EDITORIAL
1
Editorial
T
his is a very special edition of Creativity and Innovation Management. It marks the transfer of the editorship of the journal to a new team, Professor Olaf Fisscher and Dr Petra de Weerd-Nederhof at the University of Twente, The Netherlands. With the editorial and production team at Blackwell Publishing, they will be taking the journal to a new stage in its development. The relationship with Blackwell has been a long and fruitful one, and the contact with Olaf and Petra has been developed within the context of the European creativity community. It is therefore fitting that the theme of this issue of the journal is ‘Interacting for Creativity’. In our first article, Cathleen McGrath, Charles Vance and Edmund Gray report on a fascinating study amongst software executives, which explores how these individuals manage advice sharing networks. A major finding was the importance of establishing trust in advice sharing relationships. The theme of information sharing is developed in the article from Jozée Lapierre and Vincent-Pierre Giroux. Their work examines the factors important in a work climate that foster creativity, and they suggest work atmosphere, vertical collaboration, autonomy/freedom, respect, alignment and lateral collaboration are key dimensions. Kristensson and Norlander explore the context in which creative interaction takes place. They describe the impact of information technology on the creative performance of groups and individuals. They suggest that group communication support systems (GCSS) can assist in the exchange and development of ideas providing there is a fully understood and well-prepared problem, attained or strengthened through face-to-face interaction. The importance of problem definition and good quality interactions is at the heart of the Synectics system. We are especially pleased that Vincent Nolan, a distinguished Synectics practitioner has written an article about his experience with the approach. His comment about the need for greater dialogue between academics and practitioners is one we can only echo, and one of the compelling reasons for © Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
the origination of Creativity and Innovation Management. Tudor Rickards responds to Vincent’s observations with comments from his own experience of the academic and practitioner communities. Establishing an organizational structure which will encourage innovation, has concerned academics and practioners alike for many years. In their article, Dieter Gebert, Sabine Boerner and Ralf Lawehr offer intriguing observations concerning the relationship between employee empowerment (characterized by the authors as situation control) and the stimulation of innovation. They suggest that without careful management empowerment, because of the increased need for co-ordination and integration of activities, can lead to a decline in organizational innovativeness. The relationship between human resource management and the support of innovation is the subject of our final article. Rosalind Searle and Kirstie Ball examine the coherence of organisations’ utilization of human resource (HR) recruitment, training and performance management policies to support and enhance firm’s innovation performance. Their work suggests that inconsistency in reward policies is one source of resistance which blocks the generation of new ideas, and their implementation throughout the organization. The breadth of subjects covered in this issue indicates the diversity of interest in the field of creativity and innovation. As we hand over the editorship of the journal to Olaf, Petra and their team we are confident that they will continue to develop the depth and range of contributions. We would like to extend warm personal thanks to our contributors, readers and our colleagues at Blackwell Publishing. We will continue to take a close interest in the journal, and wish the new team well in their exciting adventure. Tudor Rickards Susan Moger Founding Editors Creativity and Innovation Management Manchester February 2003
Volume 12
Number 1
March 2003
2
CREATIVITY AND INNOVATION MANAGEMENT
With a Little Help from Their Friends: Exploring the Advice Networks of Software Entrepreneurs Cathleen A. McGrath, Charles M. Vance and Edmund R. Gray This field interview study examined patterns and content of advice sharing networks among 20 software executives to provide a clearer understanding of how advice relationships are established, the types of advice that are shared and the role that relationships play in the support of information exchange and diffusion. Most advice relationships were formed from strong tie relationships, while systematic differences were found among the types of advice sought from advice relationships resulting from strong ties, business ties and weak ties. The preference of software executives for rich communication media supports the importance of establishing trust in advice sharing relationships.
T
oday’s knowledge economy requires that successful entrepreneurs, executives and other knowledge professionals within organizations have access to a wide variety of people, information and advice, both inside and outside their organization (Dixon, 2000; Leonard-Barton, 1995; Mintzberg, 1973; Nohria, 1992; von Hippel, 1987). The collection of formal and informal advice contacts that managers establish inside and outside their organizations provides a network of information, knowledge and inspiration that may be drawn upon to advance the technical, innovative and business performance of their organizations. Formal connections can be established internally through formal work roles and externally through association with individuals who are members of customer, supplier or joint-venture organizations. In addition, connections can be established informally through personal contacts maintained both inside and outside the organization. This study presents a close look at the advice networks maintained by software managers whose companies are spin-outs of Carnegie Mellon University. These software managers provide an interesting look at advice sharing in an industry that is both rapidly innovating and knowledge intensive, with close connections to academic research.
Volume 12
Number 1
March 2003
A network of connections, both internal and external, is especially critical for managers in new, small businesses such as the software companies studied here. The structure and quality of individual managers’ networks can have significant influence on the overall performance of their organizations (Dubini and Aldrich, 1991). Individual networks maintained by technical managers provide them with ready access to the information and assistance necessary for rapid response to opportunities and difficulties that might arise in their business. Also, in entrepreneurial organizations, large and small, advice sharing represents a vital source of new ideas behind innovation and change and leading to competitive advantage (Carter, 1989; Saxenian, 1994; von Hippel, 1987). Advice sharing contacts are particularly important to entrepreneurs and other knowledge professionals in new software ventures that compete in a very uncertain, rapidly changing competitive environment. For example, Allen’s (1984) study of communication among innovating organizations found that project engineers spent about one-third of their total communication time with people outside their laboratory. He found that outside advice contacts were important sources for ideas and solutions to problems. In fact, informal advice ties have been shown
© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
ADVICE NETWORKS OF SOFTWARE ENTREPRENEURS
to facilitate the innovative process even when the necessary information is available from formal sources (Harmon et al. 1997). Advice sharing across organizational boundaries is common and even essential for new and small enterprises that lack a larger, developed internal structure and set of resources that provide information to aid in decision-making. We define advice as a subset of general knowledge generation in which individuals seek out or give specific assistance. Our definition is consistent with the accepted definition of knowledge as part of a hierarchy from data – unrelated facts, information – facts connected to each other, and knowledge – information made meaningful by individuals (see, for example, Davenport & Prusak, 1998; Dixon, 2000). When individuals seek advice, they use interpersonal connections to fill gaps in their knowledge or experience to resolve problems or take advantage of opportunities more quickly. Information obtained directly from another person in the form of advice allows managers to react quickly to generate innovative solutions to problems or opportunities that emerge. Advice sharing depends heavily on social interaction. For instance, an individual might learn about business-plan development by listening to a lecture or by participating in casual conversation at a conference (formal and informal forms of general knowledge generation), but to obtain advice about a particular business plan or solution to a specific problem he or she must seek out some knowledgeable other. For advice to be shared successfully, individuals must seek such information from others who are willing and able to give them advice. When individuals share advice, they become vulnerable to each other. Mayer et al. (1995) describe three components of trustworthiness as competence, benevolence and integrity. Each of these components plays a role in the advice sharing relationship. Those seeking advice count on advice contacts being competent enough to give useful advice, and having benevolence and integrity to not exploit the advice seeker’s vulnerability or purposely provide inadequate advice. Similarly, those giving advice depend on the ability of advice contacts to use the advice they are giving without requiring an overwhelming amount of assistance, and acting with benevolence and integrity in not purposely exploiting the advice giver’s willingness to share information and insight. The development and nature of advice sharing relationships can be critical to organizations’ efforts to optimize innovation for long-term survival and competitiveness. This
© Blackwell Publishing Ltd, 2003
3
study closely examines patterns and content of advice sharing relationships among a sample of top managers in software companies to provide a clearer understanding of how advice relationships are established, the types of advice that are shared and the role that relationships play in the support of information exchange and diffusion. In this study we classify advice connections based on the source of their generation as strong ties, business ties or weak ties. We argue that systematic differences in the basis of tieformation lead to differences in the type of advice that is shared. Previous literature provides evidence for the importance of weak ties when novel information is needed (Burt, 1992; Granovetter, 1973), strong ties (Krackhardt, 1992) when the trustworthiness of individuals is most important and a portfolio of weak and strong ties (Uzzi, 1997) when strategic management of business relationships is most important. We extend that work by considering the source of advice relationship formation and the type advice conveyed through the relationship as well as the nature of relationship; weak, strong or business; to develop a more full explanation of advice sharing relationships and their contribution to organizations’ innovative performance.
Establishing Ties Clearly, maintaining a network of advice contacts is important to the performance of technical managers. It is also important to understand where advice relationships are likely to form. The social-network literature provides two perspectives on where a technical manager is likely to find his or her advice contacts. In his classic article, ‘The Strength of Weak Ties’, Grannovetter (1973) argues that people with whom individuals maintain weak ties, spending little time, having infrequent contact and little investment in the relationship, are more important for transmitting new information and ideas than people with whom an individual maintains strong ties. Burt (1992) elaborates on the benefits of weak ties with his formalization of the benefits to individuals of spanning structural holes. Individuals span structural holes by cultivating relationships with others who are not themselves connected. These relationships are beneficial because they provide individuals with a rich source of novel, non-redundant information. On the other hand, Krackhardt (1992) demonstrates the importance of strong ties, which he refers to as ‘philos relationships’
Volume 12
Number 1
March 2003
4
CREATIVITY AND INNOVATION MANAGEMENT
for work-related outcomes when trust is an important factor. Strong tie relationships develop when individuals have the opportunity for repeated contact over a long period of time. Also, strong ties may emerge when individuals are not only connected to each other but also share a common connection to a third person. Krackhardt (1999) refers to these connections as Simmelian ties. Simmelian ties are a form of strong ties because the common connection to a third party tends to constrain and strengthen the relationship between the individuals in a Simmelian-tie triad. A third perspective on advice relationships takes a strategic approach, suggesting that advice connections emerge from the business ties that bind organizations. For example, studies of biotechnology alliances (Powell, Koput and Smith-Doerr, 1996) assume that informal relationships follow from the formal alliance relationships established by organizations. The social network literature would suggest that if novel information is most important to managers involved in advice sharing, then they would seek advice from individuals with whom they are weakly tied. At the same time, if trust is important to managers involved in advice sharing, then they would seek advice from individuals with whom they have strong ties. Finally, if information and learning results from the business process, then software managers will draw individuals with whom they have business ties for advice. In this study, we are able to consider the relative contribution of weak ties, strong ties and business ties that make up individuals’ advice networks. We find that most advice relationships form from strong tie relationships. Furthermore, we find systematic differences among the types of advice sought from advice relationships that result from strong ties, business ties and weak ties. Finally we find that software managers’ choices of rich communication media support the importance of the establishment of trust in advice sharing relationships.
Sample Methodology All of the software executives (most with the title of founder, chief executive officer, or chief technical officer) interviewed for this study belonged to companies that were located in or around Pittsburgh, Pennsylvania and had some past or present connection to Carnegie Mellon University. At the time of the study, software accounted for 60 per cent of all technology-based Carnegie Mellon-related firms in the Pittsburgh area (Carnegie Mellon Center for Economic Development, 1997). A
Volume 12
Number 1
March 2003
total of 36 executives belonged to organizations that matched the criteria to be included in the study. Twenty of those executives agreed to be interviewed and answered openended questions about their perspectives on sharing information, and of those 20, 17 individuals answered detailed questions about their own advice sharing network. The Pittsburgh High Technology Council provided information about the original 36 organizations’ locations, sales, employment and ages. The median age of companies in the study population was eight years. The oldest company in the study was 34 years while three companies were only one year old. Companies less than one year old were not included in the study. The average number of employees in the sample of companies was 57 while the median was 16. There was no statistically significant difference in age or size between companies that did and did not participate in the study.
Interviews Respondents were asked to name up to fifteen people – inside or outside their organization – with whom they regularly shared workrelated advice. They were told that the people they named could be anyone including coworkers, peers, customers, suppliers, competitors or academics, located anywhere in the world. The survey instrument was a modified version of an instrument developed by Schott (1994) to study patterns of interaction among academic scientists in the field of engineering. We presented respondents with categories of advice that would most likely be sought by managers of small, innovative software companies. These categories of advice were constructed using preliminary interviews with research software scientists to identify the most salient forms of advice related to software development. We focused on advice that related to the technical and innovative performance of individuals within the software start-ups. The categories included technical and business topics such as advice related to software tools, referrals to the best people to contact to answer a specific question, help with finishing deliverable products or services, suggestions for new deliverable products or services and general business advice about issues that arise in the day-to-day process of running a software company. The average number of advice network contacts identified by participants was twelve, and ranged from seven to 15. Information was collected from the participants regarding the formation and nature of a total of 202 advicecontact relationships. Although research has
© Blackwell Publishing Ltd, 2003
ADVICE NETWORKS OF SOFTWARE ENTREPRENEURS
shown that individuals cannot always accurately record specific instances in which they communicate with others (Bernard, Kilworth, and Sailer, 1980, 1984), they can recall overall patterns of communication effectively (Freeman, Romney, and Freeman 1987). For the purposes of this study, retrospective accounts of advice sharing relationships provided a good measure of advice sharing contacts.
Advice Relationship Formation Software executives reported that advice contacts were more likely to emerge from strong tie relationships formed through studying at the same school or working in the same company (although only 18 per cent of all advice contacts identified currently worked at the same company as respondents), and less likely to emerge from business relationships such as customer or supplier relationships or weak tie relationships such as contact at conferences or organization membership. Table 1 shows that work and school were the two most prevalent places to form advice sharing contacts. Together they accounted for nearly half of all advice contacts. School and work ties can be considered strong ties because they are formed in the context of long-term interaction with a shared history and experience that allows for the individuals to evaluate each other’s trustworthiness. Slightly less than
5
half (47 per cent) of the advice contacts who respondents met at work still worked in the same organizations as the respondents. (Less than 20 per cent of all advice contacts identified by respondents were currently working in the same organization as the respondent.) The next most likely place to establish advice relationships was through mutual contacts. Again, using a mutual connection embeds the advice contact in the way that Krackhardt (1999) describes Simmelian ties. These three strongtie connections together account for 60 per cent of the advice relationships reported by software managers. Business relationships emerging from contacts who were part of supplier, customer or joint-project organizations were the next most likely source of advice contacts. Combined, they comprised 18 per cent of the respondents’ advice contacts. Weak ties emerging from contacts made at conferences and organization co-membership contributed the smallest proportion (7.5 per cent) of advice contacts. The fact that respondents met over half of their advice contacts by working or studying together or through mutual connections suggests once again the importance of the opportunity to develop repeated interactions over a long time period to develop advice sharing relationships. Both school and work provide opportunities for repeated contact and observation that would allow a trusting relationship to form. Sharing connections with a mutual third person also allows a trusting relationship to form.
