use of a matched control (C): 7/38 = 18% group and validated questionnaire Limitations: Nonstandard definition
Prevalence
Methodological strengths and limitations
P2: FLF/FOM/FJQ
Volunteers for an intrusive study of sexual functioning N = 74 Age: Range NR (M: 44) Gender: Male Race/ethnicity: 93% White, 5% Black Health status: NR Recruitment: Inpatient alcoholism rehabilitation program
Derogatis Sexual Based on Nonvolunteers N = 108 Functioning Sexual Inventory (Derogatis Functioning Age: Range NR (M: 44) Gender: Male and Meyer, 1979) Inventory; Current Race/ethnicity: 93% White, 5% Black Health status: NR Recruitment: Inpatient alcoholism rehabilitation program
Procedure; Assessment
Nirenberg et al. (1991)
Comparison sample N = 47 Random selection of Pain severe Age: Range NR (M: 39) patients with irritable enough to Gender: Female bowel syndrome interfere Race/ethnicity: NR from a community with sexual Health status: Healthy data base. Control intercourse; Recruitment: group was obtained Current Community/general from general practices practitioners as well as community (“buddy controls”). Sexual function assessed by a questionnaire validated in a previous study
Sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Moody and Mayberry N = 50 (1993) Age: Range NR (M: 38) Gender: Female Race/ethnicity: NR Health status: Ulcerative colitis (UC) Recruitment: Community
Authors
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK January 25, 2001
198
19:42 Style file version Nov. 19th, 1999
Erectile dysfunction: Erections little or none of the time when sexually stimulated in the last month (6-point scale) Hypoactive sex drive: No instance of sexual drive in last month
DSM-III-R; Current lifetime
Erectile dysfunction: Age 40–49: <1%; Age 70 and over: >25% Hypoactive sex drive: Age 40–49: 0.6%; Age 70 and over: 26%‡
HSDD: HIV+: Current: 12/95 = 13%;† Past: 1/95 = 1% ARC: Lifetime: 0/33 = 0%
January 25, 2001
199
19:42
(Continued)
Strengths: Large population survey; clear definition of sexual dysfunctions Limitations: Did not report size of the age defined subsamples; did not report prevalence statistics adequately; reported a positive correlation between lack of sex drive and age but gave a negative coefficient
Strengths: Good size clinical sample; use of a control group and DSM criteria Limitations: Group demographics inadequately reported; unable to ascertain the appropriateness of the control group
PP016-291183
‡ Significant correlation with age (r = −.53, p < .001).
Random population sample stratified by age and geographic location within Minnesota; Questionnaire
Random selection of military referrals; Interview
P2: FLF/FOM/FJQ
.05.
N = 2115 Age: 40–79 (M: NR) Gender: Male Race/ethnicity: NR Health status: NR Recruitment: Community
Panser et al. (1995)
N = 33 Age: NR Gender: NR Race/ethnicity: NR Health status: NR Recruitment: Alcohol rehabilitation center
Archives of Sexual Behavior [asb]
∗ Higher prevalence among volunteers, p < † Significantly greater than ARC, p < .05.
N = 95 Age: 20–55 (M: NR) Gender: 21% female Race/ethnicity: NR Health status: HIV+ Recruitment: U.S. Air Force medical center
Pace et al. (1990)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Women selected in a stratified sample from a Netherlands city registrar. 2/3 selected between age 45–64 (i.e., around and slightly beyond age of menopause); Questionnaire
NR; Current
200
19:42
Strengths: Large representative sample Limitations: Criteria for dysfunction not reported
Strengths: Examined correlates of dysfunctions Limitations: Criteria not specified; small sample from a single practice; correlates are based upon small subsamples
January 25, 2001
Dyspareunia: Premenopausal: 18/308 = 6%; Postmenopausal: 70/435 = 16%†
Male Erectile dysfunction: 12/72 =17%;∗ Premature ejaculation: 22/72 = 31% Female Anorgasmia: 41/98 = 42%; Vaginismus: 29/98 = 30%
Strengths: Statistics on an underrepresented population Limitations: Sampling strategy not well described
PP016-291183
NR; Current
Dyspareunia (12-month prevalence): Full sample: 60/500 = 12%; Premenopausal: 23/251 = 9%; Perimenopausal: 4/16 = 25%; Postmenopausal: 22/146 = 15%
P2: FLF/FOM/FJQ
N = 1299 Age: 35–80 (M: NR) Gender: Female Race/ethnicity: Dutch Health status: NR Recruitment: Community
Questionnaire
Read et al. (1997) N = 170 Age: 18–65+ (M: 40) Gender: 58% female Race/ethnicity: NR Health status: NR Recruitment: Consecutive attendees at a general practitioner
NR; 1 year
Prevalence
Methodological strengths and limitations
Archives of Sexual Behavior [asb]
Rekers et al. (1992)
Questionnaire
Sample
Ramoso-Jalbuena N = 500 (1994) Age: Range NR (M: 47) Gender: Female Race/ethnicity: 98% Malay Health status: NR Recruitment: Community sample
Authors
Criteria for dysfunction; Comparison sample Procedure; Assessment Time frame
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Endorsement Lack of interest in or desire of lifetime/ for sex: current Lifetime (L): 97/197 = 49%; problem Current (C): 32/197 = 16% Difficulty getting an erection: (L): 79/197 = 40%; (C): 26/197 = 13% Difficulty maintaining an erection: (L): 91/197 = 46%; (C): 30/197 = 15% Ejaculating too soon/too quickly: (L): 87/197 = 44%; (C): 37/197 = 19%‡ Difficulty ejaculating: (L): 77/197 = 39%; (C): 32/197 = 16% Painful receptive anal sex: (L): 120/197 = 61%; (C): 32/197 = 16%¶ Painful insertive anal sex: (L): 28/197 = 14%;¶ (C): 6/197 = 3%
(Continued)
Strengths: Assessed both lifetime and current prevalence in an underrepresented group; although DSM criteria were not used the assessment instrument and problem list were clearly described Limitations: Not a representative sample; confounds of correlates of dysfunctions were not explored
PP016-291183 January 25, 2001
201
19:42
with known sexual side effects, and annual visits to GP, p’s < .05.
Convenience sample from a health seminar for gay men; Questionnaire
P2: FLF/FOM/FJQ
‡ Prevalence varied across religious groups. ¶ Lower prevalence among college graduates.
N = 197 Age: 20–70 (M: 37) Gender: Male Race/ethnicity: 94% White Health status: NR Recruitment: Health seminar
Archives of Sexual Behavior [asb]
∗ Positively associated with age, low social class, medication † Higher prevalence among postmenopausal group.
Rosser et al. (1997)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Sample
N = 20 Age: 27–54 (M: 46) Gender: Male Race/ethnicity: 95% White Health status: Healthy non-alcohol dependent Recruitment: Community
Hypoactive sexual desire: Alcohol dependent group (A): 3/20 = 15%; Control (C): 0/20 = 0% Male erectile disorder: (A): 2/20 = 10%; (C): 0/20 = 0% Reported behavioral problems: Decreased sexual desire: (A): 5/20 = 25%; (C): 3/20 = 15% Loss of erections during sex: (A): 3/20 = 15%; (C): 3/20 = 15%
Reported coital Erectile dysfunction: failure on 17/70 = 24% 50% or more of attempts at intercourse; Current
Prevalence
202
Strengths: Clear inclusion/exclusion criteria to rule out confounding effects of nonalcohol related illnesses and drugs on sexual functioning; use of DSM criteria Limitations: Small self-selected sample; behavior problems not well defined; study lacked sufficient power to detect differences given the low base rates
Strengths: Clear criteria for dysfunction; examined relations with numerous biological indices of sleep disorder Limitations: Relatively small self-selected sample Comments: Study found little evidence of relations between sleep disorders and erectile dysfunction
PP016-291183
Semi-structured DSM-III-R; interview of subject Current and female sexual partner
Participants recruited through media announcement of a health and marital satisfaction study; interview
Procedure; Assessment
P2: FLF/FOM/FJQ
Schiavi et al. N = 20 (1995) Age: 28–59 (M: 40) Gender: Male Race/ethnicity: 75% White Health status: 10+ years of problem drinking, Lifetime DSM-III-R alcohol dependence diagnosis, abstinent for 2–36 months prior to study Recruitment: Community
Comparison sample
Methodological strengths and limitations
Archives of Sexual Behavior [asb]
Schiavi et al. N = 70 (1991) Age: 45–74 (M: NR) Gender: Male Race/ethnicity: 95% White Health status: Healthy Recruitment: Community
Authors
Criteria for dysfunction; Time frame
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK January 25, 2001 19:42 Style file version Nov. 19th, 1999
203 NR; Current
Erectile dysfunction: Patient (P): 29/48 = 60%;∗ Control (C): 12/32 = 37% Ejaculatory disturbance: (P): 4/48 = 8%; (C): 1/32 = 3% Decreased sex drive: (P): 2/48 = 4%; (C): 1/32 = 3%
(Continued)
Strengths: Clinical sample of men with Parkinson’s disease Limitations: Diagnostic criteria not provided; recruitment strategy not provided; few demographic characteristics provided
Strengths: Anonymous questionnaire, compared dysfunctions across populations Limitations: Poorly defined criteria for dysfunctions; did not report statistics comparing specific dysfunctions across groups
19:42
prevalence in patient group, p < .05.
N = 32 Interview Age: Range NR (M: 70) Gender: Male Race/ethnicity: Health status: Healthy Recruitment: Community
Erectile difficulties: Rural: 7/52 = 13%; Urban: 2/41 = 5% Premature ejaculation: Rural: 13/52 = 25%; Urban: 10/41 = 24% Dyspareunia: Rural: 4/33 = 12%; Urban: 3/36 = 8% Orgasmic dysfunction: Rural: 3/33 = 9%; Urban: 13/36 = 36%
January 25, 2001
∗ Higher
N = 48 Age: Range NR (M: 66) Gender: Male Race/ethnicity: NR Health status: Parkinson’s disease Recruitment: Hospital?