Table 1. Advice network formation Where respondents met their advice contacts (%)
Strong ties Work School Mutual contact Business ties Respondent’s company was customer Respondent’s company was supplier Joint project Weak ties Attending conference Member of same organization Other way Total
© Blackwell Publishing Ltd, 2003
Reciprocal (%)
Gives advice only (%)
Receives advice only (%)
29.2 19.3 11.9
69 67 67
17 10 4
14 23 29
8.4 5.4 4.0
91 47 87
0 12 13
9 41 0
5.0 2.5 14.4
80 20 58
20 0 21
0 80 21
100.0
66
13
21
Volume 12
Number 1
March 2003
6
CREATIVITY AND INNOVATION MANAGEMENT
Table 2. Communication media for advice sharing Media
Face-to-face conversation Telephone E-mail Fax Post Publication Lecture Web Usenet
All contacts (%) n = 202
Local contacts (%) n = 119
Distant contacts (%) n = 83
74.3 65.3 52.0 13.9 5.0 5.0 4.0 3.5 2.0
80.7 63.0 51.3 10.9 4.2 3.4 5.0 3.4 0.0
65.1 68.7 53.0 18.1 6.0 7.2 2.4 3.6 4.8
Communication Media Our sample of software executives demonstrated a heavy reliance on face-to-face and telephone communication for advice sharing. Even among technologically savvy users, e-mail was less preferred than more personal and ‘cue rich’, real-time communication patterns that may include more non-verbal forms of communication that promote communication clarity and message confidence. In fact, executives’ preferences for communication media closely mirrored Daft and Lengel’s (1986) continuum of media richness, for most to least rich. Seeking advice may involve incremental questions and answers that are more easily facilitated in real-time interactions. Also, since advice sharing is heavily dependent on trust, face-to-face and telephone conversations that encompass a wider range or richness of communication forms (e.g. tone of voice, inflections, facial expressions) are more likely to allow the establishment of trust for any single advice request. The more casual, implicit communication afforded by multiple reinforcing and redundant cues of face-to-face and telephone interactions allows individuals the comfortable freedom to ask questions and respond less concisely, yet still maintain a clear understanding. Also, face-to-face communication allows for real-time instant correction to avoid misunderstanding, wasteful confusion, and possible more serious outcomes including miscues, conflict and damaged relationships. In addition, face-to-face advice sharing contexts can create a fertile setting and greater
Volume 12
Number 1
March 2003
Table 3. Using advice How software managers use advice from their personal advice networks (%) Providing business advice Providing advice contact referrals Suggesting new deliverables Using tools Setting up technical environment Completing deliverables Other use
60.2 50.0 43.8 33.0 30.1 34.7 9.1
opportunity for unexpected valuable learning and serendipity (Vance et al., 1991).
Using Advice Respondents were asked to consider the kinds of advice that they received from their contacts, and they were allowed to select as many categories as appropriate. Of the 176 advice relationships reported to give advice, the average number of types of advice given was three. Table 3 shows that the software executives obtained general business advice most often (60.2 per cent), followed by advice about who might be a useful contact (50.0 per cent). Advice about using tools, setting up the work environment and completing deliverables was obtained less often. It is interesting to observe that even within highly technical industries
© Blackwell Publishing Ltd, 2003
ADVICE NETWORKS OF SOFTWARE ENTREPRENEURS
the types of advice most frequently sought related to general business issues and not specific technical issues.
7
This finding is consistent with the origin of their advice-giving relationship through some other third party. Advice contacts made through work on joint projects were likely to provide advice used to complete projects. These results suggest that the source of advice relationship formation does influence the type of advice received from an individual. To explore the relationship between the source of contact formation and the content of advice shared, we considered the strong tie, weak tie and business tie formation discussed earlier. Table 4 shows that the type of advice received from contacts differed depending on the initial source of the advice relationship. Suggestions for new products are most likely to come from connections made through business ties, while help with software tools is most likely to come from advice connections made through strong ties. The only type of advice that respondents were more likely to seek inside their own organization was advice on completing deliverable products or services (61 per cent of all contacts within the
Connecting to Advice Figure 1 shows the pattern of the type of advice received by individuals from advice contacts made in the eight different contexts discussed above. While business advice makes up a significant component of advice received from all sources, some differences exist in the relative magnitude of types of advice received from different sources. Advice about using tools and setting up work environments is a larger part of the advice received from contacts made at work and school than advice received from other contacts. Again, having studied or worked together provided a shared background and competence that would make this pattern likely. Advice contacts made through mutual connections provided a large proportion of advice in the form of referrals. 35 30
Environment Tools Complete New Ideas Referral Biz Advice
25 20 15 10 5
on lC
ua M ut
ta ct Su pp lie r C us to m Jo er in tP ro je ct C on fe re nc M e em be rs hi p
ol ho
Sc
W
or
k
0
Figure 1. Distribution of type of advice respondent received by different advice contact sources Table 4. Type of advice received by strength of tie Type of advice respondent received by contact category (%) Tie type
Business
Referrals
Suggest**
Tools**
Work Env
Complete*
Strong Business Weak
64 60 62
52 55 46
40 70 31
40 24 0
23 33 8
36 42 8
* Chi-square p < 0.10. ** Chi-square p < 0.01.
© Blackwell Publishing Ltd, 2003
Volume 12
Number 1
March 2003
8
CREATIVITY AND INNOVATION MANAGEMENT
same organization provided this sort of advice while only 29 per cent of external advice contacts provided this advice). Finally, help completing projects is likely to come from business ties. There is no single category in which contacts made through weak ties are the most likely to provide advice, but weak-tie contacts are as likely as strong-tie and business-tie contacts to give advice on business issues, referrals and setting-up work environments.
Software Executives’ Decisions to Share Advice In order to fully understand how advicesharing relationships are established and maintained, the decision to seek advice must be considered in addition to the decision to give advice. Study participants answered open-ended questions about the potential costs and benefits of sharing information with others. They identified the potential costs of advice seeking as the time consumed, the potential of receiving bad advice and the vulnerability or loss of status associated with advice-seeking. Reported benefits of seeking advice included learning and the infusion of new ideas and perspectives. Reported costs for giving advice included concerns about giving away proprietary information and spending too much time helping people in other companies. Finally, reported benefits of giving advice were the establishment of goodwill, leads on new ideas and personal satisfaction. Interestingly, when answering an open-ended question about benefits and costs of sharing information, building reciprocity ‘credits’ that could be called upon in the future was not a benefit mentioned by respondents. However, two-thirds of the advice relationships reported were reciprocal. This suggests that the reciprocity may emerge more as a result of the social nature of advice sharing rather than the economic perspective of advice exchange. Two common themes emerged from the concerns raised by the software executives. First, sharing advice makes people more vulnerable to each other, and second, sharing advice requires a time commitment from both parties. When seeking advice, software executives may let other companies know what they do not know. Exposing ignorance may cause discomfort, or it may be as costly as harming one’s reputation within the venture capital community. Also, those seeking advice may receive cavalier advice. Those giving advice become vulnerable to the possibility of harming their own competitive position. For both those seeking and giving advice, trust becomes a critical issue in the advice sharing
Volume 12
Number 1
March 2003
relationship. Second, the success of an advicesharing interaction depends upon the competence of both the advice seeker and advice giver. Sharing advice can be costly for the advice-giver if the individual seeking advice is significantly less competent than the advicegiver expected. What seems to be a simple request for advice can become a taxing commitment if the advice seeker requires a great deal of assistance to incorporate the advice offered. On the other hand, receiving advice from someone who is not competent or aware of his or her level of competence may lead advice seekers down a fruitless or detrimental path. This is one way that the relational nature of advice sharing differs from the more general case of information as a public good. In the latter case, if the information seeker does not have the ability to use information – what Cohen and Levinthal (1990) refer to as absorptive capacity – his or her opportunity to benefit will be lost but with no cost to the generator of the information. In the former case of advice sharing, the advice-giver may bear some of the cost of the lack of knowledge of the adviceseeker. When advice sharing is embedded within a broader social context, such as a common connection to a university that projects a high standard of integrity and education, it may be easier to establish trust in both participant motives and abilities and therefore easier to establish advice sharing relationships. This common social context may also address the second concern of advice-seekers and givers by reducing the time it takes to share advice when advice contacts are repeated and embedded within a common culture and context.
Discussion This exploratory research reveals a fundamental tension between software executives’ need to trust advice contacts, which requires a considerable time and relationship investment, and their need to maintain a network of advice contacts that are diverse and unconnected. Our findings suggest that software executives who are looking for advice tend to call upon people they know well, and mostly from relationships formed through previous work settings, educational settings or mutual contacts. At the same time, research (Burt, 1992; Granovetter, 1973) has shown that weak ties are important sources of information and control that lead to innovation. While weak ties play an important role in information generation overall, when it comes to the specific practice of seeking and giving
© Blackwell Publishing Ltd, 2003
ADVICE NETWORKS OF SOFTWARE ENTREPRENEURS
advice, we find that strong ties play a more important role. Important implications emerge from this research for managers of innovative organizations. First, we see that advice relations are more likely to emerge from situations in which people have an extended period of interaction and the opportunity to observe each other’s work performance. The extended contact allows individuals to determine the trustworthiness in terms of competence, benevolence and integrity (Mayer, et al., 1995) of potential advice contacts. This trend highlights the potential contribution of the university as a trusted social-tie generator as well as an idea generator, particularly important for software innovative industries. Although the provision by universities of long-term education and training opportunities to executives associated with advice-sharing gatherings presents a significant cost to innovative companies, the cost may be offset by the access to valuable advice and information that is created through participation in these educational and subsequent work interactions. Future research should continue to examine this potentially important contribution of universities as facilitators of advice sharing networks, taking them far beyond their more traditional theoretical and basic research role. Finally, software executives in innovative companies can learn important lessons from this study. First, it demonstrates the importance of balancing the need to maintain advice networks, whose members are close and trusted contacts, with the need to maintain advice networks of diverse, unconnected contacts. Second, it highlights the variety of information that is exchanged within advice networks. Advice networks are an important source of information about business, technical and social issues. Understanding that advice spans a broad range of contexts may help software and other entrepreneurial executives to foster efficient and effective advice networks. Considering both the content and structure of advice networks allows software executives to pattern the relationships and advice sharing behaviour in ways that benefit their own performance as well as the innovative performance of their companies.
References Allen, T. (1984) Managing the Flow of Technology: Technology Transfer and the Dissemination of Technological Information within the R&D Organization. MIT Press, Cambridge, MA. Bernard, H., Kilworth, P. and Sailer, L. (1980) Informant accuracy in social network data, a compar-
© Blackwell Publishing Ltd, 2003
9
ison of clique-level structure in behavioral and cognitive data. Social Networks, 2, 191–218. Bernard, H., Kilworth, P. and Sailer, L. (1984) On the validity of retrospective data: the problem of information accuracy. Academy of Management Journal, 34, 205–214. Burt, R. (1992) Structural Holes: The Social Structure of Competition. Harvard University Press, Cambridge, MA. Carnegie Mellon Center for Economic Development (1997) A Partner in Strengthening Our Region Through Business Growth. Carter, A. (1989) Knowhow trading as economic exchange. Research Policy, 18, 155–163. Cohen, W. and Levinthal, D. (1990) Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly, 35, 128–152. Daft, R. and Lengel, R. (1986) Organizational information requirements, media richness and structural design. Management Science, 32, 554–571. Davenport, T. and Prusak, L. (1998) Working Knowledge: How Organizations Manage What They Know. Harvard Business School Press, Boston, MA. Dixon, N. (2000) Common Knowledge: How Companies Thrive by Sharing What They Know. Harvard Business School Press, Boston, MA. Dubini, P. and Aldrich, H. (1991) Personal and extended networks are central to the entrepreneurial process. Journal of Business Venturing, 6, 305–313. Freeman, L., Romney, A. and Freeman, S. (1987) ‘Cognitive structure and informant accuracy’, American Anthropologist, Vol. 89, pp. 310–325. Granovetter, M. (1973) The Strength of Weak Ties. American Journal of Sociology, 78, pp. 1360–1380. Harmon, B., Ardishvili, A., Cardozo, R., Elder, T., Leuthold, J., Parshall, J., Raghian, M. and Smith, D. (1997) Mapping the university technology transfer process. Journal of Business Venturing, 12, 423–434. Krackhardt, D. (1999) Simmelian ties. Super strong and sticky. In Kramer, R. and Neale, M. (eds), Power and influence in Organizations. Sage, Thousand Oaks, CA, pp. 39–66. Krackhardt, D. (1992) The strength of strong ties: The importance of philos in organizations. In Nohria, N. and Eccles, R. (eds), Networks and Organizations. Harvard Business School Press, Boston MA, pp. 240–261. Leonard-Barton, D. (1995) Wellsprings of Knowledge: Building and Sustaining the Sources of Innovation. Harvard Business School Press, Boston. Mayer, R. C., Davis, J. H. and Schoorman, F. D. (1985) An integration model of organizational trust. The Academy of Management Review, July. Mintzberg, H. (1973) The Nature of Managerial Work. Harper & Row, New York. Nohria, N. (1992) Information and search in the creation of new business ventures: the case of the 128 venture group. In Nohria, N. and Eccles, R. (eds), Networks and Organizations. Harvard Business School Press, Boston MA, pp. 240–261. Powell, W., Koput, K. and Smith-Doerr, L. (1996) Interorganizational collaboration and the locus of innovation: networks of learning in biotechnology. Administrative Science Quarterly, 41, 116–145.
Volume 12
Number 1
March 2003
10
CREATIVITY AND INNOVATION MANAGEMENT
Saxenian, A. (1994) Regional Advantage: Culture and Competition in Silicon Valley and Route 128. Harvard University Press, Boston, MA. Schott, T. (1994) Emerging and Declining Centers of Engineering Science Japan and the United States. Knowledge: Creation, Diffusion, Utilization, 15, 417–456. Uzzi, B. (1997) Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42, 35–67. Vance, C. M., Boje, D., Mendenhall, M. and Kropp, H. R. (1991) A Taxonomy of Learning Benefits from External Knowledge Sharing Meetings. Human Resource Development Quarterly, 2, 3752–3757. Von Hippel, E. (1987) ‘Cooperation between rivals: Informal knowhow trading. Research Policy, 16, 291–302.
Volume 12
Number 1
March 2003
Cathleen A. McGrath is at the College of Business Administration, Loyola Marymount University (One LMU Drive, Los Angeles, California 90045, USA, phone (310) 338–4585, e-mail:
[email protected]). Charles M. Vance is at the College of Business Administration, Loyola Marymount University (One LMU Drive, Los Angeles, California 90045, USA, phone (310) 338–4508, e-mail:
[email protected]). Edmund R. Gray is at the College of Business Administration, Loyola Marymount University (One LMU Drive, Los Angeles, California 90045, USA, phone (310) 338–7408, e-mail:
[email protected]).
© Blackwell Publishing Ltd, 2003
CREATIVITY AND WORK ENVIRONMENT
11
Creativity and Work Environment in a High-Tech Context Jozée Lapierre and Vincent-Pierre Giroux Creativity is essential for success in business, especially in the high-tech field where knowledge is the key resource. This study addresses the ways in which creativity is fostered in hightech organizations. It melds the different perspectives on organizational creativity into one six-dimensional model that defines the creativity work environment. Those dimensions are: work atmosphere; vertical collaboration; autonomy/freedom; respect; alignment; and lateral collaboration. They are valid, reliable predictors of the creativity achieved in high-tech organizations.