Self-report; Current
PP016-291183
Singer et al. (1992)
Anonymous questionnaire given to all patients over 18 at two family practice clinics during a 2-week period
P2: FLF/FOM/FJQ
N = 77 Age: 18–60+ (Men, M: 46; Women, M: 39) Gender: 47% female Race/ethnicity: Eastern European Health status: Recruitment: Urban family practice
Archives of Sexual Behavior [asb]
Shahar et al. N = 85 (1991) Age: 18–60+ (Men, M: 42; Women, M: 33) Gender: 39% female Race/ethnicity: Jewish–Yemenite Health status: NR Recruitment: Rural family practice
Premature ejaculation: (A): 4/20 = 20%; (C): 1/20 = 5% Difficulty ejaculating: (A): 0/20 = 0 %; (C): 0/20 = 0% (No significant differences)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
Sample
204
Reduced sexual desire: Male (M): 20/626 = 3%; Female (F): 34/686 = 5% Pain or discomfort during intercourse: (M): 3/626 = 0.4%; (F): 21/686 = 3% Unable to achieve orgasm: (M): 5/626 = 0.8%; (F): 47/686 = 7%∗ Erectile dysfunction: (M): 34/626 = 5%†
Strengths: Large representative sample Limitations: Criteria not well defined
Strengths: Representative population sample; study provides some information about the influence of assessment strategies Limitations: Criteria and time frame for premature ejaculation, delayed ejaculation, painful ejaculation and decreased sexual desire are not specified
January 25, 2001
Representative Endorsement of population sample brief problem of 1904–74 birth description; cohorts in Denmark; Current Questionnaire
QUESTIONNAIRE Erectile dysfunction more than occasionally: 16/411 = 4% INTERVIEW Erectile dysfunction more than occasionally: 7/100 = 7% (None of these individuals reported this on the questionnaire) Premature ejaculation: 14/100 = 14% Delayed ejaculation: 2/100 = 2% Painful ejaculation, decreased sexual desire, other: 10/100 = 10%
PP016-291183
N = 1494 Age: 18–88 (M: NR) Gender: 50% female Race/ethnicity: Danish Health status: Recruitment: Community
Report of erectile dysfunction obstructing intercourse more than occasionally; criteria not provided for other dysfunctions; 1 year
Prevalence
P2: FLF/FOM/FJQ
Ventegodt (1998)
All participants completed a questionnaire; 103 were randomly selected to be interviewed (100 accepted)
Comparison sample Procedure; Assessment
Methodological strengths and limitations
Archives of Sexual Behavior [asb]
Solstad & Hertoft N = 439 (1993) Age: 51 Gender: Male Race/ethnicity: Danish Health status: NR Recruitment: Random population sample of birth cohort
Authors
Criteria for dysfunction; Time frame
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK 19:42 Style file version Nov. 19th, 1999
Consecutive patients were given a standard interview
Venerophobia— persistent fear of having contracted an STD despite evidence to the contrary; abnormal sensation in genitalia— ill-defined pain, tingling, creeping of insects in the absence of objective abnormality; Current
MALE Premature ejaculation: 743/964 = 77% Delayed ejaculation: 5/964 = 0.6% Erectile dysfunction: 228/964 = 24% Venerophobia: 126/964 = 13% Abnormal sensation in genitalia: 14/964 = 14% Guilt about masturbation: 322/964 = 33% Nocturnal emission: 684/964 = 71% FEMALE Sexual arousal disorder: 0/36 = 0% Vaginismus: 0/36 = 0% Dyspareunia: 0/36 = 0%
PP016-291183 January 25, 2001
205
19:42
(Continued)
Strengths: Large sample of an underrepresented population; defined uncommon dysfunctions that were examined Limitations: Did not define criteria for some dysfunctions; conclusions are limited to sex clinic attendees; small number of women in sample
P2: FLF/FOM/FJQ
p < .01. p < .001.
N = 1000 Age: Most between 20–40 Gender: 96% male Race/ethnicity: North Indian clinic Health status: Healthy Recruitment: Sex therapy clinic
Archives of Sexual Behavior [asb]
∗ Prevalence decreased with age, † Prevalence increased with age,
Verma et al. (1998)
Premature ejaculation: (M): 31/626 = 5% Vaginismus: (F): 3/686 = 0.5%
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
206 Medical evaluation; Inability to achieve Medical history questionnaire given and/or maintain an to patients at a preventive erection suitable for medicine clinic on two occasions intercourse; 6–48 months apart Current
N = 3250 Age: 25–83 (M: 51) Gender: Male Race/ethnicity: 98% White Health status: Primary care clinic
Wei et al. (1994)
Positive report of symptom; Current
NR; Current
19:42
Strengths: Large clinical sample; operational definition; provides incidence statistics Limitations: Does not use DSM
Strengths: Examined anatomical and other potential covariates of sexual functioning Limitations: Did not report prevalence statistics for other dysfunctions assessed
Strengths: Relatively large sample representing diverse socioeconomic groups Limitations: No diagnostic criteria provided, though authors reported using a standard terminology for all interviews. It is unclear whether dyspareunia percentages are for the full group or sexually active women
January 25, 2001
Erectile dysfunction: Age <45: 4/939 = 0.4%; Age 45–54: 12/1173 = 1%; Age 55–64: 26/845 = 3%; Age ≥65: 29/293 = 10%
Dyspareunia/vaginal dryness: 30/73 = 41%
Dyspareunia Group 1: 20/250 = 8%;∗ Group 2: 34/250 = 14%; Group 3: 32/150 = 21% Loss of Libido Group 1: 93/250 = 37%; Group 2: 75/250 = 30%; Group 3: 53/150 = 35%
Prevalence
PP016-291183
Questionnaire, physical exam
N = 250 (Group 2) All women were Age: 30–55 (M: NR) postmenopausal; Gender: Female Structured Race/ethnicity: Pakistani interview Health status: Recruitment: Middle class health clinic N = 150 (Group 3) Age: 30–55 (M: NR) Gender: Female Race/ethnicity: Pakistani Health status: NR Recruitment: Wives of retired military officers
Procedure; Assessment
Methodological strengths and limitations
P2: FLF/FOM/FJQ
N = 73 Age: Range NR (M: 53) Gender: Female Race/ethnicity: NR Health status: Healthy Recruitment: Gynecology clinic
N = 250 (Group 1) Age: 30–55 (M: NR) Gender: Female Race/ethnicity: Pakistani Health status: NR Recruitment: Urban slum
Wasti et al. (1993)
Comparison sample
Criteria for dysfunction; Time frame
Archives of Sexual Behavior [asb]
Weber et al. (1995)
Sample
Authors
Table I. (Continued)
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
207 workers.
Erectile dysfunction: 29/55= 53%
Strengths: Reports a higher prevalence than previous studies with comparably aged men suggesting relationship between Alzheimer’s and E.D; examined potential confounds Limitations: Did not include a matched control group
January 25, 2001 19:42
∗ Significantly less than groups 2 and 3, p < .05. † Significantly lower than comparison groups of DAS
Persistent inadequate erections for intromission and sexual intercourse; Current
Strengths: Use of a measure with known psychometrics; use of two matched control groups Limitations: Self-selected sample
PP016-291183
Volunteers for a longitudinal study on Alzheimer’s disease; Interview
Volunteers for a study Not experiencing Erectile dysfunction: on the effects of a full erections 2/35 = 6%† stilbene derivative during sexual (DAS); Brief activity; Current sexual function questionnaire (Reynolds et al., 1988)
P2: FLF/FOM/FJQ
N = 55 Age: 50–80+ (M: 70) Gender: Male Race/ethnicity: 98% White Health status: Alzheimer’s disease Recruitment: Volunteers for a longitudinal study on Alzheimer’s disease
N = 30 Age: Range NR (M: 46) Gender: Male Race/ethnicity: NR Health status: Current stilbene derivative exposure (DAS) Recruitment: Factory workers N = 20 Age: Range NR (M: 45) Gender: Male Race/ethnicity: NR Health status: Former stilbene derivative exposure (DAS) Recruitment: Factory workers
Archives of Sexual Behavior [asb]
Zeiss et al. (1990)
Whelan et al. N = 35 (1996) Age: Range NR (M: 39) Gender: Male Race/ethnicity: NR Health status: Healthy (Additives workers) Recruitment: Factory workers
P1: FOM/LZR QC: FOM/FTK Style file version Nov. 19th, 1999
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
208
Style file version Nov. 19th, 1999
Simons and Carey
influence of such variation when nonstandard or nonexistent criteria coincide with extreme prevalence estimates. However, at the outset, it should be noted that lack of standard criteria is an important factor affecting comparisons among the majority of studies reviewed. Terminology utilized by the respective investigators is retained in Table I. Prior to each section we provide a brief description of the respective dysfunctions as defined in the DSM-IV. It is understood that in addition to the sexual dysfunction, the DSM system uses two additional criteria to determine a sexual disorder, (1) the disturbance causes marked distress or interpersonal difficulty, and (2) the dysfunction is not better accounted for by another Axis I disorder (except another sexual dysfunction) and is not due to the physiological effects of a substance or general medical condition. Thus, despite our adoption of the DSM nomenclature, the majority of the prevalence data pertain to sexual dysfunctions rather than DSM disorders.