Introduction
C
reativity lies at the heart of successful companies, particularly those operating in the high technology sector. Businesses distinguish themselves through their capacity for continuous innovation (Nonaka and Takeuchi, 1995). Without a continuous level of innovation, many of them would simply cease to exist. Creativity is present at every stage of the process culminating in innovative products or services (Couger, 1995). Ciardi (1956, in Couger, 1995) and Koestler (1960) define creativity as the imaginatively gifted recombination of known elements into something new. Kao (1996) explains that creativity adds value to knowledge and progressively makes it more useful. Hence, creativity is a by-product of the knowledge economy, where knowledge is the key resource (Handy, 1996). Knowledge needs creativity. Fostering creativity is therefore a critical aspect of effective management. Creativity is hindered far more often than it is aided and abetted, however (Robinson and Stern, 1997; Amabile, 1998). This makes it essential to understand how to nurture creativity. Different managerial practices influence the work environment, which is the major factor fostering creativity within business organizations (Bharadwaj and Menon, 2000; Couger, 1995; Robinson and Stern, 1997; Sweetman, 2001). Furthermore, authors have recognized the need for more research on the creativity work © Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
environment (Montuory, 1998; Pearson and Chatterjee, 1997; Tesluk, Farr and Klein, 1997). This paper expands on this line of thinking on the organizational environment and creativity. More specifically, we focus on the work environment of high-tech firms so as not to dilute our findings through a multi-industry study. High-tech is a specific sector for which innovation is the raison d’être since knowledge is the key input. Consequently, we feel that for high-tech firms, the creativity work environment cannot be dissociated from knowledge or from knowledge sharing and management. Further to this, Ekvall’s study (1997) showed that the creative climate is affected by the type of industry involved. The objective of this study is to develop and validate an evaluation tool for the creativity work environment in order to assist managers in high-tech settings.
Theoretical Framework In seeking to understand how creativity is fostered in a work environment, previous studies have put the primary emphasis on psychological tools and management practices (see Table 1), all arguing that their own particular viewpoint best determines what exerts the strongest influence on organizational creativity. Previous studies have focused on one perspective and excluded arguments from the other perspective. To produce a broader picture of creativity in the high-tech setting,
Volume 12
Number 1
March 2003
12
CREATIVITY AND INNOVATION MANAGEMENT
Table 1. Theoretical background Psychological perspective
– – – –
– Climate – Autonomy Work ambiance Time resource Dynamism / openness
Amabile et al., 1996; Barrett, 1998; Ekvall, 1996, 2000. – Individuals – – Skills – Expertise – Creative problem solving Couger, 1995; Demory, 1990; Tan, 1998.
our research combines both perspectives into one conceptual model (see Table 1). As regards the first perspective, the psychological perspective, we find that psychology literature about creativity is divided into two dimensions, i.e. climate and individuals. The climate dimension has been operationalized mostly with collections of creative climate questionnaires (CCQs) by Ekvall (1996, 2000) and Amabile’s KEYS: Assessing the Climate for Creativity (Amabile et al., 1996). The CCQ instrument covers ten dimensions: challenge, freedom, idea support, trust/openness, dynamism/liveliness, playfulness/humor, debates, conflicts, risk taking and idea time. The ten dimensions address the creative climate but not the leadership and management aspects, which are nonetheless of serious concern to practitioners (Ekvall, 1996). Amabile’s KEYS (1998) evaluate six dimensions of the work climate: challenge, freedom, resources, work-group features, supervisory encouragement and organizational support. Amabile (1998) argues that the fostering of intrinsic motivation, through the sixdimensional KEYS work climate, is what will yield the most immediate outcomes in terms of organizational creativity. Climate is one dimension of our theoretical framework. The climate dimension underlies the psychological perspective and expresses one aspect of the psychology approach to creativity.
Volume 12
Number 1
March 2003
Management perspective – Knowledge management – – Cross-pollination – Knowledge sharing – Diversity of stimuli Earl and Scott, 1999; Gendreau, 1998; Jassawalla and Sashittal, 1998; Kao, 1996; Nonaka and Takeuchi, 1995; Robinson and Stern, 1997. – – – –
– Strategy communication – Alignment Communication spiral Attitude towards change Explicit encouragement
Gendreau, 1998; Jassawalla and Sashittal, 1998; Kao, 1996; Lucas, 2000; Nonaka and Takeuchi, 1995; Robinson and Stern, 1997; Groth and Peters, 1999.
The other side of the psychological perspective is found in the ‘individuals’ dimension. Couger (1995), one leading author on creativity, argues that providing employees with creative problem solving courses is one of the best improvement programmes an organization can espouse. In that vein, Couger (1995) and Demory (1990) propose many different creativity tools that individuals may use in their everyday work. An organization can enhance the creativity of its individuals by providing creativity training and assigning work to people best skilled to do that work. This defines the second dimension of our theoretical framework, i.e. the individuals. Management literature, too, provides many insights into organizational creativity. It consists of books on business written by scholars and centred on case studies and experiences, both of which are valuable approaches but differ significantly in terms of the psychometric/socio-psychological approach used in justifying the first two dimensions. The creativity definition used here is based on knowledge and, as Gurteen (1998) argues, creativity is at the very heart of knowledge management, which is our third dimension. For Nonaka and Takeuchi (1995), knowledge creation and innovation stem first and foremost from knowledge management as that concept pertains to sharing knowledge among indi-
© Blackwell Publishing Ltd, 2003
CREATIVITY AND WORK ENVIRONMENT
viduals through formal communication channels. Cross-pollination defines a work environment that allows employees to meet formally and informally, an essential aspect of the processes of knowledge exchange and creation (Kao, 1996; Paulus, 2000). Kao (1996) and Robinson and Stern (1997) emphasize the importance of serendipity. In order to live a serendipitous moment, one must be open to diverse stimuli. For instance, an organization may provide training and information sources unrelated to the main areas of interest of its employees. Knowledge management is therefore about collaboration and information sharing, formally and informally. Many of the authors who define knowledge management as an essential aspect of organizational creativity also emphasize the need for strategy communication, the fourth dimension of our theoretical framework. First, alignment is a recurring theme in creativity management literature. For Robinson and Stern (1997), a well-communicated strategy produces strong alignment. Alignment is associated with long-term objectives, corporate missions and value precision and comprehensiveness. For Gendreau (1998), alignment has to do with dreams, metaphors and building employees’ desires to possibly enhance their sense of autonomy. To produce this type of alignment, employees have to share in defining alignment, which entails a communication spiral (Nonaka and Takeuchi, 1995). The communication spiral is concerned with superiors’ attitudes towards communicating with subordinates and implementing participative management. Lastly, all of these components of the dimensions of strategy communication and knowledge management may prove worthless if a company is not prepared to accept and deal with change. Do employees have the general feeling that the company has a positive attitude toward change (Kao, 1996). In short, strategy communication refers to management’s ability to communicate a purpose.
Method We conducted an empirical study to explore the dimensionality of the creativity work environment construct. Our motivation was the need to operationalize the proposed model, which results from a lack of empirical findings in creativity combining both psychological and management perspectives. Thus, we are faced with an exploratory study requiring scale purification (Churchill, 1979) and a principal component analysis
© Blackwell Publishing Ltd, 2003
13
(Hair et al., 1987) in order to reduce the number of items and provide an initial indication of scale dimensionality (Babin, Boles and Robin, 2000).
Item Generation The original instrument consisted of 80 questions. Given the bilingual nature of the majority of companies in the outlined sample, respondents were given the choice of either an English or a French questionnaire. Some items were worded positively and others negatively so as to reduce response bias (Amabile, 1996). Some subjects were addressed by more than one item but worded differently. The 69 questions on the work environment were multiple-choice questions scaled from 1 (totally disagree) to 4 (totally agree). A fourpoint response scale was also used by Amabile et al. (1996), Ekvall (1997) and Shalley, Gilson and Blum (2000). Our intention was to avoid a neutral default option. Twenty-two of the 69 items were patterned directly after Ekvall’s creativity climate questionnaire (CCQ) (1996, 2000). The list of creativity inhibitors drawn up by Groth and Peters (1999) helped provide a general perspective to articulate many items concerning the overall management perspective. In addition to the 69 items operationalizing the four dimensions, we used four items to measure the level of creativity achieved. Creativity achieved was evaluated from two standpoints: the usefulness of the behavioural and creativity products in a business context, where novel ideas are the creative products (Couger, 1995). Lastly, the seven remaining questions concerned demographics and are not covered in the current paper, neither is the language chosen by respondents.
Research Context and Sampling Data was collected in summer 2000 by means of a multiple-choice survey e-mailed to a convenient sampling of individuals employed by Canadian high-tech firms. Key respondents were professionals and technicians of different hierarchical levels, sex, age and seniority. Individuals from ten high-tech firms in the information technology (IT) sector participated in the survey. Out of 160 questionnaires, 122 were returned, giving a response rate of 76%. Eighteen of the responses were in English. The companies were of varying size and age, and operated in the sectors of telecommunication, fibre-optics, networking, computer software and electronics.
Volume 12
Number 1
March 2003
14
CREATIVITY AND INNOVATION MANAGEMENT
Data Analysis Step 1: Dimensionality Scale Purification. We purified the measure using Churchill’s procedure (1979). First, we analysed the correlations between the 69 items; response to items involving one single construct – creativity work environment in the current study – should strongly correlate. We therefore eliminated 23 items with low interitem correlation. Second, we analysed the internal consistency of the 46 remaining items. Most items had a corrected item-total correlation higher than 0.5 with a standardized item alpha of 0.95. We set the threshold at 0.4 item-total correlations in order to eliminate some items. Seven other items were then withdrawn. The descriptive statistics singled out four out of the remaining 39 items that should be deleted owing to low variance and non-uniform distribution. One item deemed insufficiently clear was also eliminated. Using Cronbach’s alpha coefficient, we ran a second consistency analysis on the 36 remaining items. We decided to be stricter to lend maximum internal consistency to the instrument. Two-thirds of the items had a corrected item-total correlation higher than 0.55, and the standardized item alpha was 0.95. We then decided to set a decision point at 0.5 of the corrected item-total correlation. That threshold eliminated six items. Among the 30 remaining items, four had been taken directly from Ekvall’s CCQ factors (1996, 2000). Two items were from Ekvall’s ‘idea support’ factor and two from the ‘dynamism’ factor. Those two pairs of items were aggregated to improve the ratio of respondents to number of items. Aggregation of the concepts is acceptable because they make up the essence of the underlying construct (Bagozzi and Heatherton, 1994). The two new items are: idea support (IS) and dynamism (DY). All items have a normal probability plot with respect to kurtosis and skewness coefficients, as well as frequency histograms. Principal Component Analysis (PCA). After the scale purification, we used exploratory factor analysis to eliminate items with low inter-item correlation and gain an initial indication of scale dimensionality (Babin, Boles and Robin, 2000). PCA is particularly useful when researchers seek to identify underlying factors that potentially characterize a specific group of variables. The sampling size (n = 111) was sufficient for this analysis because it obeys the rule of being four to five times larger than the number of variables (27). This is stricter than is mentioned in Hair, Anderson and
Volume 12
Number 1
March 2003
Tatham (1987). The PCA was performed with the varimax factor rotation procedure because the four dimensions are expected not to correlate (Iaobucci, 1994). As shown in Table 2, all eigenvalues exceed the value one, and the cumulative percentage of total variance explained was 66.9 per cent. These results indicate that six independent dimensions with 27 items represent the creativity work environment. To name and describe the six dimensions, we use the items with a factor loading of 0.5 or more. This eliminates items that crossload on more than one dimension, thus reducing the number of items to 23. Though the six dimensions do not exactly represent the theoretical model with four dimensions, they nonetheless include both management and psychological dimensions. None of the six factors relates to the ‘individuals’ dimension, the only dimension of the theoretical framework not specifically related to the work environment. Now more than ever, we can regard the ‘creativity work environment’ as the research domain. Following is a description of the six dimensions. D1: Work Atmosphere The dimension called ‘work atmosphere’, i.e. the human aspect of the organization, is the most important. People feel that they belong to the organization; they feel that they have a role to play, a purpose, and that their ideas will be listened to and supported in a non-critical environment. Moreover, people can sense the enthusiasm of the whole organization, be stimulated by it and enjoy working in it (Lucas, 2000). Handy (1996) wrote that the present problem with big organizations is that they are hopelessly boring. The work atmosphere dimension reveals that the opposite should be happening so as to foster creativity. Work atmosphere can be seen as a subdimension of Ekvall’s climate (1996) because half of the items loading on this factor are similar to CCQ factors (2000). D2: Vertical Collaboration This dimension has to do with the quality of the communication spiral (Nonaka and Takeuchi, 1995) among the different hierarchical levels. This dimension is a mix of two dimensions of the theoretical framework: knowledge management and strategy communication. Employees feel free to talk to their bosses about their ideas, without fear of being judged. Superiors trust employees’ viewpoints and often talk with them, ask for feedback and allow employees to participate in
© Blackwell Publishing Ltd, 2003
CREATIVITY AND WORK ENVIRONMENT
15
Table 2. Principal component analysis Items
(IC) It is obvious that creativity is at is at the heart of the company.* (SP) Each person understands his/ her role in, and his/her contribution to, the company. (IS) This department enthusiastically welcomes new ideas/(–) Suggestions for improvement are regularly ignored. (ID) There is an easy, natural flow of ideas within the division. (R) People trust one another here in this division. (DY) There is a lot of enthusiasm here/The work atmosphere is highly stimulating. (P) There is a cheerful atmosphere here. (CV) The employees put the company values into practice. (FS) Employees feel free to express their ideas to their boss. (ST) The bosses have confidence in their employees’ ideas/ viewpoints. (TD) Each employee participates in determining and defining their tasks and/or objectives. (SM) People feel they are maximizing their skills and knowledge in their work.* (PS) People feel they can use many of their skills in their work.* (IE) The employers encourage employee initiatives.* (PH) It is easy to find a resource person within the division. (B)(–)The bureaucracy weighs heavily on the employees and limits their freedom to accomplish what they would like. (F) There is a lot of freedom here. (LR) People are not afraid to take risks here.
© Blackwell Publishing Ltd, 2003
Dimensions Work atmosphere
Vertical collaboration
Autonomy /freedom
Respect
Alignment
Lateral collaboration
0.47
0.21
0.34
0.18
0.11
0.27
0.64
0.41
0.04
-0.01
0.16
0.09
0.64
0.40
0.20
0.25
0.10
0.21
0.51
0.24
-0.02
0.07
0.48
0.38
0.64
0.16
0.24
0.26
0.23
0.08
0.64
0.22
0.34
0.33
0.15
0.11
0.70
0.09
0.36
0.24
0.10
0.13
0.32
0.74
-0.02
0.16
0.23
0.16
0.043
0.51
0.27
0.50
0.24
0.31
0.34
0.63
0.23
0.25
0.25
0.01
0.08
0.74
0.18
-0.08
0.13
0.29
0.33
0.48
0.29
0.37
0.15
0.03
0.38
0.44
0.28
0.37
-0.03
0.09
0.30
0.43
0.44
0.29
0.27
-0.07
0.26
0.16
0.67
0.01
0.21
0.19
0.48
0.04
0.63
-0.11
-0.04
0.25
0.13 0.18
0.19 0.22
0.63 0.60
0.39 0.25
0.24 0.41
0.12 0.17
Number 1
March 2003
Volume 12
16
CREATIVITY AND INNOVATION MANAGEMENT
Table 2. Continued Items
Dimensions Work atmosphere
Vertical collaboration
Autonomy /freedom
Respect
Alignment
Lateral collaboration
0.17
0.36
0.33
0.59
0.00
0.39
0.10
-0.04
0.22
0.52
0.47
0.17
0.49 0.33
-0.06 0.20
0.12 -0.07
0.54 0.72
0.32 0.02
0.18 0.19
0.18
0.21
0.21
0.17
0.72
0.22
0.15
0.27
0.18
-0.05
0.75
0.09
0.28
0.27
0.22
0.27
-0.10
0.52
0.14
0.23
0.08
0.14
0.25
0.74
0.13
0.01
0.23
0.19
0.25
0.73
11.80 43.7 43.7
1.47 5.5 49.2
1.40 5.2 54.4
1.27 4.7 59.1
1.08 4.0 63.1
1.04 3.9 66.9
(SC) Generally speaking, employees feel their bosses listen to them. (T) Employees are not worried that people will talk behind their backs. (M) (–) People feel exploited. (ES) The selection of employees is carefully scrutinized. (MS) The corporate mission is specific and clear to all employees. (LT) The long-term company objectives are known and clear to everyone. (OW) Employees are interested in the work of others. (WI) People are kept informed of the work and progress of other teams. (WA) Information about different company projects is readily available to all employees. Eigenvalue % variance Cum. Variance N MSA-KMO Bartlett
111 0.90 1766.29
Notes: (–) indicates reverse scoring * indicates items with a factor loading < 0.5
determining and defining their tasks and objectives. The quality and high frequency of vertical communication also drive tacit knowledge, such as corporate values and mission (Nonaka and Takeuchi, 1995). Gurteen (1998) calls this general perspective which employees get with vertical collaboration ‘know-why’: ‘know-why is often more important than know-how as it allows you to be creative – to fall back to principles – to reinvent your know-how and to invent new know-how’.
Volume 12
Number 1
March 2003
D3: Autonomy/freedom The dimension called ‘autonomy/freedom’ represents a sense of ownership and control over one’s own work and ideas (Amabile et al., 1996). Bureaucracy does not weigh on the shoulders of individuals, dictating the manner in which they must do their job. In a work environment that fosters creativity, people can readily contact experienced people to discuss problems and alternatives. People feel free to be who they are and are encouraged to use
© Blackwell Publishing Ltd, 2003
CREATIVITY AND WORK ENVIRONMENT
17
their own initiative. This implies that individuals are not apprehensive about venturing outside established boundaries and taking risks.
much of the effort in developing innovative ideas is social in nature: exposure to different kinds of information in order to stimulate lateral (divergent) thinking.
D4: Respect
Summary
The dimension ‘respect’ refers to the emotional and intellectual comfort zone. Individuals have no fear of being stabbed in the back; they feel they can trust their work mates as well as their superiors. Indeed, people feel that their superiors listen with careful consideration. The treatment of employees is seen as fair and equal for everyone (Katz, 1997). In addition, the organization is very careful in selecting its employees, and this enhances the feeling of respect by sending the message that people are trustworthy and not taken for granted (Couger, 1995).
As presented in the theoretical framework, the outcomes just described indicate that the creativity work environment is multidimensional and, as expected, is well explained by socio-psychological factors and by management and knowledge management factors. Interestingly, this study overlaps a study by Rickards, Chen and Moger (2001), who investigate six team-effectiveness dimensions as predictors of two team-performance criteria – creativity and productivity. Specifically, the most important predictor of creativity achieved in this study, work atmosphere, overlaps three of the six factors, namely, creative climate, idea owners and platform of understanding. Another overlapping finding concerns our lateral collaboration dimension with their learning from experience factor. Results of both studies carried out into different countries with different samples and units of analysis are revealing. Note that the work environment (climate, atmosphere) of individual workers or teams is of primary importance when dealing with organizational creativity.
D5: Alignment Understanding of the long-term organizational orientations represents the alignment dimension. According to Robinson and Stern (1997), alignment is crucial to creativity. When employees feel that the corporate mission is clear and precise, they understand the essence of their work; they picture the castle for which they are carving stones (Arad, Hanson and Schneider, 1997). In a relevant jazz metaphor, Barrett (1998) called the alignment for creativity in organization a mutual orientation around the basic root movement of the chord patterns. When that mutual orientation is achieved, even the basic chords themselves can be altered, augmented or replaced. Alignment builds a desire, a common dream (Gendreau, 1998) in which everyone in the organization can invest in his or her own way. D6: Lateral Collaboration The lateral collaboration dimension concerns information access and sharing among employees and teams. This dimension is based on the cross-pollination and knowledge sharing components of the theoretical framework. Without a competition mindset, people are interested in the work of others. Employees want to know what other teams do and are continually kept informed of other teams’ achievements and needs. Information exists; it is not hidden. It is easy to find, and people are curious. Information is shared through formal and informal channels. Informal channels become important since creative ideas may not fit the formal processes and may in fact suggest new processes (Couger, 1995). As Csikszentmihalyi and Sawyer (1994) suggest,
© Blackwell Publishing Ltd, 2003
Step 2: Reliability and Validity Reliability As indicated in Table 3, the reliability analysis computed with Cronbach’s alpha coefficients suggests that the six dimensions and creativity achieved are reliable: alpha values are greater than 0.60, the threshold recommended for testing scale reliability for exploratory study (Lockhart and Russo, 1994). Convergent Validity Principal component analysis is concerned with exploring the patterns of relationships among a number of variables, where the researcher has only limited control over which variables are indicators of which latent construct (Hair et al., 1998). Conversely, in confirmatory factor analysis, the researcher has full control over the specification of indicators. Moreover, confirmatory factor analysis allows for statistical testing of the goodness-of-fit of the proposed factor solution, which is not possible with principal component analysis. Confirmatory factor analysis is particularly useful for validating scales for the measurement of
Volume 12
Number 1
March 2003
18
CREATIVITY AND INNOVATION MANAGEMENT
specific constructs (Steenkamp and van Trijp, 1991). The remaining 23 items evaluating the creativity work environment were used to compute the variance and covariance matrix using the PRELIS procedure, which served as input for all confirmatory analyses. The validation analysis was computed using LISREL 8 (Jöreskog and Sörbom, 1993). We estimated a confirmatory factor model – in which the six
dimensions were allowed to freely correlate – to assess overall convergent validity (Bagozzi and Baumgartner, 1994). Assuming the exploratory nature of the research, correlations between some measurement errors were necessary to strengthen the LISREL parameters. All measurement error correlations were theoretically defensible to preserve the model interpretation (Venkatraman, 1985). Table 3 gives the results. All items are still loading sig-
Table 3. Reliability analysis and convergent validity Dimensions†
Reliability analysis Corrected item-total correlation
Alpha if item deleted
D1 Work atmosphere SP (Sense of purpose) IS (Idea support) ID (Idea sharing) R (Reliance) DY (Dynamism) P (Playfulness)
0.60 0.77 0.60 0.68 0.74 0.72
0.87 0.84 0.87 0.85 0.84 0.85
D2 Vertical collaboration CV (Application of corporate values) FS (Free speech between sup. and empl.) ST (Superiors’ trust) TD (Employees participate in work description)
0.70 0.64 0.71 0.62
0.78 0.80 0.77 0.82
D3 Autonomy/freedom PH (Peer help) B (Bureaucracy) (–) F (Freedom) LR (Low risk aversion)
0.63 0.51 0.60 0.63
0.70 0.77 0.73 0.70
D4 Respect SC (Superiors’ consideration of employees) T (Trust) M (Mistreatment) (–) ES (Cautious employee selection)
0.63 0.47 0.65 0.53
D5 Alignment MS (Mission statement clear, understood) LT (Long-term objectives clear, understood) D6 Lateral collaboration OW (Interest in others’ work) WO (Others’ work info actively shared) WA (Others’ work info available)
Convergent validity
Components standardized item alpha
l (ML)
T-Value
0.88
0.42 0.48 0.49 0.52 0.52 0.55
7.04**** 10.68**** 7.53**** 9.00**** 9.73**** 9.00****
0.84
0.43 0.51 0.51 0.40
8.49**** 9.45**** 10.34**** 6.52****
0.79
0.56 0.51 0.49 0.63
7.35**** 6.10**** 8.28**** 9.48****
0.68 0.76 0.66 0.73
0.77
0.47 0.40 0.59 0.45
9.40**** 5.74**** 8.63**** 7.57****
0.70 0.70
– –
0.82
0.75 0.65
10.55**** 8.37****
0.47 0.61 0.51
0.68 0.48 0.64
0.71
0.41 0.51 0.57
7.57**** 7.53**** 7.51****
N = 111; c2(195) = 191.37; P = 0.56; RMSR = 0.027; GFI = 0.88; AGFI = 0.83; D = 1.004; CFI = 1.00. † Summated scale. **** p £ 0.001, *** p £ 0.01, ** p £ 0.05.
Volume 12
Number 1
March 2003
© Blackwell Publishing Ltd, 2003
CREATIVITY AND WORK ENVIRONMENT
19
Table 4. Discriminant validity F Factors
F 1. F 2. F 3. F 4. F 5. F 6.
Work Atmosphere Vertical Collaboration Autonomy/Freedom Respect Alignment Lateral Collaboration
Phi (F) matrix unconstrained F1
F2
F3
F4
F5
F6
1.00000 0.81881 0.81368 0.84155 0.53600 0.67423
1.00000 0.71184 0.78218 0.57720 0.54614
1.00000 0.74816 0.68191 0.74648
1.00000 0.57876 0.81703
1.00000 0.54213
1.00000
n = 111. Unconstrained: c2(195) = 191.37; (p = 0.56). Constrained (All F = 1): c2(210) = 348.44; (p = 0.00).
nificantly and substantially (t-Value > 2.0) on their underlying dimensions. All l – estimated with maximum likelihood (ML) – are higher than 0.30, the generally accepted level (Venkatraman, 1985), and the model fit is adequate (GFI = 0.88; AGFI = 0.83; RMSR = 0.03; D = 1.00), providing evidence for convergent validity (Bagozzi and Baumgartner, 1994). Therefore, all 23 items and 6 dimensions were retained. Discriminant Validity Discriminant validity refers to the degree to which the measures of the six constructs are unique. This is achieved when measures of each construct converge on their corresponding true scores (F) and the correlation between pairs of dimensions is significantly different from unity. This requires comparing a model with this correlation constrained to equal one with the unconstrained model. A significantly lower c2 value for the model with the unconstrained correlation associated with a p-value < 0.05 (Jöreskog, 1971), when compared with the constrained model, provides support for discriminant validity (Zaheer and Venkatraman, 1994). As indicated in Table 4, the test for the constrained model yielded the following statistics: c2(210) = 348.44 (p = 0.00), indicating poor data fit. Furthermore, the difference between the two models is c2(15) = 157.07 (p = 0.00), providing strong support for the discriminant validity. Results indicate that despite the four-point scales used in the questionnaire, the variance is large enough to draw out six strongly discriminated dimensions.
Step 3: Creativity Achieved Now that we have six valid, reliable dimensions for the creativity work environment,
© Blackwell Publishing Ltd, 2003
Table 5. Multiple regression with enter method Dependent Variable = Creativity Achieved Independent variables
b
D1 Work atmosphere D2 Vertical collaboration D3 Autonomy/freedom D4 Respect D5 Alignment D6 Lateral collaboration
0.48**** 0.31**** 0.18*** 0.30**** 0.16*** 0.30****
R2 = 56.28%. Adjusted R2 = 53.76%. Overall F = 22. P = ****. Durbin-Watson = 1.77; Ho accepted, p < 0.01 (Dillon and Goldstein, 1984). **** p £ 0.001, *** p £ 0.01. One-tailed test.
we can test if all creativity work environment dimensions are predictors of creativity achieved. We performed a multiple regression analysis using the enter method (Table 5). As expected, all creativity work-environment dimensions explain creativity achieved with a high degree of significance.
Implications for Managers We believe this research stands out well, as it operationalizes the creativity work environment with dimensions that involve both psychological and management approaches to that environment. The model presents dimensions that have not been articulated before, particularly the ‘vertical collaboration’ and ‘lateral collaboration’ dimensions, both of
Volume 12
Number 1
March 2003
20
CREATIVITY AND INNOVATION MANAGEMENT
which derive from the strategies’ communication and knowledge-management theoretical background. This model can be used in different ways by its practitioners.
Evaluation and Application Tools The model can bring insights into ways of fostering creativity. The dimensions can be used as indicators of the creativity achieved by employees. It could be used to ‘audit’ creativity. Depending on the scores obtained in the different dimensions, a company can determine its strengths and weaknesses in fostering creativity, and share that knowledge with employees and supervisors. The model, used over time in a ‘before and after’ fashion, can measure the impact of managerial practices on the creativity work environment. It can also serve as an unbiased tool for comparing groups within an organization, as well as for understanding employee behaviour in terms of creativity and the work environment. Furthermore, the evaluation results can be applied easily to the work environment, since the six dimensions defined are close to worklife considerations. The tool gives not only an evaluation of the creativity achieved but also indicates which aspect of the work environment the manager of the evaluated group should work on. As Tan (1998) suggested, a total system approach to managing creativity in organizations requires tremendous commitment from top managers. The model can help to focus that effort and share it with employees.
Knowledge Management As previously mentioned, the inclusion of dimensions inspired by knowledgemanagement theories follows the theory that creativity is one of the main goals of knowledge management (Gurteen, 1998). However, creativity is rarely addressed by knowledgemanagement academicians and practitioners. In the current model, knowledge management is included in the definition of the creativity work environment. Specifically, practitioners can find in the creativity work environment a relevant factor in calculating return on investment (ROI) resulting from knowledgemanagement investment. Indeed, ROI calculations are a major issue for knowledgemanagement practitioners (McDermott and Rush, 2001), and Couger (1995) explains that a creativity program may have a very high ROI. The impact of knowledge management over the creativity work environment is of high value for practitioners.
Volume 12
Number 1
March 2003
Human Factors This research assesses the findings that creativity is a matter of ‘human’ factors, otherwise known as the ‘soft stuff’ (Bentley, 1998). The model refers to human aspects of management. Allocating resources and processes is not enough to foster creativity; management – where knowledge is the main resource and innovation, the main goal – is a function of human complexity and subtlety. Therefore, as indicated by Amabile (1998), a manager should spend a good amount of time evaluating who should be in what team and considering human factors instead of simple resources and competencies allocation. It is not only intrinsic factors that are important to creativity; extrinsic motivators can also restrain it. Finally, Shalley, Gilson and Blum (2000) mentioned that when a job’s creativity is higher, job satisfaction increases and employees’ intentions to leave decrease as well. High turnover being a subject of great concern in high-tech organizations (Lucas, 2000), this other impact of creativity should be of interest to high-tech managers.