FEMALE DYSFUNCTIONS Female Orgasmic Disorder is characterized by unusual difficulty in attaining orgasm. Determination of dysfunction depends upon age, sexual experience, and the adequacy of sexual stimulation. Prevalence rates of female orgasmic disorder range from 4% to 7% across three large population samples (Ernst et al., 1993; Lindal and Staffansson, 1993; Ventegodt, 1998). Each of these studies was conducted in a northern European country. Ernst et al. (1993) and Fugl-Meyer and Sjogren Fugl-Meyer (1999) reported 1-year prevalence statistics, Lindal and Stefansson (1993) reported lifetime prevalence, and Ventegodt (1998) reported current prevalence. Despite the more restrictive time frame, Ernst et al. (1993), Fugl-Meyer and Sjogren Fugl-Meyer (1999), and Ventegodt (1998) reported higher prevalence rates (7%–10%) than did Lindal and Stefansson (4% lifetime, 1993). A fifth study in the U.S. reported that 24% of women have been unable to reach orgasm for several months over the past year (Laumann et al., 1999). Only Lindal and Stefansson (1993) used DSM criteria whereas the other investigators used idiosyncratic definitions. These studies demonstrate the potential influence of nonstandard diagnostic criteria on prevalence estimates. It is uncertain whether the high estimate obtained by Laumann and colleagues (1999) reflects differences in assessment technique or population differences. Results from three studies in four primary care populations (Chandraiah et al., 1991; Read et al., 1997; Shahar et al., 1991) estimate the prevalence of female orgasmic disorder between 5% (lifetime; Chandraiah et al., 1991) and 42% (current; Read et al., 1997). The study by Chandraiah et al. (1991) was the only study to use DSM-III criteria and reported a 5% lifetime prevalence in 43 women attending a premenstrual syndrome (PMS) clinic. As expected, the prevalence of female orgasmic disorder appears to be higher in primary care settings compared with the
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
Sexual Dysfunctions
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
209
general population. However, methodological problems, especially poorly defined criteria, preclude definitive conclusions in this regard. Two studies in primary care settings report high prevalence rates but do not report the criteria used for assessing the dysfunction. In contrast, Chandraiah et al. (1991) reported a prevalence estimate based upon DSM-III criteria that is within the range of recent northern European population based estimates. Prevalence estimates of female orgasmic disorder obtained in sexuality clinics are 0% (Bhui et al., 1994), 22% (Goldmeier et al., 1997), and 41% (Bhui et al., 1994; Jindal and Dhall, 1990). However, Bhui et al. (1994) had a sample of three women and thus can not be taken as a useful estimate of prevalence. Neither Jindal and Dhall (1990) nor Goldmeier and colleagues (1997) used DSM criteria, making these estimates difficult to reconcile. Female Sexual Arousal Disorder is characterized by an insufficient lubrication–swelling response to sexual excitement. Little research has been conducted on female sexual arousal disorder. Lindal and Stefansson (1993) reported a lifetime prevalence of 6% in a large random population sample. This study used DSM-III criteria enhancing confidence in this estimate. Fugl-Meyer and Sjogren Fugl-Meyer (1999) reported a 1-year prevalence of 8% in a large Swedish sample. In contrast, Laumann and colleagues reported a 1-year prevalence of 19% in a representative U.S. sample. Lack of standard criteria across studies makes this large disparity difficult to interpret. Chandraiah et al. (1991) reported a lifetime prevalence of 21% in a primary care setting based upon DSM-III criteria. One study examined women attending a sex therapy clinic (Verma et al., 1998). This was a small sample from a North Indian clinic, in which no cases of female dysfunctions were reported. Thus, more research on the prevalence of arousal disorders in women is needed. Hypoactive Sexual Desire Disorder (HSDD) is defined as deficient sexual fantasies and desire for sexual activity. Determination of dysfunction is relative to age and the context of the person’s life. Prevalence estimates for HSDD among females range from 5% (Ventegodt, 1998) to 46% (Chiechi et al., 1997) across seven studies. The two highest estimates (46%; Chiechi et al., 1997; 37%, Wasti et al., 1993) are from samples of postmenopausal women. The other studies report 1-year estimates ranging from 14% (Fugl-Meyer and Sjogren Fugl-Meyer) to 33% (Laumann et al., 1999). Only Lindal and Stefansson (1993) used DSM-III criteria, and they reported a significant higher lifetime prevalence among women (16%) than among men (4%). The many methodological differences across few studies makes interpretation of this large range difficult. Six studies examined HSDD among women in primary care settings. Brown and colleagues reported a current prevalence of 20% (Brown and Rundell, 1993) and 31% (Brown and Rundell, 1990) among HIV+ women. These two studies are part of a 5-year longitudinal study and may reflect overlap of participants between studies. Jamieson and Steege (1996) reported a current prevalence of 10% among
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
210
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
Simons and Carey
women in a gynecology clinic. Goggin et al. (1998) reported a positive relation between HSDD and depressive symptoms, low life satisfaction, and perceived risk for HIV infection. This study reported a current estimate of 39% in a sample obtained from the community and HIV/AIDS health clinics. Chandraiah et al. (1991) studied women in a PMS clinic and reported a lifetime prevalence of 21% based upon DSM-III criteria. Chiechi et al. (1997) reported a current prevalence of 46% among postmenopausal women. Thus, current estimates range from 10% (Jamieson and Steege, 1996) to 46%. Dyspareunia is characterized by persistent genital pain during sexual intercourse. Dyspareunia is not diagnosed if the pain is exclusively due to vaginismus or lack of lubrication. Sexual pain disorders have been the focus of a relatively large number of studies. Prevalence estimates range from 3% (Lindal and Stefansson, 1993; Ventegodt, 1998) to 18% (Moody and Mayberry, 1993) in the general population. This relatively large range is difficult to explain in terms of methodological differences. For example, percentages at the high end were reported by Moody and Mayberry (1993) and Glatt et al. (1990), who report point prevalence as opposed to the 3% lifetime prevalence reported by Lindal and Stefansson (1993). Similarly, both high (Glatt et al., 1990; Laumann et al., 1999; Moody and Mayberry, 1993) and low (Ernst et al., 1993; Rekers et al., 1992; Ventegodt, 1998) estimates have been reported based on operational as opposed to DSM criteria. Only Lindal and Stefansson (1993) used DSM-III criteria. The only consistent difference is that the lower estimates are from Northern European countries whereas the higher ones are from the U.S. In general practice settings, current estimates range from 3% (Heisterberg, 1993) to 46% (Jamieson and Steege, 1996) across six studies. Jamieson and Steege (1996) reported point prevalence (in a large primary care sample) of “pain during or after intercourse.” The high prevalence in this and the Weber et al. (1995) sample (41%) is three times greater than the next highest estimate of 14% (Heisterberg, 1993). These high estimates may be due to the operational definition of the disorder. The DSM-IV diagnostic criteria exclude pain that is associated exclusively with lack of lubrication or vaginismus. Neither of these high estimates exclude pain secondary to these causes. In fact, Weber and colleagues (1995) specifically included vaginal dryness as a sufficient criterion. Chandraiah et al. (1991), using DSM-III criteria, reported a lifetime prevalence of 12%. The range of estimates obtained illustrate the differences that result from nonstandard definitions and the importance of clearly specifying the definitions used. Two studies reported prevalence of dyspareunia in sexuality clinic settings. No female dysfunctions were reported in a large sample of women attending a North Indian clinic (Verma et al., 1998). Jindal and Dhall (1990) reported a current prevalence of 13% in an infertility clinic. In relation to other sexual dysfunctions, dyspareunia appears to be a less common presenting problem at sexuality clinics. This may be because pain disorders are more likely to present at primary care rather than at sexuality specific clinics.
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Sexual Dysfunctions
Style file version Nov. 19th, 1999
211
Older women have been a population in which dyspareunia has been a focus of research. There is evidence that dyspareunia is more prevalent in postmenopausal women (Rekers et al., 1992). Prevalence estimates in community samples of postmenopausal women range from 2% (Barlow et al., 1997) to 21% (Wasti et al., 1993). Barlow et al. (1997) reported 2-year prevalence data of “painful intercourse.” However, only 19% of the sample reported having penetrative intercourse. Thus, the prevalence of pain during intercourse may be more appropriately estimated at 9%. Thus, prevalence estimates range from 9% to 21% across five studies (Barlow et al., 1997; Diokno et al., 1990; Ramoso-Jalbuena, 1994; Rekers et al., 1992; Wasti et al., 1993). Vaginismus is defined as persistent involuntary spasm of the vagina interfering with intercourse. Prevalence data for vaginismus are scant without the benefit of multiple studies within populations. Read et al. (1997) reported a current estimate of 30% in a primary care setting. Community estimates range from 0.5% to 1% (Fugl-Meyer and Sjogren Fugl-Meyer, 1999; Ventegodt, 1998). Verma et al. (1998) reported a prevalence of 0% in a sexuality clinic sample. Goldmeier et al. (1997) reported a current prevalence of 25% in a STD clinic. Thus, no clear estimate emerges.
MALE DYSFUNCTIONS Male Orgasmic Disorder is characterized by persistent difficulty in attaining orgasm. Determination of dysfunction depends upon age and the adequacy of sexual stimulation. The limited data available regarding male orgasmic disorder suggest that prevalence rates are relatively low. Community estimates of male orgasmic disorder range from 0% (Schiavi et al., 1995) to 3% across six studies (Fugl-Meyer and Sjogren Fugl-Meyer, 1999; Lindal and Stefansson, 1993; Singer et al., 1992; Solstad and Hertoft, 1993; Ventegodt, 1998). A much higher 1-year prevalence estimate of 8% is reported by Laumann et al. (1999). Estimates from eight primary care samples across four studies range from 0% (Catalan et al., 1992a) to 36% (El-Rufaie et al., 1997) with a median of 9% (Shahar et al., 1991). Estimates from a sexuality clinic sample across four studies range from 0% (Bhui et al., 1994) to 38% (Catalan et al., 1992b). One study reported a lifetime prevalence of 39% in gay men (Rosser et al., 1997). Estimates from samples of gay men were notably higher than from other samples (i.e., ≥38% Catalan et al., 1992b; Rosser et al., 1997 vs. ≤6% Bhui et al., 1994; Verma et al., 1998). We hypothesize that this difference may reflect greater recognition of the threat of infection with HIV. In this regard, two studies reported an increased prevalence of male orgasmic disorder among men with HIV. Current estimates were 20% (vs. 0%; Catalan et al., 1992a) and 38% (vs. 9%; Catalan et al., 1992b). Premature ejaculation is defined as ejaculation with minimal stimulation before the person wishes it. Age, novelty of the sexual situation, and recent frequency of sexual activity are considered in determining premature ejaculation.
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
212
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
Simons and Carey
Community estimates of the current/1-year prevalence of premature ejaculation range from 4% (Ernst et al., 1993; Fugl-Meyer and Sjogren Fugl-Meyer, 1999) to 5% (Schiavi, 1995; Ventegodt, 1998) across three studies. Two additional studies reported significantly higher 1-year estimates of 14% (Solstad and Hertoft, 1993) and 29% (Laumann et al., 1999). The reason for these discrepant estimates is not clear. Schiavi et al. (1995) reported a current estimate of 20% among former alcohol-dependent men. In primary care settings current estimates range from 2% (Nirenberg et al., 1991) to 31% (Read et al., 1997) across four studies. The lowest estimate was among alcohol-dependent individuals choosing not to participate in additional research that would have involved intrusive measurements. Volunteers for the study reported a prevalence of 24% (Nirenberg et al., 1991). Thus, prevalence of premature ejaculation may be estimated at between 4% (Catalan et al., 1992b) and 31% (Read et al., 1997). There is an extremely large range of estimates of the current prevalence of premature ejaculation in sexuality clinic samples. Current estimates range from 0% (Bhui et al., 1994; Catalan et al., 1992b) to 77% (Verma et al., 1998) across five studies. The highest estimate (77%; Verma et al., 1998) is from a Northern Indian population. This figure is nearly four times the next highest estimate of 22% (Goldmeier et al., 1997). If the estimate of Verma et al. (1998) is considered an outlier, a more accurate range is 0% (Bhui et al., 1994; Catalan et al., 1992b) to 22% (Goldmeier et al., 1997). Prevalence of premature ejaculation does not appear to be higher among individuals attending sexuality clinics than in primary care settings. Erectile Disorder is characterized by inadequate erections for sexual activity. The individual may have difficulty either attaining or maintaining an erection. Current/1-year prevalence estimates in the general population range from 0% (Ernst et al., 1993) to 10% (Laumann et al., 1999) across 10 studies. The prevalence of erectile disorder increases with age, history of heart disease, diabetes, treated hypertension, untreated ulcer, arthritis, allergy, and smoking (Feldman et al., 1994; Mannino et al., 1994; Panser et al., 1995; Ventegodt, 1998). Five studies have examined current erectile disorder in older men in the community (Cogen and Steinman, 1990; Feldman et al., 1994; Jonler et al., 1995; Panser et al., 1995; Schiavi et al., 1991). Estimates range from 20% reporting erections less than half the time when sexually stimulated in the last year (Jonler et al., 1995) to 52% (Feldman et al., 1994). The estimate by Feldman et al. (1994) combines “minimal, moderate, and complete” erectile dysfunction. The prevalence of moderate erectile dysfunction in the sample is 25%, closer to that of Jonler et al. (1995). In general practice settings, current estimates of erectile disorder range from 0.4% (Wei et al., 1994) to 37% (Singer et al., 1992) across seven studies. This wide fluctuation can be attributed to differences among assessment criteria and presence
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
Sexual Dysfunctions
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
213
of important risk factors in the samples, including advanced age, medications (Read et al., 1997), diabetes, and medicated hypertension (Modebe, 1990). The highest rates were reported among patients with Parkinson’s disease (60%; Singer et al., 1992) and Alzheimer’s disease (55%; Zeiss et al., 1990). In sexuality clinics current rates of erectile dysfunction range from 1% (Jindal and Dhall, 1990) to 53% (Bhui et al., 1994) across seven studies. The lowest estimate was based upon interviews of women regarding the sexual functioning of both themselves and their male partner. It is worth noting that not one of these studies utilized the DSM criteria and few provided operational definitions. Thus, it is not surprising that there exists such disparity in prevalence estimates. Goldmeier et al. (1997), Rosser et al. (1997), and Verma et al. (1998) may be the most methodologically sound studies with respect to sample size and criteria. These studies provide current estimates of 19%, 15%, and 24% respectively. Rosser et al. (1997) reported lifetime estimates of 40% (getting an erection) and 46% (maintaining an erection). Hypoactive Sexual Desire Disorder is characterized by deficient sexual fantasies and desire for sexual activity. Determination of dysfunction is relative to age and the context of the person’s life. Estimates range from a current/1-year prevalence of 0% (Schiavi et al., 1995) to 7% (Ernst et al., 1993) across seven community samples studies. Panser et al. (1995) reported a significant positive correlation between age and HSDD (age 70 and over prevalence = 26%). Laumann et al. (1999) reported the highest estimate in the general population (1-year; 16%). Three studies reported prevalence data on HSDD among men in primary care settings. Estimates range from a current prevalence of 3% (Jamieson and Steege, 1996) to 55% (Catalan et al., 1992a) among individuals with HIV. Three studies also examined HSDD in sexuality clinic samples (Bhui et al., 1994; Catalan et al., 1992b; Rosser et al., 1997). Current estimates of HSDD ranged from 0% (Bhui et al., 1994) to 75% (among HIV+ males; Catalan et al., 1992b). Three studies examined HSDD in relation to HIV infection status (Catalan et al., 1992a; Catalan et al., 1992b; Pace et al., 1990). Estimates of current HSDD among HIV+ persons range from 13% (Pace et al., 1990) to 75% (Catalan et al., 1992b). The two studies by Catalan et al. (1992a,b) reported substantially higher estimates than the other studies (75% and 55% respectively). Unfortunately they did not report any criteria and it is unclear to what extent their “loss of interest in sex” corresponds to HSDD. Catalan and colleagues did not find statistically significant differences of prevalence of HSDD across HIV+/− groups. However, Pace et al. (1990) did find a higher prevalence among HIV+ persons than among a control sample from an alcohol treatment center. Dyspareunia is characterized by recurrent genital pain during sexual intercourse. Sexual pain disorder among men appears to be significantly less prevalent than in women (Fass et al., 1998). Estimates across seven studies range from a
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
QC: FOM/FTK PP016-291183
January 25, 2001
214
19:42
Style file version Nov. 19th, 1999
Simons and Carey
lifetime prevalence of 0.2% in a random population sample (Lindal and Stefansson, 1993) to a lifetime prevalence of 8% in a combined community and clinical sample (Metz and Seifert, 1990). One study examined painful insertive and receptive anal sex in gay men (Rosser et al., 1997). The study reported current prevalence estimates of 3% (insertive) and 16% (receptive).
DISCUSSION Our review of recent prevalence estimates for the sexual dysfunctions is largely consistent with that reported 10 years ago by Spector and Carey (1990). Community samples indicate a current prevalence of 0%–3% for male orgasmic disorder, 0%–5% for erectile disorder, and 0%–3% for male HSDD. Pooling current and 1-year figures provides community prevalence estimates of 7%–10% for female orgasmic disorder and 4%–5% for premature ejaculation. For the point of comparison, Spector and Carey (1990) reported a current prevalence of 4%–10% for male orgasmic disorder, 4%–9% for male erectile disorder, 5%–10% for female orgasmic disorder, and 36%-38% for premature ejaculation. Thus, only the prevalence of premature ejaculation is markedly different. The high estimate for premature ejaculation reported by Spector and Carey (1990) was based upon two relatively small samples. The much lower estimate obtained in this review is based upon four studies with a total of over 2000 men and is thus more representative of the general population. The current review was able to provide an estimate of the prevalence of male HSDD, a figure unavailable previously. Stable community estimates of other sexual dysfunctions remain uncertain. Spector and Carey (1990) made four suggestions for new research in this area. Specifically, they called for increased use of (1) stratified samples representative of the general population; (2) psychometrically sound assessment techniques to facilitate interpretation and replication; (3) a common classification system to aid comparison across studies; and (4) collection of incidence data. There have been some notable studies over the past 10 years that have incorporated these methodological recommendations. For example, Ernst et al. (1993) and Rekers et al. (1992) used stronger sampling techniques, stratifying by psychological distress and age, respectively. The field has also had the benefit of several larger scale (>1000 participants) random population surveys (e.g., Barlow et al., 1997; Fugl-Meyer and Sjogren Fugl-Meyer, 1999; Laumann et al., 1999; Ventegodt, 1998). There has also been progress in assessment techniques. For example, Brown et al. (1990) and Goggin et al. (1998) used a modified version of the Structured Clinical Interview for DSM-III-R (Brown and Rundell, 1993). Reports of interrater reliability in several studies provide a measure of reliability of diagnoses (Brown and Rundell, 1990; Meyer-Bahlburg et al., 1993). Additional assessment instruments with known psychometric characteristics are also being used (e.g., the DISS-IIIA
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
Sexual Dysfunctions
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
215
was used by Robins, 1986; Chandraiah et al., 1991; Lindal and Stefansson, 1993; and Meyer-Bahlburg et al., 1993; the GRISS was used by Rust and Golombok, 1986; and Goldmeier et al., 1997). Review of the studies in which the most psychometrically sound assessment techniques were used also demonstrates a trend toward using the DSM as a common classification system. Incidence data continue to be sparse. Wei et al. (1994) is one exception. These authors report incidence data on erectile dysfunction stratified by age. There have been a small number of excellent studies that have incorporated many important methodological features into study design (e.g., Fugl-Meyer and Sjogren Fugl-Meyer, 1999; Lindal et al., 1993). However, despite the increased attention in the past decade to the study of sexual dysfunctions there appears to have been relatively little methodological improvement overall. We identify three successive strategies for improvement in relation to assessment criteria. First, the criteria for determining a dysfunction need to be clearly reported. Although several investigators have used operational definitions, many studies failed to report the criteria they used in the paper. Lack of consistent reporting of assessment criteria make comparisons across studies difficult and hinder the accumulation of data across studies to enhance knowledge. Second, standard criteria for the sexual dysfunctions need to be adopted. The use of standard diagnostic criteria appears to be the exception rather than the rule among the studies reviewed. The DSM and the multiaxial system proposed by Schover et al. (1982) provide two potential options. The third avenue for development is to examine sexual disorders rather than simply the dysfunction. The omission of psychological sequelae of the sexual dysfunctions is a significant methodological concern. According to the DSM-IV (APA, 1994), sexual disorder diagnoses need to be based on three criteria: (A) sexual dysfunction (i.e., physical/psychological manifestation [e.g., lack of orgasm, lack of erection, pain during intercourse, lack of sexual interest, etc.]), (B) the disturbance causes marked distress or interpersonal difficulty, and (C) the dysfunction is not better accounted for by another Axis I disorder (except another sexual dysfunction) and is not due to the physiological effects of a substance or general medical condition. Criterion A has frequently been incorporated into most investigators’ operational definitions. However, criteria B and C are typically omitted from the diagnostic criteria. From a clinical standpoint, the report of accompanying distress and/or interpersonal difficulty is important. However, this criterion was rarely addressed in the reported criteria in the studies reviewed. Typically, only the prevalence of symptoms is reported although some exceptions can be noted. For example, Amr et al. (1997) report the prevalence of both erectile disorder determined by DSM-III-R criteria as well as prevalence of erectile dysfunction symptoms. The latter led to a prevalence rate of 27% whereas the former resulted in a much lower (5%) rate. This additional level of detail provides especially helpful information regarding the underlying development of the disorder. For example,
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
216
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
Simons and Carey
Amr et al. (1997) reported significant increases in erectile dysfunction symptoms but not erectile disorder in relation to pesticide exposure. Such findings may be informative in understanding the biological and psychological contributions to the development of sexual disorders. The study by Fugl-Meyer and Sjogren Fugl-Meyer (1999) is particularly informative in respect to the relations between sexual dysfunction and sexual disorder characterized by resultant perceived psychosocial problems. This study assessed the prevalence of the dysfunctions as well as the percentage of participants who perceived their sexual dysfunction as problematic. For some dysfunctions, there was a high concordance between the presence of a dysfunction and perceived problems. For example, 69% of the men reporting erectile dysfunction reported that it was problematic. In contrast, only 45% of women with orgasmic dysfunction perceived it as problematic. Thus, in this study, if one defined female orgasmic disorder as the inability to attain orgasm, the 1-year prevalence rate is 22%. In contrast, the prevalence rate is only 10% if one defines the disorder as the presence of the dysfunction and the dysfunction causing a problem (marked distress or interpersonal difficulty in DSM-IV terminology). This study clearly differentiates between sexual functioning on the one hand and a psychological disorder defined in part by subjective distress and disturbance in interpersonal relations. This study provides a clear demonstration of how differences in diagnostic criteria can have a profound effect on prevalence estimates. Such differences contribute to the wide discrepancies seen across some studies. We acknowledge that most studies were designed only to obtain data on the occurrence of a symptom and that investigators did not claim to be assessing a disorder defined in the DSM. Determination of the appropriateness of assessing a sexual dysfunction versus a disorder (in the DSM sense) rests upon the goals of the study. Assessing solely the dysfunction is appropriate if the potential accompanying distress or interpersonal conflict is not of interest. In some cases, the symptom is an important focal point as in the relationship between erectile dysfunction and health problems such as diabetes mellitus (Weinhardt and Carey, 1996). In such studies, biological precursors are of interest. In more clinically focused research, determination of whether the dysfunction is accompanied by significant distress or interpersonal conflict is relevant. It is such psychosocial problems that are the impetus for intervention not variation in sexual functioning per se. For symptoms such as reduced sex drive, the importance of the symptom in isolation from DSM criteria is unknown, and the prevalence estimates are less useful. The wealth of studies conducted over the past 10 years is encouraging as is the adoption of the methodological suggestions that were outlined in Spector and Carey (1990). We noted some of the studies that had particularly sound methodological designs. These studies are exemplars that could guide the continuing development in the study of sexual functioning. These exemplars are unfortunately few in comparison to the full collection of studies. Many continue to have methodological problems that limit their potential usefulness. With continued attention to
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
QC: FOM/FTK PP016-291183
January 25, 2001
Sexual Dysfunctions
19:42
Style file version Nov. 19th, 1999
217
statistical design, it is hoped that methodologically rigorous studies will no longer be the exception to the rule.