Another Way to Consider the Creativity Work Environment Another way to consider the creativity work environment is to analyse whether or not the organization is ready to accept creative ideas. Ideally organizations want to foster creativity, but in practice not all of them are ready to accept it. According to Robinson and Stern (1997), most of the employees’ creativity impacts processes. In fact creativity may affect several aspects of the organization: processes, vision/mission, products and division of labour. As an example, it can be quite tough on an experienced senior manager to be shown a new and more efficient way to do something by a junior employee, particularly when the senior has been doing his best in a certain manner for many years (Handy, 1996). The courteous junior employee might even avoid spreading his/her new ideas in order to avoid embarrassing the senior manager. Creativity can also be a stress on a budget. Rationally, if an idea is very creative, i.e. it suggests something that has not been thought of before, the budget does not expect it. The creative employee might be encouraged to be creative but he/she could still have to face many hurdles in order to have the idea considered seriously. Thus creativity presents an opportunity, but a dangerous one (Couger, 1995).
© Blackwell Publishing Ltd, 2003
CREATIVITY AND WORK ENVIRONMENT
Limitations and Implications for Future Research The main contribution of this research is that it proposes a general perspective to the creativity work environment. The model is a relevant step forward in operationalizing the creativity work environment. However, future research should also take the following points into consideration: • In terms of methodology, this research is solid, especially considering that it represents exploratory research. However, in order to give more strength to the PCA analysis, future investigations should have more respondents, and these respondents should come from a higher number of organizations. • Even though the measuring format has been proven valid and reliable in this study, it would be necessary to measure the creativity-achieved construct with existing and validated evaluation tools. In order to measure creativity achieved, i.e. to evaluate the product of creativity, Couger (1995) suggests observing the product of the creative activity and making some subjective judgement regarding that product by applying an agreed-upon definition of what constitutes creativity. • A further version of the survey should consider financial resources. Amabile (1998) argues that deciding how much to give to a team or project is a sophisticated judgment call that can either support or kill creativity. The budget has to be large enough to provide the necessary resources to accomplish the task. However, too much money can become a source of extrinsic motivation and undermine one’s intrinsic motivation (Robinson and Stern, 1997). Hence, money is a complex issue, but is an important part of fostering creativity. • A general survey on the creativity work environment should also deal with the concept of stability. Some respondents commented that their company had been through too many changes in too little time, and many of them felt overwhelmed. As a result, employees had no more energy to devote to generating new ideas; continual adaptation to a new job was already asking enough from them. As Amabile (1998) wrote, it is difficult, if not impossible, to work creatively toward a target if it keeps moving. However, too much stability can destroy creativity (Robinson and Stern, 1997). The right dosage is not easily reached. • In order to have a total perspective over creativity, a complete model must also include
© Blackwell Publishing Ltd, 2003
21
the concept of innovation, which is the essential goal of creativity in organizations. Creativity assists innovation, which is defined as ‘turning ideas into products, services [or processes]’ (Couger, 1995). Therefore, a high-performance creativity work environment involves, on the one hand, an environment that brings a higher sense of creativity achieved, and generates, on the other hand, concrete high-impact innovations of the organization’s products, services and processes. While the model presented here addresses the first part of this process, there is a need to operationalize a full model that will also cover the transformation of ideas into products, services, and processes. But, as suggested by Nohria and Gulati (1996), there is an inverse U-shaped relationship between slack and innovation in organizations: both too much and too little slack may be detrimental to innovation. Since these authors have validated the slack concept in a high-tech setting, the full model should also take into account such a possible constraint on creativity and on innovation as well.
Conclusion The six dimensions cover three out of four of the theoretical background dimensions: climate, knowledge management and strategy communication, and concern the creativity work environment, which largely explains the creativity achieved. By contrast with Ekvall’s (1996) CCQ, our model also addresses the concepts of leadership and management, along with the dimensions of alignment and vertical and lateral collaboration. It enhances Amabile’s (1998) KEYS, mostly by including and measuring knowledge-management elements. Since knowledge is at the very heart of both our creativity definition and the growth of high-tech firms, emergence of the lateral and vertical communication dimensions in the creativity work environment model might be attributed to the specific situation of high-tech organizations.
References Amabile, T.M. (1998) How to kill creativity. Harvard Business Review, (September-October), 77–87. Amabile, T.M., Conti, R., Coon, H., Lazenby, J. and Herron, M. (1996) Assessing the work environment for creativity. Academy of Management Journal, 39(5), 1154–1184. Arad S., Hanson, M.A. and Schneider, R.J. (1997) A framework for the study of relationships between organizational characteristics and organizational
Volume 12
Number 1
March 2003
22
CREATIVITY AND INNOVATION MANAGEMENT
innovation. Journal of Creative Behavior, 31(1), 42–58. Babin, J.B., Boles, J.S. and Robin, D.P. (2000) Ethical work climate. Journal of the Academy of Marketing Science, 28(3), 345–358. Bagozzi, R.P. and Baumgartner, H. (1994) The evaluation of structural equation models and hypothesis testing. In R.P. Bagozzi (ed.), Principles of Marketing Research. New York, Blackwell Publishers. Bagozzi, R.P. and Heathberton, T.F. (1994) A general approach to representing multi-faceted personality constructs: Application to state self-esteem. Structural Equation Modeling, 1, 35– 67. Barrett, F.J. (1998) Creativity and improvisation in jazz and organizations. Organization Science, 9(5), 605–622. Bentley, T. (1998) The soft stuff is really the hard stuff (human relations in the workplace). Management Accounting, February. Bharadwaj, S. and Menon, A. (2000) Making innovation happen in organization: individual creativity mechanisms, organizational creativity mechanisms or both? Journal of Product Innovation Management, 17, 424–434. Churchill, G.A. Jr. (1979) Paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 16, 64–73. Ciardi, J. (1956) What every writer must learn. Saturday Review, Dec. 15, 17. Couger, J.D. (1995) Creative Problem Solving and Opportunity Finding. Boyd & Fraser Publishing Company, Toronto, Canada. Csikszentmihalyi, M. and Sawyer, K. (1994) Creative insight: the social dimension of a solitary moment. In Sternberg, R.J. and Davidson, J.E. (eds.), The Nature of Insight. Bradford Book, The MIT Press, Boston, USA, pp. 329–364. Demory, B. (1990) Créativité? Créativité . . . Créativité! Méthodes et outils. Les Éditions Agence d’Arc Inc, Montréal, Canada. Dillon, W. and Goldstein, M. (1984) Multivariate Methods and Applications. John Wiley and Sons. N.Y., USA. Earl, M.J. and Scott, I.A. (1999) Opinion: What is a chief knowledge officer? Sloan Management Review (winter). Ekvall, G. (2000). CCQ: Creative climate questionnaire. Innovation Centre Europe www.iceurope.com. Ekvall, G. (1997) Organizational conditions and levels of creativity. Creativity and Innovation Management, 6(4), 195–205. Ekvall, G. (1996) Organization climate for creativity and innovation. European Journal of Work and Organizational Psychology, 5(1), 105–123. Gendreau, S. (1998) La Cité des intelligences. Les Éditions Céra, Montréal. Groth, J.C. and Peters, J. (1999) What blocks creativity? A managerial perspective. Creativity and Innovation Management, 8(3), 179–187. Gurteen, D. (1998). Knowledge, creativity and innovation. Journal of Innovation Management, 2(1), 5–13. Hair, J.F. Jr., Anderson, R.E. and Tatham, R.L. (1987) Multivariate Data Analysis, 2nd Edition. MacMillan Publishing Company, New York.
Volume 12
Number 1
March 2003
Hair, J.F. Jr., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998). Multivariate Data Analysis, 5th Edition. Prentice Hall, New Jersey. Handy, C. (1996, 1989) L’Âge de déraison (The Age of Unreason). Village Mondial. Iaobucci, D. (1994) Classic factor analysis. In Bagozzi, R.P. (ed.), Principles of Marketing Research. Blackwell Publishers, Cambridge, pp. 279–316. Jassawalla, A.R. and Sashittal, H.C. (1998) An examination of collaboration in high-technology new product development processes. Journal of Product Innovation Management, 15, 237–254. Jöreskog, K.G. and Sörbom, D. (1993) Lisrel 8: User’s Reference Guide, SSI, Chicago. Jöreskog, K.G. (1971) Statistical Analysis of Congeneric Tests. Psychometrika, 36, 109–133. Kao, J. (1996) Jamming: The Art and Discipline of Business Creativity. Harper Business. Katz, R. (1997) Organizational socialization and the management of uncertainty. In Katz, R. (ed.), The Human Side of Managing Technological Innovation. Oxford University Press, Oxford, pp. 355– 369. Kirton, M.J. (1987) Adaptors and innovators. Cognitive style and personality. In Isaksen, S.G. (ed.). Frontiers of Creativity Research, Bearly Ltd, Buffalo. Lockhart, D.C. and Russo, R.J. (1994) Mail and telephone surveys. In Bagozzi, R.P. (ed.). Principles of Marketing Research, Blackwell Publishers, Oxford, pp. 116–161. Lucas, W.A. (2000) Challenge and personnel strategy for technical professionals. International Center for Research on the Management of Technology, (February). McDermott, R. and Rush, D. (2001) Assessing the impact of knowledge management: combining qualitative and quantitative measures. American Productivity and Quality Center, (September). Montuori, A. (1998) Book review on Couger’s creative problem solving and opportunity findings. Journal of Creative Behavior, 32(3), 223– 225. Nohria, N. and Gulati, R. (1996) Is slack good or bad for innovation. Academy of Management Journal, 39(5), 1245–1264. Nonaka, I. and Takeuchi, H. (1995) The knowledge creative company: How Japanese companies create the dynamics of innovation. Oxford University Press, Oxford. Paulus, P.B. (2000) Groups, teams, and creativity: The creative potential of idea-generating groups. Applied Psychology: An International Review, 49(2), 1237–1262. Pearson, C.A.L. and Chatterjee, S.R. (1997) Participant characteristics and organizational processes: a Hong Kong case study. Empowerment in Organizations, 5(3), 151–165. Rickards, T., Chen, M-H. and Moger, S. (2001) Development of a self-report instrument for exploring team factor, leadership and performance relationships. British Journal of Management, 12(3), 243–250. Robinson, A.G. and Stern, S. (1997). Corporate creativity: how innovation and improvement actually happen. Berrtet-Koehler Publishers Inc.
© Blackwell Publishing Ltd, 2003
CREATIVITY AND WORK ENVIRONMENT
Shalley, C.E., Gilson, L.L. and Blum, T.C. (2000) Matching creativity requirements and the work environment: Effects on satisfaction on intention to leave. Academy of Management Journal, 43(2), 215–223. Steenkamp, J.E.M. and van Trijp, H.C.M. (1991) The use of lisrel in validating marketing constructs. International Journal of Research in Marketing, 8(4), 283–299. Sweetman, K. (2001) Management mistakes squelch employee innovation. MIT Sloan Management Review, (Summer), 9–10. Tan, G. (1998) Managing creativity in organizations. Journal of Creativity and Innovation Management, 7(1), 23–31. Tesluk, P.E., Farr, J.L. and Klein, S.R. (1997). Influences of Organizational Culture and Climate on Individual Creativity. Journal of Creative Behaviour, 31(7), 27–41. Venkatraman, N. (1985). Strategic orientation of business enterprises: the construct and its measurement, U.M.I. Dissertation Services.
© Blackwell Publishing Ltd, 2003
23
Zaheer, A. and Venkatramen, N. (1994). Determinants of electronic integration in the insurance industry: an empirical test. Management Science, 40(5), 549–566.
Jozée Lapierre Ph.D in marketing is professor of high-technology marketing at the École Polytechnique of Montréal. (Case Postale 6079, Succursale Centre-Ville, Montréal (Québec) Canada H3C 3A7, phone (514) 340–4711 (5864), e-mail:
[email protected]). Vincent-Pierre Giroux holds a B.Sc. in Computer Engineering and Master’s degree in Management of Technology from École Polytechnique de Montréal (Case Postale 6079, Succursale Centre-Ville, Montréal (Québec) Canada H3C 3A7).
Volume 12
Number 1
March 2003
24
CREATIVITY AND INNOVATION MANAGEMENT
Whatever Happened to Synectics? Vincent Nolan The article brings the story of the Synectics creative problem-solving techniques (originating in 1960) up to date. It identifies its origins with a group of inventors, who tape-recorded their meetings to research their creative process, and the resulting benefits and weaknesses. Synectics today is a set of process tools which can be used successfully in a variety of situations, either in a specific sequence (as in the original invention model) or individually according to the needs of the situation. The result is a variety of meeting models and techniques for enhancing personal effectiveness
F
rom time to time, academic friends who are interested in creativity ask me ‘Whatever happened to Synectics? It has disappeared off our radar screens completely’. The short answer is that Synectics is alive and well, and flourishing in the world of international business. Maintaining a presence on academic radar screens has not been a priority for the company. But it is becoming one for Synectics Education Initiative, the registered charity dedicated to promoting the use of Synectics skills in the education system. So this article brings the Synectics story up to date and, hopefully, removes some misunderstandings that seem to exist among some members of the academic community. The academic world, it seems to me, gets most of its information from the written word. But there is another world out there, the world of experience, people using and developing creative methods like Synectics. The written word, in my experience, tends to lag what might be called ‘the real world’ by an interval of approximately ten years – simply because the practitioners are too busy doing it to have the time to write about it! Unfortunately, there is only one book with the word ‘synectics’ in its title, the one written by W.J.J. Gordon in 1960. This is how it is possible for a PhD thesis on creativity published in 1995 to contain a chapter on Synectics derived entirely from Gordon’s 1960 book! It is like deriving your knowledge of computers from a book written in 1960 called ‘Computers’. Excellent and still relevant though Gordon’s book is (particularly on the subject of
Volume 12
Number 1
March 2003
‘metaphorical process’) there are several later books by Synectics practitioners, which complement Gordon and bring the subject up to date. In chronological order, they are: The Practice of Creativity by George Prince (1970), (co-founder with Gordon of the Synectics company), my own Open to Change (Nolan, 1981) and Innovators Handbook (Nolan, 1989), Innovation and Creativity by Jonne Ceseranni and Peter Greatwood (1995), Creative Education (Synectics Education Initiative (ed. Nolan), 2000), and Your Life is a Series of Meetings by George Prince (2002). A book by Richard Harriman (Chairman of Synectics Corp) and Jeff Mauzy is scheduled for publication in Spring 2003. Not all academics fall into the ‘Synectics is Gordon’ trap. Michael J. Hicks (1991) covers Synectics accurately in his excellent Problem Solving in Business and Management as a result of talking to practitioners and users of Synectics. Likewise, Dr John Martin and Jane Henry of the Open University collaborated with Synectics Ltd to generate the material contained in the OU’s module on Creative Management (1992) for their MBA Course (including filming live sessions for the accompanying video). Dr R.C. Parker, in The Management of Innovation (1982) gives an accurate description of the Synectics process together with transcripts of live sessions and follows up the results. Michael J. Hicks is currently assembling a comprehensive bibliography of writings about Synectics; it will be accessible from the website www.synecticsworld.com. To be fair, much of the problem arises from the word ‘synectics’ itself. It was made up by
© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
WHATEVER HAPPENED TO SYNECTICS?