REFERENCES Aizenberg, D., Zemishlany, Z., Dorfman-Etrog, P., and Weizman, A. (1995). Sexual dysfunction in male schizophrenic patients. J. Clin. Psychiat. 56: 137–141. American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders, 4th Ed., Author, Washington, DC. Amr, M. M., Halim, Z. S., and Moussa, S. S. (1997). Psychiatric disorders among Egyptian pesticide applicators and formulators. Environ. Res. 73: 193–199. Barlow, D. H., Cardozo, L. D., Francis, R. M., Griffin, M., Hart, D. M., Stephens, E., and Sturdee, D. W. (1997). Urogenital aging and its effect on sexual health in older British women. Br. J. Obs. Gyn. 104: 87–91. Bhui, K., Herriot, P., Dein, S., and Watson, J. P. (1994). Asians presenting to a sex and marital therapy clinic. Int. J. Soc. Psychiat. 40: 194–204. Boulet, M. J., Oddens, B. J., Lehert, P., Vemer, H. M., and Visser, A. (1994). Climacteric and menopause in seven South-east Asian countries. Maturitas 19: 157–176. Brown, G. R., and Rundell, J. B. (1993). A prospective study of psychiatric aspects of early HIV disease in women. Gen. Hosp. Psychiat. 15: 139–147. Brown, G. R., and Rundell, J. R. (1990). Prospective study of psychiatric morbidity in HIV-seropositive women without AIDS. Gen. Hosp. Psychiat. 12: 30–35. Catalan, J., Klimes, I., Bond, A., Day, A., Garrod, A., and Rizza, C. (1992a). The psychosocial impact of HIV infection in men with haemophilia: Controlled investigation and factors associated with psychiatric morbidity. J. Psychosom. Res. 36: 409–416. Catalan, J., Klimes, I., Day, A., Garrod, A., Bond, A., and Gallwey, J. (1992b). The psychosocial impact of HIV infection in gay men. A controlled investigation and factors associated with psychiatric morbidity. Br. J. Psychiat. 161: 774–778. Chandraiah, S., Levenson, J. L., and Collins, J. B. (1991). Sexual dysfunction, social maladjustment, and psychiatric disorders in women seeking treatment in a premenstrual syndrome clinic. Int. J. Psychiat. Med. 21: 189–204. Chiechi, L. M., Granieri, M., Lobascio, A., Ferreri, R., and Loizzi, P. (1997). Sexuality in the climacterium. Clin. Exp. Obstet. Gynecol. 24: 158–159. Cogen, R., and Steinman, W. (1990). Sexual function and practice in elderly men of lower socioeconomic status. J. Fam. Pract. 31: 162–166. Derogatis, L. R. (1977). Administration, scoring, and procedures for the R (revised) version and other instruments of the psychopathology rating scale series. SCL-90, Johns Hopkins Press, Baltimore. Derogatis, L. R., and Meyer, J. (1979). A psychological profile of the sexual dysfunctions. Arch. Sex. Behav. 8: 201–223. Diokno, A. C., Brown, M. B., and Herzog, A. R. (1990). Sexual function in the elderly. Arch. Intern. Med. 150: 197–200. El-Rufaie, O. E. F., Bener, A., Abuzeid, M. S. O., and Ali, T. A. (1997). Sexual dysfunction among type II diabetic men: A controlled study. J. Psychosom. Res. 43: 605–612. Ernst, C., Foldenyi, M., and Angst, J. (1993). The Zurich Study: XXI. Sexual dysfunctions and disturbances in young adults. Data of a longitudinal epidemiological study. Eur. Arch. Psychiat. Clin. Neurosci. 243: 179–188. Fass, R., Fullerton, S., Naliboff, B., Hirsh, T., and Mayer, E. A. (1998). Sexual dysfunction in patients with irritable bowel syndrome and non-ulcer dyspepsia. Digestion 59: 79–85. Feldman, H. A., Goldstein, I., Hatzichristou, D. G., Krane, R. J., and McKinlay, J. B. (1994). Impotence and its medical and psychosocial correlates: Results of the Massachusetts Male Aging Study. J. Urol. 151: 54–61. Fugl-Meyer, A. R., and Sjogren Fugl-Meyer, K. (1999). Sexual disabilities, problems, and satisfaction in 18–74 year old Swedes. Scand. J. Sexol. 3: 79–105.
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
218
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
Simons and Carey
Glatt, A. E., Zinner, S. H., and McCormack, W. M. (1990). The prevalence of dyspareunia. Obstet. Gynecol. 75: 433–436. Goggin, K., Engelson, E. S., Rabkin, J. G., and Kotler, D. P. (1998). The relationship of mood, endocrine, and sexual disorders in human immunodeficiency virus positive (HIV+) women: An exploratory study. Psychosom. Med. 60: 11–16. Goldberg, D. P., and Hillier, A. (1979). Manual of the General Health Questionnaire, Egyptian Psychiatric Association, Cairo. Goldmeier, D., Keane, F. E., Carter, P., Hessman, A., Harris, J. R., and Renton, A. (1997). Prevalence of sexual dysfunction in heterosexual patients attending a central London genitourinary medicine clinic. Int. J. STD AIDS 8: 303–306. Heisterberg, L. (1993). Factors influencing spontaneous abortion, dyspareunia, dysmenorrhea, and pelvic pain. Obstet. Gynecol. 81: 594–597. Ismael, N. N. (1994). A study on the menopause in Malaysia. Maturitas 19: 205–209. Jamieson, D. J., and Steege, J. F. (1996). The prevalence of dysmenorrhea, dyspareunia, pelvic pain, and irritable bowel syndrome in primary care practices. Obstet. Gynecol. 87: 55–58. Jindal, U. N., and Dhall, G. I. (1990). Psychosexual problems of infertile women in India. Int. J. Fertil. 35: 222–225. Jonler, M., Moon, T., Brannan, W., Stone, N. N., Heisey, D., and Bruskewitz, R. C. (1995). The effect of age, ethnicity and geographical location on impotence and quality of life. Br. J. Urol. 75: 651–655. Laumann, E. O., Paik, A., and Rosen, R. C. (1999). Sexual dysfunction in the United States. JAMA 281: 537–544. Lindal, E., and Stefansson, J. G. (1993). The lifetime prevalence of psychosexual dysfunction among 55 to 57-year-olds in Iceland. Soc. Psychiat. Psychiatr. Epidemiol. 28: 91–95. Mannino, D. M., Klevens, R. M., and Flanders, W. D. (1994). Cigarette smoking: An independent risk factor for impotence? Am. J. Epidemiol. 140: 1003–1008. Metz, M., and Seifert, M. H. (1990). Men’s expectations of physicians in sexual health concerns. J. Sex Marital Ther. 16: 79–88. Meyer-Bahlburg, H. F., Nostlinger, C., Exner, T. M., Ehrhardt, A. A., Gruen, R. S., Lorenz, G., Gorman, J. M., el-Sadr, W., and Sorrell, S. J. (1993). Sexual functioning in HIV+ and HIV− injected drugusing women. J. Sex Marital Ther. 19: 56–68. Modebe, O. (1990). Erectile failure among medical clinic patients. Afr. J. Med. Med. Sci. 19: 259– 264. Moody, G. A., and Mayberry, J. F. (1993). Perceived sexual dysfunction amongst patients with inflammatory bowel disease. Digestion 54: 256–260. Nirenberg, T. D., Wincze, J. P., Bansal, S., Liepman, M. R., and et al. (1991). Volunteer bias in a study of male alcoholics’ sexual behavior. Arch. Sex. Behav. 20: 371–379. Pace, J., Brown, G. R., Rundell, J. R., and Paolucci, S. (1990). Prevalence of psychiatric disorders in a mandatory screening program for infection with human immunodeficiency virus: A pilot study. Mil. Med. 155: 76–80. Panser, L. A., Rhodes, T., Girman, C. J., Guess, H. A., Chute, C. G., Oesterling, J. E., Lieber, M. M., and Jacobsen, S. J. (1995). Sexual function of men ages 40 to 79 years: The Olmsted County Study of Urinary Symptoms and Health Status Among Men. J. Am. Geriatr. Soc. 43: 1107–1111. Ramoso-Jalbuena, J. (1994). Climacteric Filipino women: A preliminary survey in the Philippines. Maturitas 19: 183–190. Read, S., King, M., and Watson, J. (1997). Sexual dysfunction in primary medical care: Prevalence, characteristics and detection by the general practitioner. J. Pub. Health Med. 19: 387–391. Regier, D. A., Farmer, M. E., Rae, D. S., Zocke, B. Z., Keith, S. J., Judd, L. L., and Goodwin, F. K. (1990). Comorbidity of mental disorders with alcohol and other drug abuse. J. Am. Med. Assoc. 264: 2511–2518. Rekers, H., Drogendijk, A. C., Valkenburg, H. A., and Riphagen, F. (1992). The menopause, urinary incontinence and other symptoms of the genito-urinary tract. Maturitas 15: 101–111. Reynolds, C. F., Frank, E., Thase, M. E., Houck, P. R., Jennings, J. R., Howell, J. R., Lilienfeld, S. O., and Kupfer, D. J. (1988). Assessment of sexual function in depressed, impotent, and healthy men: Factor analysis of a brief sexual function questionnaire for men, Psychiat. Res., 24: 231–250. Robins, L. N., Helzer, J. E., and Croughan, J. (1981). The NIMH Diagnostic Interview Schedule: Version III, Public Health Service (HSS).
P1: FOM/LZR
P2: FLF/FOM/FJQ
Archives of Sexual Behavior [asb]
Sexual Dysfunctions
QC: FOM/FTK PP016-291183
January 25, 2001
19:42
Style file version Nov. 19th, 1999
219
Rosser, B. R., Metz, M. E., Bockting, W. O., and Buroker, T. (1997). Sexual difficulties, concerns, and satisfaction in homosexual men: An empirical study with implications for HIV prevention. J. Sex Marital Ther. 23: 61–73. Rust, J., and Golombok, S. (1986). The Golombok-Rust Inventory of Sexual Satisfaction (GRISS), NFER/Nelson, Windsor. Schiavi, R. C., Mandeli, J., Schreiner-Engel, P., and Chambers, A. (1991). Aging, sleep disorders, and male sexual function. Biol. Psychiat. 30: 15–24. Schiavi, R. C., Stimmel, B. B., Mandeli, J., and White, D. (1995). Chronic alcoholism and male sexual function. Am. J. Psychiat. 152: 1045–1051. Schover, L. R., Friedman, J. M., Weiler, S. J., Heiman, J. R., and LoPiccolo, J. (1982). Multiaxial problem-oriented system for sexual dysfunctions: An alternative to DSM-III. Arch. Gen. Psychiat. 39: 614–619. Shahar, E., Lederer, J., and Herz, M. J. (1991). The use of a self-report questionnaire to assess the frequency of sexual dysfunction in family practice clinics. Fam. Pract. 8: 206–212. Singer, C., Weiner, W. J., and Sanchez-Ramos, J. R. (1992). Autonomic dysfunction in men with Parkinson’s disease. Eur. Neurol. 32: 134–140. Solstad, K., and Hertoft, P. (1993). Frequency of sexual problems and sexual dysfunction in middle-aged Danish men. Arch. Sex. Behav. 22: 51–58. Spector, I. P., and Carey, M. P. (1990). Incidence and prevalence of the sexual dysfunctions: A critical review of the empirical literature. Arch. Sex. Behav. 19(4): 389–408. Ventegodt, S. (1998). Sex and the quality of life in Denmark. Arch. Sex. Behav. 27: 295–307. Verma, K. K., Khaitan, B. K., and Singh, O. P. (1998). The frequency of sexual dysfunctions in patients attending a sex therapy clinic in north India. Arch. Sex. Behav. 27: 309–314. Wasti, S., Robinson, S. C., Akhtar, Y., Khan, S., and Badaruddin, N. (1993). Characteristics of menopause in three socioeconomic urban groups in Karachi, Pakistan. Maturitas 16: 61–69. Weber, A. M., Walters, M. D., Schover, L. R., and Mitchinson, A. (1995). Vaginal anatomy and sexual function. Obstet. Gynecol. 86: 946–949. Wei, M., Macera, C. A., Davis, D. R., Hornung, C. A., Nankin, H. R., and Blair, S. N. (1994). Total cholesterol and high density lipoprotein cholesterol as important predictors of erectile dysfunction. Am. J. Epidemiol. 140: 930–937. Weinhardt, L. S., and Carey, M. P. (1996). Prevalence of erectile disorder among men with diabetes mellitus: Review of the empirical literature. J. Sex Res. 33: 205–214. Whelan, E. A., Grajewski, B., Wild, D. K., Schnorr, T. M., and Alderfer, R. (1996). Evaluation of reproductive function among men occupationally exposed to a stilbene derivative: II. Perceived libido and potency. Am. J. Ind. Med. 29: 59–65. Zeiss, A. M., Davies, H. D., Wood, M., and Tinklenberg, J. R. (1990). The incidence and correlates of erectile problems in patients with Alzheimer’s disease. Arch. Sex. Behav. 19: 325–331.
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
QC: FTK PP016-291184-86
January 25, 2001
11:30
Style file version Nov. 19th, 1999
Archives of Sexual Behavior, Vol. 30, No. 2, 2001
BOOK REVIEWS Sexual Harassment: Theory, Research, and Treatment. Edited by William O’Donohue. Allyn and Bacon, Needham Heights, Massachusetts, 1997, 315 pp., $73.00. Reviewed by Lucia F. O’Sullivan, Ph.D.1
In the United States, the general public had its attention fixated on sexual harassment—thanks primarily to the media expos´e of the President’s intimate association with a White House intern. If one seeks an antidote to this hysteria, this book constitutes a balanced, comprehensive review of the issues surrounding judgments about sexual harassment. It covers a broad array of topics, including some of special interest, such as sexual harassment in the military and in schools. This book will appeal to a wide college audience, both graduate and undergraduate. Most other publications addressing sexual harassment are essentially resource guidebooks offering practical advice on how to confront the problem of sexual harassment (e.g., Petrocelli and Repa, 1998; Sandler, 1996) or how to create credible prevention policies (e.g., Wagner, 1992). This one serves a similar purpose, but is also empirically and theoretically rich in content. Despite its title, however, there is no coverage of treatment issues except in the documentation of damaging sequelae experienced by victims; thus, this book will not be of particular use to those seeking treatment guidelines. The first few chapters deal with definitional concerns. Fitzgerald et al. explain how the phenomenon needs to be understood primarily in terms of the impact of particular actions rather than the intentions associated with those actions, and that it is the victim’s subjective experience or interpretation of a person’s action that determines whether, in fact, sexual harassment has occurred. In addition, most of these contributors note that research in this area is plagued by inconsistencies in empirical definitions employed across studies and in the classification of behaviors. Thus, readers will need to develop a certain degree of tolerance for ambiguity to fully appreciate this book. 1 HIV
Center for Clinical and Behavioral Studies, New York State Psychiatric Institute, Unit 15, 1051 Riverside Drive, New York, New York 10032-2695. 221 C 2001 Plenum Publishing Corporation 0004-0002/01/0400-0221$19.50/0 °
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
222
QC: FTK PP016-291184-86
January 25, 2001
11:30
Style file version Nov. 19th, 1999
Book Reviews
Leeser and O’Donohue contribute an interesting, though somewhat labored, moral philosophical analysis of the concept of wrongness. Conte provides absorbing coverage of the evolution of sexual harassment law since the 1970s, which will be of considerable use to many scholars. This review includes documentation of the most significant legal precedents paving the way to our current laws. Gruber evaluates epidemiological data on sexual harassment and argues that there may be “a universality of sexual harassment experiences” (p. 88). If this is so, it is a conclusion based on data gleaned primarily from North American and Western European sources. A consistent theme throughout these chapters is that sexual harassment is a crime against women committed by men. Thus, little attention is given to the topics of same-sex harassment cases or sexual harassment of men by women. This theme may irk some readers, but the contributors have gone to considerable effort to support this perspective with epidemiological and social science data. Hotelling and Zuber discuss the “saliency of gender” and apply feminist theory to the topic of sexual harassment with emphasis on the writings of MacKinnon. Their arguments, unfortunately, make women appear hypersensitive and in need of extraordinary care. However, they conclude their chapter with some valuable suggestions for prevention which moves beyond the typical “clear policies, education, and training” formula. Tangri and Hayes provide an exceptional critique of current theories, which alone may be worth the price of the book. They easily incorporate into their arguments the complexities associated with defining harassment, rather than apologizing for them, as seemed to be the case in many of the other chapters. Pryor and Whalen provide a review of the typologies of sexual harassment, but this material was covered in greater detail in many of the preceding chapters. Dansky and Kilpatrick describe the work-related, psychological, and physical effects of sexual harassment. This chapter is dry and would have benefitted considerably from the addition of a couple of short case histories to illustrate the grievous consequences sometimes associated with these experiences and to put a human face back into the picture. Grundmann et al.’s chapter on prevention was a little disappointing. It seemed to be an opportunity to provide some creative suggestions in response to the critical issues raised in the preceding chapters. The authors seem to waive this opportunity, arguing first for more adequate causal models. (Fortunately, some of the other contributors include their own suggestions for prevention). There is also a section on “steps” women can take to reduce the likelihood that they will be sexually harassed, including “not expecting to be given any special treatment on the job based on their gender . . . not avoiding duties of the job that might be uncomfortable or dirty” (p. 181), which is somewhat disturbing. Gutek provides an excellent discussion of guidelines useful to organizations wishing to establish sexual harassment policies and to promote a harassment-free
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
QC: FTK PP016-291184-86
January 25, 2001
Book Reviews
11:30
Style file version Nov. 19th, 1999
223
work environment. She also offers sensible advice about problems that may arise in response to the increasing “rights” of complainants and “responsibilities” of organizations to prevent harassment. The following chapter by McQueen addressing investigations of sexual harassment allegations suffers by comparison, although some useful suggestions are provided in the latter half of the chapter. Adams’ historical analysis of Black women’s experiences of sexual harassment is fascinating. She argues convincingly how “(t)he perception of black women as available and sexually promiscuous combines with sexist views of women in general to result in harassment that includes both racial and sexual connotations” (p. 220). Paludi covers the issue of sexual harassment in schools, although her focus on the behavior of teachers and professors is incongruent with her review (and the review of other contributors) that most girls and women experience sexual harassment from peers (or coworkers). Niebuhr examines the efforts taken in the military to measure and control the occurrence of sexual harassment and comments on the much-needed “wake-up call” the military has faced in this regard. In the final chapter, Barak argues that it would be “completely erroneous” and “impossible” to compare sexual harassment data cross-culturally, but he does so anyway, and ultimately provides readers with a useful context within which to understand much of the material presented in this book. A broader historical analysis of sexual harassment is missing from this book, although Conte’s legal analysis and Adams’ historical analysis of the experiences of Black women remedy this situation to some extent. There is little review of the reams of perception research that has been published in the last decade despite the contributors’ documentation of research indicating that subjective criteria are required to determine whether harassment has occurred. There is also no coverage of the recent interest in child-to-child sexual harassment. Further, these chapters could have used some serious additional editing: there is an unfortunate degree of redundancy beyond that which is required for adequate coverage of the individual topics. In sum, this edited volume addresses an important, challenging, and timely subject in a scholarly manner. It is a book that will be of use to a wide academic audience and certainly will not disappoint those seeking a broad education on this critical social issue and practical advice on tackling sexual harassment in schools or the workplace. REFERENCES Petrocelli, W., and Repa, B. K. (1998). Sexual Harassment on the Job, 3rd Ed., Nolo Press, Berkeley, CA. Sandler, B. R., Shoop, R. J., and Bratt, C. S. (eds.). (1996). Sexual Harassment on Campus: A Guide for Administrators, Faculty, and Students, Allyn and Bacon, Needham Heights, MA. Wagner, E. J. (1992). Sexual Harassment in the Workplace: How to Prevent, Investigate, and Resolve Problems in Your Organization, Amacom, New York.