the founders of Synectics Inc. to convey ‘the bringing together of diverse forces’, from the Greek, ‘syn’ and ‘ektos’. It was, says George Prince (1981), ‘our first big mistake’ and the mistake was compounded by applying it both to the methods, and to the company (not to mention Gordon’s book). By registering the word as a trademark, the Synectics company gave itself the task of ensuring that everything that was done in the name of Synectics came up to their standards of quality – at the same time as running training courses and training in-company trainers. Understandably, course graduates (and indeed readers of Gordon’s book) tend to believe that they are practising ‘synectics’ without necessarily matching Synectics Inc.’s quality standards. In this article, I am using a capital S for Synectics to refer to the company and the work done by its accredited personnel and a lower case ‘s’ for the generic synectics as practised by people who are not accredited. The attempt to protect the trademark has not been completely successful – the word has become a generic term for creative problemsolving in groups (see The New Shorter Oxford English Dictionary) – and a web search on the word Synectics produces a bewildering array of diverse sites, only one of which is that of the Synectics Corporation, www.synecticsworld. com. To complicate matters further, Synectics is an evolving body of knowledge, from the very nature of its origins and its continuing ethos. Synectics resulted from the decision by a group of inventors (the Invention Design Group of the Arthur D. Little consulting company) in the 1950s, to research their own creative process by tape-recording their meetings and discussions. (That is the official story; there is an alternative explanation, which I like better – whenever they had a breakthrough in their invention activity, there were furious arguments about whose idea it was and how it had happened. The tape-recording was set up to settle these arguments.) Its origin with a group of inventors is a great strength of the Synectics process (and also a weakness). It is a strength because invention requires a strong process to carry an idea through to implementation, through an often lengthy process of development. Brainstorming, by contrast, came from an advertising context, where the idea itself can be the solution (Osborn, 1953), requiring further craft work in copywriting, graphics and TV production but little further creativity. Whatever the true story (and both could be true), recordings of live meetings became an integral part of the Synectics research and learning process (video having replaced audio
© Blackwell Publishing Ltd, 2003
25
as soon as it became available). Consequently, new insights are always being discovered, even though a core body of knowledge has stood the test of application to real-world situations over a long period of time. (Unfortunately, each new generation of Synectics practitioners has a tendency to re-invent the terminology, which can generate further confusion.) A classic example is the (possibly apocryphal) story of the creation of the pre-war advertisement for Player’s cigarettes: Player’s Please. The sales director is reported as saying in a brainstorming session, ‘What we need is something dead simple, like Players Please!’ and scrawled it up on a flipchart. Immediately it was seen as a solution – a simple alliterative catchphrase, with an attractive double meaning. It appeared in advertisements and on posters, in the sales director’s original scrawl. James Dyson, on the other hand, built over 5,000 prototypes before launching his cyclone vacuum cleaner! His ‘development engineering’ required constant injections of creativity and a sustained climate of open-minded experiment (Dyson, 1998). The Synectics process views the initial approaches generated in a period of suspended judgement (similar to, but more sophisticated, than brainstorming) as starting points for further development rather than as finished solutions to be accepted or rejected. They call them ‘springboards’: they are takingoff points, not (normally) end points (Nolan, 1989). At the end of the explicitly divergent process to generate springboards, Synectics moves on to the development process, which converts initially new, but not feasible approaches into new and feasible courses of action. It does so by using a developmental type of judgement, to be made by the individual who will be responsible for implementing the solution. A developmental judgement is a temporary, provisional judgement, which articulates what is, or might be, attractive about an idea and goes on to express its major shortcoming as a ‘need for improvement’. There are two important elements to this apparently simple process: the first is the positive emphasis of finding, even searching for, all the value in the idea. This ensures that every idea that is evaluated is positively received, maintaining the safe climate created by the initial suspending judgement. There is no risk of rejection in expressing any idea. The second is the conversion of the one major negative into a ‘direction for improvement’: ‘what I need now is a way to . . .’, invit-
Volume 12
Number 1
March 2003
26
CREATIVITY AND INNOVATION MANAGEMENT
ing ideas directed at this new target. By continued iteration between the ideas and their constructive evaluation, the process moves (probably) to a course of action that is new and feasible and has the commitment of the individual who has been doing the directing (Nolan, 1989). (The process should not be confused with the traditional, binary screening of ideas as good and bad, or even classifying them as ‘plus, minus, interesting’. There are no negatives in Synectics – except perhaps for the inelegant label ‘itemised response’, which is given to this method of evaluation.) The responsibility for the direction of this process goes to the individual who will be responsible for implementing the solution, eliminating at a stroke the ‘not invented here’ syndrome, the unwillingness of people to implement new ideas unless they have played a major part in inventing them. In the Synectics model, that person, alone, switches into a convergent mode of thinking, after the divergent phase; everyone else continues to suspend judgement, but focuses their thinking on the ‘needs for improvement’ being articulated by the problem owner or ‘client’ (the label originally used, following Carl Rogers’ ‘client-centred’ therapy (1961)). The enthusiasm and commitment required for successful implementation of new ideas comes from the emotional ownership generated by the process of producing the invention and the responsibility for deciding which routes to explore and which solution to adopt. If I feel it is my idea (and it genuinely is) then I will make sure it works. Many brainstorming/lateral thinking sessions come up with exciting new solutions and then look for someone to implement them. (They remind me of the Aesop fable about the mice: to stop the depredations of the household cat, they came up with the solution of tying a bell round the cat’s neck. But then the question came up, ‘Who will bell the cat?’.) And if someone is ‘tasked’ to implement the solution, they are unlikely to do so with the commitment and enthusiasm needed to make a new solution successful. I mentioned earlier that its origin in invention is also a source of weakness for synectics. The standard nine-step model (as published on the Synectics website, www.synecticsworld. com) starts from the assumption that there is a problem owner looking for help from others to solve a problem they cannot solve satisfactorily on their own. As Parker (1975) points out, to make such an admission is seen as a confession of failure in the culture of most organizations. If the organization has ‘problems’ they are ascribed to the shortcom-
Volume 12
Number 1
March 2003
ings of others (‘the hole is at the other end of the ship’). Managers solve their own problems privately (perhaps in consultation with a trusted colleague or external consultant) and arrive at what they consider to be the best solution. They then try to ‘sell’ their solution to their colleagues, whose support they will probably need. These colleagues, not having shared in the problem-solving process, will probably want to explore alternatives; ‘Have you thought of . . . ?’, ‘Why not . . . ?’. The solution owner will tend to hear these questions as a challenge and will begin to defend his solution. A debate ensues, using time and energy that could be better used to build a consensus by either improving the proposed solution, or inventing a better one. So the opportunities for the direct application of the standard synectics model tend to be fairly limited. However the tools which make up the standard model can be readily adapted to the situation as it exists within the organization, in this case to recognize that the problem owner is actually a ‘solution owner’. The proposed solution (or ‘best current thinking’ in Synectics language) can be subjected to a constructive evaluation to identify the common ground and also the ways in which it needs to be improved. The creative techniques can then be directed to inventing the necessary improvements. Alternatively (or additionally), the original solution can be set aside (temporarily) while the group addresses the original problem to see if better solutions can be devised. Either way, the group are engaged in a co-operative, win-win process, rather than an adversarial, win-lose debate. And, in this situation, Synectics has evolved into a conflict resolution model as well as an invention model (Nolan, 1990, 2002). The same principles are also applied to deal with the wide spread of responsibility for decision-making and action within organizations. To apply Synectics effectively calls for clarity about ‘who does what’ in the organization; establishing that clarity is often a useful step in itself. The key principle of the client/problem owner concept is that each individual is responsible for what they do individually, within the framework of agreed strategies and policies (which may be decided at a different level of hierarchy). Each individual is also responsible for ensuring that their actions do not cause problems elsewhere in the organization. The Synectics model provides a framework for ‘creative consultation’ on the lines of the conflict resolution model described above.
© Blackwell Publishing Ltd, 2003
WHATEVER HAPPENED TO SYNECTICS?
In fact, Synectics is a powerful alternative to the argument and debate which wastes so much time and energy in organizations of all kinds (Nolan, 2002). Equally, the elements of the Synectics toolkit can be reshuffled to deal with other, not explicitly creative, situations. For example, the Synectics agenda meeting is a model which demonstrably reduces the time spent in conventional meetings as well as improving the quality of the output and the satisfaction of the participants (‘we achieve twice as much in half the time’, they say) (Nolan, 1989). The itemised response tool can be used to ensure that feedback of any kind is constructive in its effect as well as its intent (in, for example, an appraisal situation) (Nolan, 1989). And so on . . . So what, then, is Synectics today? Precise definition of such a rich and evolving body of knowledge is difficult, but here is my ‘best current thinking’: Synectics is a set of process tools derived from video analysis of the methods used successfully in a variety of situations. The tools may be used in a specific sequence (as in the original Invention Model) or individually according to the needs of the situation, resulting in a variety of meeting models and techniques for enhancing personal effectiveness It follows from such a broad definition that Synectics will overlap with other bodies of knowledge. It defies the urge to pigeon-hole – perhaps because it is an eagle, not a pigeon that can be picked up by the academic radar screens.
References Ceserani, J., and Greatwood, P. (1995) Innovation and Creativity. Kogan Page Ltd, London.
© Blackwell Publishing Ltd, 2003
27
Dyson, J. (1998) Against the Odds: An Autobiography, Trafalgar Square, North Ponfret, Vermont. Gordon, W.H.H. (1960) Synectics, The Development of Creative Capacity. Harper & Row, New York. Hicks, M.J. (1991) Problem Solving in Business and Management: Hard Soft and Creative Approaches. Chapman & Hall, London. Martin, J. (ed.) (1992) Creative Management Course B882, Video 2 and Media Book, Open University, Milton Keynes. Nolan V.C. (2002) Creativity: the Antidote to the Argument Culture in proceedings of The Conference on Creative and Cultural Diversity, Halifax: The Creativity Centre. Nolan, V.C. (ed.) (2000) Creative Education: Educating a Nation of Innovators. Synectics Education Initiative, Stoke Mandeville. Nolan, V.C. (1989) The Innovators Handbook. Sphere Books Ltd, London. Nolan, V.C. (1981) Open to Change. MCB Publications Ltd, Bradford. Osborn, A.F. (1953) Applied Imagination. Charles Scribner’s Sons, New York. Parker, R.C. (1975) Creativity: a case history. Engineering. February. Parker, R.C. (1982) The Management of Innovation. John Wiley & Sons Ltd, Chichester. Prince, G.M. (2002) Your Life is a Series of Meetings . . . Get Good at Life. 1st Books Library, Weston MA. Prince, G.M. (1981) A Presentation made to the staff of Synectics Ltd. Prince, G.M. (1970) The Practice of Creativity. Macmillan, New York. Rogers, C. (1961) On Becoming a Person. Houghton Mifflin, Boston.
Vincent Nolan is at the Synectics Education Initiative, 13 Marsh Lane, Stoke Mandeville HP22 5VZ, UK.
Volume 12
Number 1
March 2003
28
CREATIVITY AND INNOVATION MANAGEMENT
Synectics: Reflections of a Little-s Practitioner Tudor Rickards Vincent Nolan provides a valuable update on the body of knowledge known as Synectics. The contribution deserves attention as an account of a fascinating body of knowledge of a method seeking deliberately to stimulate creativity, particularly in small, facilitated groups. He also indicates a lack of awareness of such practical work from academic establishments. In this response I offer another perspective, from a practitioner of creative problem-solving techniques based in an academic environment.
C
reativity and Innovation Management began over a decade ago with the declared intention of acting as a bridge between the academic and practitioner communities. Since that time, co-editors Susan Moger and myself have found it a tough challenge to achieve. Vincent Nolan’s article is the type we rarely came across, and yet was the kind we hoped to publish, in as much as it did seek to cross the academic practitioner communication divide.
The personal and the public As a long-standing supporter of Synectics, ‘from the outside’ I welcome Vincent Nolan’s fascinating article updating the reader on Synectics from the perspective of a highly experienced practitioner. It leaves some scope for additional comments from the perspective of someone who would, in Vincent’s terms, be a practitioner of little-s synectics (a term distinguishing people like myself from the ‘accredited personnel’ who may claim to be applying capital-s Synectics). This is a distinction with important practical implications. There are indeed quality and intellectual property reasons for preserving the difference. Unfortunately, the vocabulary hints at a situation in which the flow of ideas in both directions is not as strong as it might be. Most academics are interested in the subject in a way which favours its examination ‘from the outside’. There have been a few careful studies of the dynamics and outcomes of such for-
Volume 12
Number 1
March 2003
mally executed trials – for example in order to understand the ‘why’ of aspects of group dynamics, or perhaps of the metaphoric mode of ideation in problem solving. These have tended to be in the nature of little-s synectics studies. Quite correctly, those who have been trained – the capital-s Synectics practitioners – may challenge the legitimacy and relevance of results to their work. This is an issue for other management and educational practitioners. Edward de Bono has always made efforts to exercise quality control over those who claim to be marketing Lateral Thinking (or lateral thinking) methods. One could cite other examples. One of the consequences is the emergence of practitioners offering alternatives to the earlier technique systems, often rather reminiscent of the original, with a few modifications. Sometimes these are deliberate attempts to ‘beat the system’. There may also be other motives. The later entries into the market may feel there are no intellectual property problems, as they all can be traced back to a much early body of knowledge. From our experiences, the successful introduction of a new technique is akin to introducing any innovation. It often begins with a highly committed innovator. Indeed, if the idea was the brainchild of two ‘parents’ there may often be a messy divorce, with each parent seeking custody, and the right to bring up the child according to his or her principles. This fierce sense of ownership has other implications, as the parent also seeks to introduce the idea/technique into the most
© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
SYNECTICS: REFLECTIONS OF A LITTLE-S PRACTITIONER
favourable circles. Yet this risks another set of ways of losing control over the child’s development. In less metaphoric terms, there is often a tension between the understandable proprietary feelings of the initiator, and the aspirations to seek widespread approval. The article by Vincent Nolan set me thinking about my own introduction to the youthful synectics technique.
My early experiences My earliest contact with Synectics practioners occurred during a period when I was working in a new products group, in the Unilever organization, as far back as the 1960s. One of my colleagues (Dr Peter Ellington) had learned about an exciting method of stimulating ideas for new products, called Synectics. He contacted one of the pioneers, George Prince, who by then had left the Invention Design Group of Arthur D Little, and was busy developing the company Synectics Inc. His original partner, at ADL, Bill Gordon, had, by then, formed a separate company, Synectics Learning Systems Inc. George paid several visits to Unilever, to explain, and to offer preliminary training in his methods. George probably had as high-voltage an impact on the Unilever researchers as he had on many other audiences. I suspect his influence is particularly potent in small groups. Then, (and for many years afterwards) my impression was of a man of enormous authenticity in his methods of encouraging others to release their creative potential. He gave us new-product technologists a glimpse of his methods, applying them to laboratory problems of an applied chemistry kind. We learned the basic principles he had been developing to permit teams to work in a more constructive fashion. We also learned some principles and practices for escaping previously imposed assumptions, and for discovering new insights. He made the principles sound easy enough, but so are the principles behind the world’s great religions. When we tried our own creativity-stimulating efforts we become more aware of the practical difficulties involved. That is not a criticism of the approach – merely an indication of the gap between a beginner applying a new approach, and an expert operating the approach.
‘Something borrowed, something new . . . ’ Enthusiasm for Synectics persisted in Unilever for many years. However, shortly after
© Blackwell Publishing Ltd, 2003
29
meeting George Prince, I left the company, and began what was to be a career-long association with an international business school. My main work was to support the training of R&D managers. I was keen to introduce some of my industrial experiences, although at first I had little influence on the timetables of the various MBA programmes. I also discovered painfully that creativity did not have as much credibility among many of my new colleagues as I might have expected. For some years I worked mainly on the specialist programmes for technical managers. At the time (the 1970s) there were various competing approaches for stimulating creativity and innovation. A stream of visitors arrived at Manchester, often with their own ‘recipes’. Our American visitors included Sid Parnes and colleagues from Buffalo – who introduced us to the internationally wellknown creative problem solving stage model, that had evolved from Alex Osborn’s original brainstorming approach – Sid Shore, Will McWhinney, and Stan Gryskiewicz (who brought with him Don McKinnon, the doyen of creativity research). Later, the European Association for Creativity and Innovation brought a flow of creativity researchers and teachers from all over the world. These included Edward de Bono, Mark Brown, Jan Buijs, Patrick Coleman, Helga Hohn, Horst Geschka, Margie Parker and dozens of others whose contributions are to be found in the five volumes of proceedings of the Association. To return to my earlier metaphor, I and my colleagues were being regularly being introduced to the latest young talent or brainchild, often with the proud parent – so many brilliant and energetic workers each with something unique to contribute. When the time came for me to introduce creativity training into our own business school programmes I was overwhelmed for choice, yet confined by the proprietary nature of the most attractive offerings. I still considered Synectics as a promising candidate, appropriate for our needs. However, at the time, as is the case today, it was difficult to work out a deal between a university and a training consultancy. Issues include the formal quality control systems, examination procedures and so on, within the academic institute. Like others in my position, I settled for developing a hybrid system for structuring project work in business school applications. The approach owed something to a large number of these influences. From Buffalo, I borrowed the basic ‘diverge-converge’ sequence across several stages in problem solving. I introduced the affirmation and
Volume 12
Number 1
March 2003
30
CREATIVITY AND INNOVATION MANAGEMENT
idea-building technique ‘Yes And’ from Sid Shore (I could easily have borrowed a related approach from Edward Do Bono, or from Synectics – I just happened to be more influenced by Sid Shore’s work, via conferences we attended). I also worked up modifications to metaphoric methods that owed much to Bill Gordon, a little to George Prince and something to Gareth Morgan (1986). As far as possible I have tried to acknowledge these intellectual debts in my teaching and writings (see for example, Rickards, 1990). Vincent talks with some emphasis about the communications gap between commercial and academic communities. My point is, that the decision to become fully engaged with synectics was not simply because Synectics had not appeared on my radar screen. There were institutional pressures that made it difficult for me the take on the role of a committed academic Godparent.
The charismatic parent: the cases of George Prince and Monty Roberts One feature of many of the techniques was the presence of a powerful advocate. Often the advocates were highly charismatic and powerful platform performers. Yet I often felt that the performance demanded too much to be taken ‘in good faith’. Among them, two names now seem to me to have had a remarkably similar and powerful effect on me. One was George Prince of Synectics. It seems worthwhile to comment briefly a comparative way on George and on the inventor of a rather different kind of system for creatively dealing with complex behavioural issues. That second person is Monty Roberts, who was portrayed by Robert Redford, in the film The Horse Whisperer. While engaged in their work (and I suspect also in their personal lives), both George and Monty convey a rare, almost paradoxical impression of nonthreatening competence. In George’s case, it involves working with people, with the intention of proving help and support to a designated ‘client’. In Monty’s case his methods involve working one-on-one, but his designated client is as likely to be a horse as a human. Nevertheless, Monty treats each creature with the deepest sense of dignity. It seemed perfectly consistent with such a style for Vincent to identify some of the Synectics vocabulary with Carl Roger’s and his client-centreed methods. Rogers has given us the notion of unconditional positive regard for the person with whom we are working, which permits easier mutual efforts to address problematic behaviours or beliefs.
Volume 12
Number 1
March 2003
George and Monty (and of course Carl Rogers) show that they, too, are deeply concerned about the ethical development of the person. They provide what Susan Moger and I (Rickards & Moger, 1999) have described to our students as an invitational style of leadership.
Vincent brings Synectics to the UK At roughly the time Unilever was becoming interested in the ideas of George Prince, so was Vincent Nolan. Vincent did not mention his own central role in bringing Synectics to the UK, operating out of his own company, Abraxas. Abraxas, and later generations of Synectics practitioners, are in his debt. Vincent went on to produce his own textbook (mentioned in his article), and to help spread awareness of Synectics, not least through the Open University Masters in Business elective, The Creative Manager. Some of those he enthused were senior industrial figures like the technical manager Dr Charles (Reg) Parker, who had already begun an early archive of creativity literature in the UK, in 1970s. It is good also to see the gradual (if painfully slow) interest shown in such matters also from university departments.
Understanding systems for stimulating creativity Vicnent’s initiatives with Abraxas coincided with a shift in my career from an inventor of new soap products, to a teacher and researcher into the processes of invention. To my regret, my contact with George and Vincent became cursory. I increasingly based my understanding on the documents available to outsiders, and in particular the two early books by Gordon (1961) and Prince (1970). There is actually an important issue here, which I do not think I successfully resolved. How was I to study and understand the various systems for stimulating creativity? These included: Synectics; lateral thinking; variants of Creative Problem-Solving being developed by Sid Parnes and his people at Buffalo, and Stan Gryskiewicz and colleagues at Greensboro, NC. Within realizing, I resorted to what some of my fellow academics would call triangulation. I tried to seek comparative judgements from across the various fields. What is similar and what is different? Over which aspects of their approaches over do the practitioners appear to be in agreement; over which do they appear to be in disagree-
© Blackwell Publishing Ltd, 2003
SYNECTICS: REFLECTIONS OF A LITTLE-S PRACTITIONER
ment? These views can be found in Rickards (1999). It seems to me that the technique systems share the implicit or explicit belief that it is possible to influence the creative process, for better or worse, though the structures or systems of our working lives. Furthermore, the most robust systems are those, like Synectics, which evolve through the interplay between emerging bodies of theory and practice. What we have is a series of related belief systems whose practices have at least some superficial differences. To resort to metaphor, the systems may be likened to beliefs and practices of religious groups, of varying degrees of fundamentalism. You chose ultimately though a leap of faith based on what makes sense to you, influenced by what the advocates proclaim, what they promise and what they expect you to do. Signs of nonfundamentalism are indications that the practices (but not necessarily the beliefs) are ‘moving with the times’.
Synectics – from the outside Taking these considerations into account I have concluded, that from the outside, Synectics appears to be, as Vincent describes it, ‘a rich and evolving body of knowledge’. Its unique historical features were those powerful operational mechanisms for stimulating creativity though various analogical methods. Plenty of people since have used variants of the original; few have combined their approach with empirical and theoretical grounding. (Gareth Morgan, a possible exception, may have gone out of his way to conceal such a grounding, in his influential book, Images of organization, Morgan, 1986.) Synectics is also highly unusual in holding on to the behavioural aspects of group dynamics while also seeking to establish specific structural features to support group behaviours stimulating creativity. In this, it seems to have stronger claims over some other systems which often appear to place faith in creativity specials – a multiplicity of special
© Blackwell Publishing Ltd, 2003
31
techniques, each seeking to produce a leap of insight.
An integrative suggestion Finally, I would like to approach one aspect of Vincent’s article in the spirit of creative conflict resolution. I agree that there are misunderstandings of the nature of Synectics by those interested but mostly occupied in academic teaching and research. That is a justification for the original article. I would argue that there are also misunderstandings among busy practitioners of what academics under different circumstances might be able to contribute. For example, academics try to collect and weigh up evidence in a pretty systematic way that would augment the more informal networks within which practitioners tend to operate. My deeply held view is that academics and practitioners have everything to gain from dialogue and communication. This journal, and the more diverse kind of conference try (sometimes with more success than others) to enable that communication. Vincent’s article is further step in the direction towards better communication and understanding.
References Gordon, W.J.J. (1961) Synectics: The development of creative capacity. Harper & Row, New York. Morgan, G. (1986) Images of organization. Sage, Newbury Park, CA. Prince, G.M. (1970) The practice of creativity. Harper & Row, New York. Rickards, T. (1990) Creativity and Problem-Solving at Work. Gower, Farnborough, UK Rickards, T. (1999) Creativity and the management of change. Blackwell, Oxford. Rickards, T. and Moger, S.T. (1999) Handbook for creative team leaders. Gower, Aldershot.
Tudor Rickards is Professor of Creativity and Organizational Change at Manchester Business School, United Kingdom, email:
[email protected]
Volume 12
Number 1
March 2003
32
CREATIVITY AND INNOVATION MANAGEMENT
The Creative Product and the Creative Processes in Virtual Environments Per Kristensson and Torsten Norlander The effects of information technology on the creative performance in groups and individuals were examined. An experimental 3 ¥ 2 design was used in order to assess the effects of computer-mediated and face-to-face interaction on creative performance. Forty-eight groups, divided into individuals or small three-person groups, worked on a heuristic task during three sessions. Creativity was operationalized in terms of the creative product and the creative process. The results regarding the creative product indicated that small groups had a higher flexibility than individuals. The small groups also proved to have significantly more incubations. Regarding the creative process, the face-to-face group showed a better flow, and this was especially valid in respect of the preparation phase. This study suggests that, providing there is a fully understood and well-prepared problem, attained or strengthened through face-to-face interaction; GCSS may serve as an important arena for exchanging and developing ideas.
Introduction
T
he quality of new products and services is heavily dependent on the creative performance of individuals working alone or in teams. Both normative and empirical research suggests that teams of individuals should be assigned to work with the development of a new product or service, from idea generation to market launch (e.g. Brown & Eisenhardt, 1995). However, contrary to what one might expect, creativity research over the past decade has indicated that creative efforts made by groups are not always as successful as suggested or imagined (e.g. Diehl & Stroebe, 1987, 1991; Pinsonneault, Barki, Gallupe & Hoppen, 1999; Stroebe & Diehl, 1994). Several studies have, as a matter of fact, revealed that groups can be outperformed by an equal number of non-interacting individuals, often referred to as nominal groups (e.g. Dennis & Valacich, 1993). Psychological process losses such as evaluation apprehension, free riding and production blocking account for a great deal of explanation for these intuitively rather contradictory findings (Pinsonneault et al., 1999). Interestingly, research indicates that the weaknesses of groups might be surmounted by the use of a
Volume 12
Number 1
March 2003
group communication support system (GCSS) (Dennis & Valacich, 1993; Hollingshead, 2001). A GCSS is a technology whose functionality provides or modifies within-group communication (Hollingshead, 2001). Today, the use of GCSSs is proliferating in organizations – among other things it has become a widely recommended tool for companies that attempt to develop a new product or service. For example, a GCSS may play an important role when a company attempts to tie together the information from customers with information about financial opportunities, technological requirements and the strategy of the senior management. The development process of new products and services has proved to be a complex and iterative process and is often performed by groups of people. This makes the communication between the parties involved crucial to success. However, product development is not just any organizational work process; it is a process highly dependent on creativity (Brown & Eisenhardt, 1995; Clark & Wheelwright, 1994). Creativity is a unique and complex human capability that, on the group level, is tightly interwoven with communication. According to researchers (e.g. Tidd, Bessant & Pavitt, 2001) many problems in the innovation processes occur due to failures in communication. Since problem solving to a
© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
ELECTRONIC INTERACTION AND CREATIVITY
large extent depends on the possibility of combining different sets of knowledge, which often may be distributed multi-directionally (up, down and lateral) in the organization, developing mechanisms for communication across different interfaces stands out as critical to innovation success. The approach taken in this study is consistent with the input-process-output paradigm (e.g. Connolly, Jessup and Valacich, 1990; Pinsonneault et al., 1999; Taggar, 2001); the input is the creative composition of the group, the process is indicated by group creativitystifling and creativity-enhancing behaviour, and the output of concern is the creative performance of the group. This framework makes it possible to predict, evaluate and design experiments investigating the effects of GCSS on creativity. According to Pinsonneault et al. (1999) creative performance in group interaction can be either enhanced or reduced with respect to quality and quantity. Building on Osborn (1957), it is inferred that the group, on the positive side, can provide encouragement, stimulation, motivation or social recognition, inspiring creative contributions (see also Connolly, Jessup and Valacich, 1990). Factors that increase and enhance groups’ idea generation productivity are referred to as process gains. On the negative side there are process losses, factors that decrease and inhibit groups’ idea generation (Pinsonneault et al., 1999). Working in a group might inhibit contributors expecting embarrassment, hostile evaluation, conformity pressures or other negative events from proposing an unusual idea. Furthermore, several researchers have also described the mechanics of the combining-communication process as an obstacle to a group, referred to as production blocking (e.g. Diehl & Stroebe, 1991; Lamm & Trommsdorff, 1973). A distinguishing feature with the research conducted on creativity in GCSS environments is that it is commonly based on measurement of ideational fluency, i.e. the quantity of ideas (Paulus, 2000; Rickards, 1999). However, the type of creative activity that is undertaken in a complex, iterative and uncertain product-development project involves more aspects of creative performance than just the ability to come up with a number of ideas. The idea also needs to be a qualitatively unique answer to existing customer needs – an original and valuable idea – and because ways to find such ideas have proved complicated, the question of how the creative process is affected also needs to be addressed. After all, since previous research has focused on processes in terms of gains and losses, research attempts to measure not only the outcome from such processes, but also the measure-
© Blackwell Publishing Ltd, 2003
33
ment of the process itself, seem justified. Typically, in creativity research, the creative process is not addressed (Kay, 1994; Nemiro, 2002). Therefore, in the present paper creativity is defined from a product and process perspective. The product is assessed not only in terms of fluency but also in terms of flexibility, originality and elaboration. Fluency represents a quantitative aspect of divergent thinking together with flexibility while originality and elaboration represents qualitative aspects of the creative product. The process is assessed through Wallas’ (1926) four phases; preparation, incubation, illumination and verification, and provides information on how creative performance was liberated or limited. In addition to the unanswered question about the effects of information technology on creative performance in groups, the question of how groups differentiate from individuals regarding creative performance is also addressed. The research question dates back to the days of Osborn (1957), who suggested that communication among members in a group would provide process gains from their stimulating collaboration. However, taking a closer look at what experimentally designed research conveys about group performance regarding problem-solving capacity, the hypothesis from Osborn may appear incorrect (see Rickards, 1999 for a discussion on this subject). When scrutinizing these studies, certain features appear that can explain the contradictory results (Ekvall, 1988). In the experiments where individuals appear as superior, the results from groups have been compared to individuals that afterwards have been “grouped” into a group result (i.e. nominal groups). Furthermore, another common feature is that the tasks given have been of a very general kind. All participants have had enough knowledge and experience to contribute and understand the problem, and thus, have been able to contribute to a final solution. Yet, this is a reality and premise that does not often exist outside the research laboratory (Rickards, 1994). The reason for group and team work in organizations is that many problems are of such a complicated nature, and entail such time limitations, that it requires a group of people with complementary expert knowledge to be able to bring them to a successful close. Also, in an organizational setting, such as in a product development project, the tasks are complex, varied and often include problem identification, idea generation and consensus decision in order to reach a final solution (Taggar, 2001). Considering that the creative performance demanded in a product development situation calls for something more than just a large
Volume 12
Number 1
March 2003
34
CREATIVITY AND INNOVATION MANAGEMENT
number of ideas, and that the task in real-life situations may be more complex than tasks given in laboratory research settings, it seems hasty to take the side of the group or the individual generally, as the most efficient creative problem solvers. Instead, reflecting on the discussion above, it is important to examine how groups perform in comparison to individuals using additional measures of creativity and employing an open-ended task requiring elaboration over a couple of days.