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
QC: FTK PP016-291184-86
224
January 25, 2001
11:30
Style file version Nov. 19th, 1999
Book Reviews
Woman Abuse on Campus: Results From the Canadian National Survey. By Walter S. DeKeseredy and Martin D. Schwartz. Sage Publications, Thousand Oaks, California, 1998, 212 pp., $59.95 (hardback), $24.95 (paperback). Reviewed by Ilsa L. Lottes, Ph.D.2
In this volume, DeKeseredy and Schwartz try to do for Canada what Koss and her colleagues have done for the United States, namely document the extent of dating violence, using a nationally representative sample of college and university students. In addition, these researchers hoped that their book, based on the Canadian National Survey (CNS), would uncover some of the major sources of male-to-female victimization, identify key risk factors of woman abuse in Canada, allow tests of theories of woman abuse in dating, permit comparisons of abuse data with the United States, and inform abuse policy and prevention programs. In the CNS, any intentional physical, sexual, or psychological assault on a female dating partner is defined as abuse. The topics of the book’s five chapters are the historical, social, and political context of the CNS on woman abuse in dating; the incidence and prevalence of woman abuse in courtship; meanings and motives for the use of violence; risk factors for dating abuse; and progressive policy proposals. This book is well-written and well-organized and describes many limitations of the research they report. The methodology is sound—a national sample of 1835 women and 1307 men was obtained with the help of generous federal funding, and the CNS includes versions of established measures of abuse, such as the Conflict Tactics Scale and the Sexual Experiences Survey. The CNS attempts to provide insights into the context and motives of courtship violence in addition to the more commonly reported and more easily obtainable incidence and prevalence data. With respect to this last goal, their efforts were limited. What they did do was obtain data on the motives for female-to-male physical violence and debunk a common myth about the reasons for this violence. DeKeseredy and Schwartz found that much of this violence was in self-defense and thus conclude that such violence should not be labeled as “physical abuse of men.” In my view, there is much more to a context analysis than simply asking women to choose among three categories of motives for physical violence against their male partner. For an adequate context analysis, many more aspects of the situation would need to be assessed and I would probably use the abusive incident as the unit of analysis. In the ideal study, the couple in which at least one of the partners was abusive would be the analysis unit. This book also summarizes much previous work in the areas of the sexual and physical abuse of women. Thus, this book would be appropriate to use in an introductory university course in women’s studies or human sexuality as well as 2 Department of Sociology and Anthropology, University of Maryland Baltimore County, 1000 Hilltop
Circle, Baltimore, Maryland 21250.
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
Book Reviews
QC: FTK PP016-291184-86
January 25, 2001
11:30
Style file version Nov. 19th, 1999
225
for courses in social work, sexual health, and the family; however, the professional who is familiar with the woman abuse literature will not find much new with respect to theoretical approaches or program initiatives. The main new finding emphasizes that Canada and the United States have similar rates of abuse, and that despite the lower rates of some types of violence in Canada, this has not led to lower rates of courtship violence. One problem I found in the book is its almost total focus on women as the victims and men as the perpetrators. This tends to support the unsupported universal simplistic dichotomy that women are always “good and innocent victims” and men are the “bad and dominant” perpetrators. Certainly, there is much evidence to support the view that women do suffer greatly because of physical, sexual, and psychological abuse by men. But that statement describes only a part of the abuse picture. First, not all men are abusers and, as DeKeseredy and Schwartz point out, the abuse prevention programs that seem to be most effective are ones that do not “indiscriminately blame all men” (p. 143). I have also found that I am more successful in getting men to be sympathetic to women’s issues when I present a more balanced picture of these topics. Second, there should be more of an acknowledgment that men are also victims of abuse, and that this book does not cover that subject. In the United States, for example, I was surprised to learn that 25% of sexual assault victims that seek help at the county Sexual Assault Center are male. Third, although psychological assault was included in the CNS definition of abuse, this type of abuse was not examined for men and was not discussed as thoroughly as the sexual and physical abuse of women. Men certainly are victims of psychological abuse by women, and this was not mentioned. Some even attribute such psychological abuse as a factor contributing to the physical abuse of women. Finally, there was no acknowledgment that dating violence occurs in gay and lesbian relationships, and that this topic has received very little attention by researchers. Another limitation of the book is its incompleteness with respect to its test of theories. I was surprised that the works of Malamuth (Malamuth, 1986; Malamuth et al., 1991) and Craig (1990) were omitted, as these scholars have presented models of sexual aggression and coercion. Thus, lack of a social conscience, lack of empathy, harsh childhood physical punishment, hostile home environment, and acceptance of violence in general were not considered as characteristics of men more likely to abuse women. The authors do stress the importance of looking for additional factors relating to the abuse of women, for their regression models of the three types of abuse account for only about 20% of the variance with patriarchal attitudes/beliefs and male peer support variables as the significant predictors. In summary, I found this book to be appropriate for use in introductory college courses. I would use it in my sociology of sexuality course and also in my cross-cultural sexuality course and supplement it with other readings and information sources. DeKeseredy and Schwartz include much valuable information about women abuse that has wide application. The methodology was sound with respect
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
QC: FTK PP016-291184-86
January 25, 2001
226
11:30
Style file version Nov. 19th, 1999
Book Reviews
to sampling, questionnaire, scale construction, and data analysis and reporting. A good portion of the theoretical perspectives are also presented along with a review of what educational institutions can do to reduce the incidence and prevalence of the abuse of women by men. REFERENCES Craig, M. E. (1990). Coercive sexuality in dating relationships: A situational model. Clin. Psychol. Rev. 10: 395–423. Malamuth, N. M. (1986). Predictors of naturalistic sexual aggression. J. Pers. Soc. Psychol. 50: 953– 962. Malamuth, N. M., Sockloskie, R. J., Koss, M. P., and Tanaka, J. S. (1991). Characteristics of aggressors against women: Testing a model using a national sample of college students. J. Consult. Clin. Psychol. 59: 670–681.
Violence Against Women: The Bloody Footprints. Edited by Pauline B. Bart and Eileen G. Moran. Sage Publications, Newbury Park, California, 1992, 294 pp., $46.00 (hardback), $23.95 (paperback). Reviewed by Lettie L. Lockhart, Ph.D.3
This edited volume began as a collection of papers in a special issue of Gender & Society published in 1989. The authors examined violence against women from a feminist perspective. They pointedly presented a cultural, sociopolitical, and structural context of violence inflicted upon females of all ages in our society within historical and contemporary time frames. The authors collectively challenged prevailing myths that men who use violence to control women are “sick individuals” and are psychologically incapable of discerning the consequences of their misogynist subjugation of women. Furthermore, they suggested that because of the sociopolitical subordinate status of females in our society, males generally are not often conscious of the “male plot” to control and keep women in “their place.” These authors poignantly asserted that females’ subjugation by males in our society is historically rooted and contemporarily reinforced in the sociopolitical fibers of all our social institutions (e.g., political, economic, educational, religious, familial, medical, and legal) and are ingrained in our social and cultural values, norms, and mores. In Part I, entitled “Types of Violence Women Experience,” the authors discussed violence against women, ranging from pornographic phone calls, sexual harassment in the workplace, incest, rape, physical and emotional battering, and femicide. Regardless of the type of violence being discussed, the authors present very persuasive arguments that men use violence as a means to terrorize, control, and constrain the lives of women and girls while empowering themselves. 3 School
of Social Work, 303 Tucker Hall, University of Georgia, Athens, Georgia 30602.
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
QC: FTK PP016-291184-86
January 25, 2001
226
11:30
Style file version Nov. 19th, 1999
Book Reviews
to sampling, questionnaire, scale construction, and data analysis and reporting. A good portion of the theoretical perspectives are also presented along with a review of what educational institutions can do to reduce the incidence and prevalence of the abuse of women by men. REFERENCES Craig, M. E. (1990). Coercive sexuality in dating relationships: A situational model. Clin. Psychol. Rev. 10: 395–423. Malamuth, N. M. (1986). Predictors of naturalistic sexual aggression. J. Pers. Soc. Psychol. 50: 953– 962. Malamuth, N. M., Sockloskie, R. J., Koss, M. P., and Tanaka, J. S. (1991). Characteristics of aggressors against women: Testing a model using a national sample of college students. J. Consult. Clin. Psychol. 59: 670–681.
Violence Against Women: The Bloody Footprints. Edited by Pauline B. Bart and Eileen G. Moran. Sage Publications, Newbury Park, California, 1992, 294 pp., $46.00 (hardback), $23.95 (paperback). Reviewed by Lettie L. Lockhart, Ph.D.3
This edited volume began as a collection of papers in a special issue of Gender & Society published in 1989. The authors examined violence against women from a feminist perspective. They pointedly presented a cultural, sociopolitical, and structural context of violence inflicted upon females of all ages in our society within historical and contemporary time frames. The authors collectively challenged prevailing myths that men who use violence to control women are “sick individuals” and are psychologically incapable of discerning the consequences of their misogynist subjugation of women. Furthermore, they suggested that because of the sociopolitical subordinate status of females in our society, males generally are not often conscious of the “male plot” to control and keep women in “their place.” These authors poignantly asserted that females’ subjugation by males in our society is historically rooted and contemporarily reinforced in the sociopolitical fibers of all our social institutions (e.g., political, economic, educational, religious, familial, medical, and legal) and are ingrained in our social and cultural values, norms, and mores. In Part I, entitled “Types of Violence Women Experience,” the authors discussed violence against women, ranging from pornographic phone calls, sexual harassment in the workplace, incest, rape, physical and emotional battering, and femicide. Regardless of the type of violence being discussed, the authors present very persuasive arguments that men use violence as a means to terrorize, control, and constrain the lives of women and girls while empowering themselves. 3 School
of Social Work, 303 Tucker Hall, University of Georgia, Athens, Georgia 30602.
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
Book Reviews
QC: FTK PP016-291184-86
January 25, 2001
11:30
Style file version Nov. 19th, 1999
227
Scully and Marolla, in their chapter “Riding the Bull at Gilley’s: Convicted Rapists Describe the Rewards of Rape,” discussed the cultural and structural factors that encourage rape and the rewards rapists receive from raping females. Based on interviews with convicted rapists (N = 114; 54% black and 46% white), they identified rewards (as these rapists themselves perceived) they received in raping women: (1) a sense of revenge or punishment—collective liability of all women, (2) rape was a “low risk act” because many women frequently fail to report rape, (3) rape was an “added bonus” in the commission of other criminal activities, (4) sex is a male’s entitlement and by raping a woman it gives rapists sexual access to what was not offered, (5) a sense of power over the woman because rape was seen as impersonal, and (6) a form of recreation and adventure (e.g., gang rape). These authors concluded that rape is a deliberate crime. They also claimed that sexual access emerged as the reward for black men who consciously chose to rape white women. They further stated that the majority of rapes in the United States today are interracial and the rate of black on white rape significantly exceeded the rate of white on black rape: “. . . the status of black women in the United States today is relatively lower than the status of white women. . . . Thus, the desire for sexual access to higher status, unavailable women, an important function in [black/white] rape, does not motivate white men to rape black women” (p. 37). The most insidious myth about rape that has been promoted and reinforced by these authors is that rape, in general, but more specifically of white women, is most likely committed by a black man. The myth of the black rapist renders people, especially white females, oblivious to the realities of rape and the fact that 90% of all rapes are intraracial rather than interracial (Davis, 1990), and that 75% of rape victims know their attacker. Another fact that these two authors failed to recognize is that more white males have access to black females than black males have access to white women because there are more white service technicians and laborers going into black neighborhoods and homes than vice versa. Nevertheless, the average white woman maintains a greater suspicion of black men than of white men as potential rapists. These distorted social attitudes, which are racist in their very nature, constitute a major obstacle to the development of a crosscultural movement capable of major victories in the struggles against rape. As Davis (1990) points out, many feminists often ignore the interconnectedness of racism and sexism in general among the white women who pioneered the women’s movements. Thus, many antirape activists have historically failed to develop an understanding of the degree to which rape and the racist use of fraudulent rape charges are inseparable. That is, rape of black women by white men has historically constituted a political weapon of terror whereas the weapon directed at black men has been frame-up rape charges. Scully and Marolla, as well as other antirape activists, must understand the complex structures reflecting the interconnectedness of race, gender, and class oppression that characterize our power structure, if we hope to develop strategies that will allow us to purge our society of oppressive misogynist violence. We must believe that the sociopolitical rhetoric presented in
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
228
QC: FTK PP016-291184-86
January 25, 2001
11:30
Style file version Nov. 19th, 1999
Book Reviews
this volume is about the oppression of women of all racial, ethnic, and economic groups, thus directing us to develop more inclusive rather than exclusive platforms to fight societal oppression of women. In Part II, entitled “Structural Supports for Violence Against Women,” the authors used a feminist perspective to emphasize the sociopolitical, cultural, and structural factors that encourage and reinforce males’ violent subjugation of females. These authors discussed how violence inflicted upon women is structurally and institutionally supported in religious, medical, educational, legal, and social organizations, such as fraternities (see McCormick, 1995; Sanday, 1990). In a chapter on the sexual politics of black womanhood, Collins “dared” to examine the sexual politics of black womanhood and the complexity of sex, race, and class in maintaining social inequity in our society. She discussed how the sexual exploitation of black women—from slavery to the present—helped to maintain the privilege of white men relative to all women, especially women of color—a chapter that Scully and Marolla should have read prior to their attempts of interpreting their study results. Young’s discussion of Maria Goretti epitomizes the value placed on female virginity and the church’s idealization of virgin martyrs as role models for girls and women in the Catholic church. The quantitative investigation by Martin and Hummer provides rich and interesting data about fraternities and rape on college campuses. They suggested that college fraternities create a sociocultural context in which the use of coercion in sexual relations with women is normative and occurs in settings where mechanisms to control these violent aggressive behaviors toward women are minimal or absent. This chapter could have been enhanced if the authors had discussed the fact that all women, in particular college women, need to always exercise judgement. These authors unintentionally portrayed women as “hair brain victims,” without the capability of logical reasoning and deduction. Women must be taught to understand these sociocultural elements of violence against women in order to make informed decisions about inherent risks of these situations and of our actions. Having said this, I agree with all the authors that women should not be blamed for these aggressive acts inflicted upon them, but that when we do not examine our behaviors to risky situations we place women at greater risk of harm. Thus, we must also empower women with an understanding of the structural supports of these dangers, such as rape, and equip women to logically deduce level of risk and plausible strategies to reduce their risk of harm. Over the past 25 years, feminists have developed organizations that have successfully challenged social institutions for failing to protect women in our society. Battered women’s advocates have loudly criticized police departments, court systems, and laws and have forced legislatures and the criminal justice systems at the local and national levels to respond to this endemic social problem. In Part III, entitled “Politics of Institutional Responses to Violence Against Women,” the authors discussed and scrutinized societal institutional responses (e.g., criminal justice system, laws, and feminist-originated services) to violence against women.
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
Book Reviews
QC: FTK PP016-291184-86
January 25, 2001
11:30
Style file version Nov. 19th, 1999
229
Particular emphasis was given to the limitations of using piecemeal approaches that address violence toward women in a society in which gender inequities are persistent and are endemic. Feminist-originated services and programs have themselves been institutionalized (e.g., applying single-issue strategies to safeguard their budgets) and are often separated from their feminist roots. These authors collectively recognized that feminist analyses and political organizing increased services to survivors of rape and battery, but suggested that the effectiveness of these services are seriously undermined by the persistence of patriarchal biases inherent in social service and criminal justice systems. Even with these assertions, these authors continue to assert the value of increased public funded services for battered women and victims of sexual assault. In Stanko’s chapter, entitled “Ordinary Fears: Women, Violence and Personal Safety,” it was argued that sexual danger is an everyday reality for females, and women routinely assume that they must always be on their guard against assaults from men. She further stated that “women’s fear of crime is a fear of men.” Such a statement, in which there were many in this volume, reinforces the myth that only men are violent toward women. Not all women have a fear of men; if this was the case, heterosexual women would not be engaged in intimate relationships with these men. In general, it may be true that women may be fearful of men, given the sociopolitical climate of gender relationships in our society. But, on a more personal level, I would question Stanko’s assertion. Leaping generalizations such as the one being referenced continue to reinforce the stereotypical notion that all men are dangerous, and all women are victims. This is not the case. Research on lesbian relationships (Causby et al., 1995; Lockhart et al., 1995; Renzetti, 1992) suggest that women in same-gender relationships also use violent acts to control their partners. Stanko and other authors failed to acknowledge their own message that violence against women is a control mechanism. If we assert and accept a control theorem of violence against women, we must recognize that violence against women can exist between same-gender individuals as well as between heterosexual individuals. Promoting a more holistic understanding of violence against women may prompt more males to examine their attitudes and behaviors toward women. We must join forces to strategically and effectively address this major social problem rather than pitting the genders against each other. In Part IV, “Research Implications of Experiencing and Studying Violence against Women,” the authors promote feminist theories and research methodologies as a new paradigm for examining the experiences of women and men. They suggest viewing our social world from a different lens. The authors suggested that future research should incorporate both qualitative and quantitative research methodologies in order to produce richer data with greater validity and utility. Further, it is suggested that study samples should include victims, perpetrators, and service providers in the attempt to understand the experiences of all players in this social problem and devise effective strategies.
P1: FYJ/GBQ
P2: FPX/FJQ
Archives of Sexual Behavior [asb]
QC: FTK PP016-291184-86
January 25, 2001
230
11:30
Style file version Nov. 19th, 1999
Book Reviews
REFERENCES Causby, V., Lockhart, L. L., White, B. W., and Greene, K. (1995). Fusion and conflict resolution in lesbian relationships. Journal of Gay and Lesbian Social Services, 3: 67–82. Davis, A. (1990). Women, Culture and Politics, Vintage Books, New York. Lockhart, L. L., White, B. W., Causby, V., and Isaac, A. (1995). Letting out the secret: Violence in lesbian relationships. J. Interpers. Viol. 9: 469–492. McCormick, N. B. (1995). Review of Fraternity Gang Rape: Sex, Brotherhood, and Privilege on Campus. Arch. Sex. Behav. 24: 355–358. Renzetti, C. (1992). Violent Betrayal: Partner Abuse in Lesbian Relationships, Sage Publications, Newbury Park, CA. Sanday, P. R. (1990). Fraternity Gang Rape: Sex, Brotherhood, and Privilege on Campus, New York University Press, New York.