Method Participants The experiment was conducted at the Cyber/IT Research Laboratory of Karlstad University (Sweden). Ninety-six individuals, 48 men and 48 women, participated. All were students recruited on the campus and efforts were taken to secure equal representation from both sexes in order to check for possible gender factors. Participants were randomly assigned to each of six experimental groups which together constituted two independent variables (see the “Design” section below). The mean age for the entire population was 24.51 years (SD = 4.64, range = 19 to 39). There were no (three-way ANOVA) significant differences between groups or between sexes with regard to age and number of academic terms spent at Karlstad University (ps > 0.05). Two personality inventories were administered to all participants in order to provide further background information. One test, the FS (change and stability) test, measured attitude-to-creativity with respect to change and stability (Holmquist, 1986). A three-way ANOVA showed no differences between the groups and no difference between the sexes, and there was no interaction (ps > 0.05). The other test, the life orientation test (LOT), measured dispositional optimism (Scheier & Carver, 1985). A three-way ANOVA showed no differences between the groups or between the sexes and there was no interaction (ps > 0.05).
Design Participants were randomly assigned to six different experimental groups, namely chat/individual, chat/group, videoconference/individual, videoconference/group, face-to-face/individual and face-to-face/ group. In this way two independent variables were obtained: IT group (chat, videoconference or face-to-face) and assemblage
Volume 12
Number 1
March 2003
(individual or group). Internet-based software (Microsoft NetMeeting) was used to represent a text-based chat form of communication. In a text-based chat only written information can be transmitted (here referred to as Chat). The software for this condition shows resemblance to similar studies (e.g. Sosik, 1997), restricting the communication interaction among participants. Another sort of Internet-based software (Whitepine Cu-SeeMe), allowing interacting individuals to see each other while communicating, was used to represent a second form of computer-mediated communication (videoconference). In the videoconference it was additionally possible to use a chat-function. Finally, the two kinds of information technology above were compared with groups that interacted verbally face-to-face. There were three people in each group and only one in the individual condition. The outcome of the dependent variable was collected through tape recordings from two meetings, one original meeting and one final Meeting, together with a notebook all participants were instructed to use.
Instruments Panels Five panels of scorers were constituted: panel A, B, C, D and E judged the creative product from Guilford’s dimensions of creativity (i.e. fluency, flexibility, originality, elaboration). The judges in all panels were doctoral candidates in psychology at the university, with the exception of panel C which judged originality and whose members were professional journalists at a newspaper. Panel E was given the task of identifying indicators of the stages of creative thought (i.e. preparation, incubation, illumination) at the original meeting, in the diary and at the final meeting. All ten judges performed all scoring completely independently of each other (see Amabile, 1996). Self-Reported Measures Besides the measures of creativity, subjective measures were collected regarding how participants perceived the process and the product. The participants were asked to fill out a short questionnaire about their perceptions of the communication process and the outcome of the group. A VAS scale, ranging from 1 to 10, was used for this objective and the individual scores were combined into a group total (ranging from 0 to 30 for process; from 0 to 60 for outcome).
© Blackwell Publishing Ltd, 2003
ELECTRONIC INTERACTION AND CREATIVITY
Procedure The experimental session was composed of three phases consisting of two meetings – an original and final meeting – with a thought period in between (cf., Norlander & Gustafson, 1996). The laboratory consisted of three separate rooms, each named after major cities in order to create a feeling and understanding of being geographically dispersed. In the face-to-face condition the group used a circular table. The Original Meeting On arrival at the laboratory for the original meeting, participants were instructed to fill out a questionnaire concerning background information, FS and LOT. Each participant then received a complex problem and was instructed to work with this in the group in one of the three conditions. The original meeting lasted for seven minutes and the time for both the original and final meetings was estimated on the basis of a pilot run. The aim of the whole experimental procedure was to try to emulate a typical project sequence early in a product or service development project. Consequently, the problem concerned how the university could improve the quality of its courses, a familiar and important topic for students. The Diary When the original meeting was concluded, all participants were presented with a small pocket-sized notebook (specially made for the occasion). In the notebook, the problem was presented, along with instructions for how to use it.1
35
The diary phase was motivated by of reallife settings, where members of a product development team are expected to elaborate upon problems and their solutions in different situations over a period of time. The Final Meeting At the final meeting the participants wer instructed to outline in detail a plan considered optimal to solve the problem given at the original meeting. The final meeting lasted for ten minutes. Also, at the final meeting, data on participant satisfaction was collected. In order to make the experimental setting more realistic, all participants were informed that the three best solutions would be rewarded with a cash prize of $100.
Scoring The Creative Product (Panel A, B, C and D) The judges of panel A, B, C and D were told to rate every participant in accordance with a selected dimension of creativity. The dimensions were scientifically derived from Guilford (1967). The dimensions measure divergent production and were fluency (number of ideational responses), flexibility (number of ideational responses within separate categories), originality (the newness of an idea) and elaboration (the amount of detail in ideational responses). Because the assessment was time-consuming there was one set of judges for every dimension; altogether eight different judges were used. Each judge was instructed to rate the finished solutions from the final meeting on a scale from one to ten. The Creative Process (Panel E)
1
‘We want you to think of this problem from now on and until after the experiment is over. Write down in this notebook every idea, notion and thought for solution of the problem that occurs for you. Carry this notebook with you wherever you go, and whenever you get an idea concerning this problem, just make a note of it, no matter how irrelevant it may seem. It is not necessary to write much, only a few sentences at a time. We would rather have more frequent and shorter memoranda than a single long report. Also, write down where you think your idea originated (for example, was it your group discussion that stimulated you, or was it perhaps an association you had when you saw or heard something?). Don’t concern yourself with the absurdity of what you make a note of or read anything peculiar on the subject or discuss it with others. We just want your own ideas.’ (Adapted from Patrick, 1938; Norlander & Gustafson 1996).
© Blackwell Publishing Ltd, 2003
The three reports (original meeting, diary and final meeting) were each typed on a separate slip of paper designated merely by number. The papers from the three chat, videoconferencing and face-to-face groups were, of course, shuffled. The scoring was carried out by trained judges according to a written instruction, to a large extent adapted from Patrick (e.g. 1938) and Norlander and Gustafson (1996). The panel reported the frequency and localization of assertions (AS), thought changes (TC), incubations (IC) and illuminations (IL). AS and TC are connected with, and represent, the preparation phase, IC the incubation phase and IL the illumination phase. The four categories/indicators were defined in the following way:
Volume 12
Number 1
March 2003
36
CREATIVITY AND INNOVATION MANAGEMENT
• AS – the assertion that the group had no ideas (i.e. “this is difficult”, “we seem to be trapped in a corner here”); • TC – any modification of thought which was sufficient to cause formation of a new sentence (e.g. “e-mail information to students could perhaps be something . . .”); • IC – the spontaneous recurrence of a notion with more or less continual modification while other topics were under consideration; • IL – the period when the plan was distinctly formulated by the group. The three reports were regarded as descriptions of one process. This means that an IC that was noticed in the diary could have its origins in the original meeting and so forth.
Results Interjudge Reliability Panel A, B, C and D The correlation statistics (Pearson’s r) showed a significant correlation regarding fluency (r = 0.97, p < 0.001), flexibility (r = 0.65, p < 0.001), originality (r = 0.63, p < 0.001) and elaboration (r = 0.74, p < 0.001) between judges. In order to enable further statistical analysis the judge’s ratings were combined into a composite for creativity (O’Quin & Besemer, 1989). Panel E The two meetings and diary of each group were also analysed separately by two independent judges for the number of assertions (AS), thought changes (TC), incubations (IC) and illuminations (IL). The correlation statistics between the two judges showed significant correlation for all variables, with the highest correlation for the number of thought changes in the diary (r = 0.91, p < 0.001) and
assertions at the original meeting (r = 0.86, p < 0.001). Most difficult to judge was the number of thought changes at the final meeting (r = 0.51), however also significant (p < 0.001). Of great interest for the investigation was the correlation of incubations at the final meeting, which showed good agreement from both judges (r = 0.81, p < 0.001).
Dependent Variables 1. The Creative Product A Pillais’ MANOVA (3 ¥ 2 factorial design) was used with the IT group and assemblage group as independent variables, and fluency, flexibility, originality and elaboration as dependent variables. The analysis showed no significant interaction effect (p = 0.481), no significant difference with regard to the assemblage group (p = 0.087) and no significant difference with regard to the IT group (p = 0.635). Univariate F-tests did however indicate a significant result for flexibility with regard to assemblage [F(1,42) = 6,82, p = 0.012] where the participants in the group condition scored higher on flexibility as compared to participants in the individual condition. For means and standard deviations see Table 1. 2. The Creative Process The AS and TC served as indicators for the preparation stage and were combined. According to the judges there were no participants with any illumination at all. It is of interest to see the distribution of the four indicators of the creative-thought process with regard to the original meeting, the diary and the final meeting. The preparation stage (AS + TC) clearly are most frequent in the original and diary sessions, while the most incubations and 94 per cent of the illuminations were noted at the final meeting (see Table 2). These findings are well in accordance with
Table 1. Means and (standard deviations) for the creative product concerning the IT group Chat
Fluency Flexibility Originality Elaboration
Volume 12
Number 1
Videoconference
Face-to-face
Individual
Group
Individual
Group
Individual
Group
9.00 (4.54) 1.44 (1.32) 4.38 (2.86) 4.56 (2.77)
7.06 (3.34) 2.62 (1.30) 5.88 (2.84) 5.88 (2.47)
4.31 (4.24) 1.88 (0.83) 5.00 (3.21) 4.88 (2.34)
6.81 (5.79) 2.00 (0.60) 3.88 (1.55) 6.31 (2.94)
5.38 (5.71) 1.44 (0.50) 4.81 (2.10) 4.25 (2.19)
7.94 (4.17) 2.38 (1.09) 4.38 (2.34) 3.81 (1.46)
March 2003
© Blackwell Publishing Ltd, 2003
ELECTRONIC INTERACTION AND CREATIVITY
Table 2. The distribution of AS + TC, IC and IL, at the original meeting, diary and final meeting (%) Content
Original
Diary
Final
AS + TC IC IL
58.44 35.15 5.21
19.13 12.12 1.04
22.43 52.73 93.75
Note: AS + TC = the percentage of assertions and thought changes, representing the preparation stage, that the group had, IC = incubations, IL = illuminations.
earlier research (Norlander & Gustafson, 1996). Univariate F-tests revealed significant differences regarding both the IT variable [F(2,47) = 4.07, p = 0.024] and the assemblage variable [F(1,47) = 44.44, p < 0.001]. For the IT variable a post hoc comparison (Scheffé) showed that the face-to-face group had more assertions and thought changes than the videoconference group, indicating a more thorough preparation stage. On the assemblage variable the small groups had more incubations than the individuals. There were no other significant differences. 3. Participants Self-Reported Perceptions of Product and Process One-way ANOVA revealed significant differences for the IT condition regarding the process [F(2,47) = 5.26, p = 0.009]. A post hoc comparison (Scheffé) indicated that the faceto-face group had a better process (M = 7.19, SD = 3.05) as compared to both the videoconference group (M = 4.47, SD = 2.34) and the chat group (M = 4.67, SD = 2.48). There was no significant difference between the two computer-mediated groups. Regarding the assemblage variable there was no significant difference (p < 0.06). Nor were any differences found regarding how participants perceived their product.
Discussion This study used an experimental design in order to investigate the effects of electronic interaction and groups versus individuals on creative performance. Overall, the experiment produced three main results: 1. Regarding the creative product, small groups showed a higher flexibility, imply-
© Blackwell Publishing Ltd, 2003
37
ing that groups are able to produce a larger variety of ideas than individuals. 2. Regarding the creative process, the faceto-face group showed significantly better preparation. The small groups produced significantly more incubations than the individuals. 3. Subjective measures regarding participant satisfaction showed that participants communicating face to face perceived their process as more satisfying than the computer-mediated groups. One widely known theory that addresses the effects of using different media is media richness theory (Daft & Lengel, 1986). The central proposition of media richness is that performance is hypothesized to improve when people use richer media to interact on equivocal tasks (Dennis et al., 1999). Unfortunately, virtually all research on media richness has focused on perceptions of media fit, and one of the few studies that exists (Dennis et al., 1999) reports only small empirical support. In this study, the examination of the creative product also fails to validate the theory, while the measurement of the process indicates little support in terms of greater easiness to prepare for creative thought in rich media. One possible explanation for the process data is that motivation and optimism, which appear as important components in a start-up phase, might be more easily communicated in a communication rich face-toface meeting. As an explanation of why these differences diminished later in the process, previous theory has indicated that the task-oriented working-style of computermediated groups sometimes can account for why CMCs might catch up with face-to-face groups (Bordia, 1997). Another explanation, somewhat similar to the above, is that the difficulties in communicating electronically are most likely to appear in the beginning of a sequence (i.e. in the preparation phase) than later, when the participants have learned how to govern the media. Therefore, while face-toface participants have been occupied identifying the problem, CMC participants have been trying to organize communication in general. Furthermore, in media richness theory the task is said to be of importance (Walther, 1996). In this study, the aim was to present a task that required both problem identification and problem solving, thus involving greater creative abilities than in a task requiring only idea generation. However, as there were no significant differences regarding the creative product, the task in this study might not have been equivocal enough to convey the effects suggested by media richness theory. Accord-
Volume 12
Number 1
March 2003
38
CREATIVITY AND INNOVATION MANAGEMENT
ing to Walther (1996) the task has to be both highly equivocal and interpersonally demanding in order to bring about the advantages for rich media. As stated in the introduction, many researchers have proposed computermediated communication as a means for bridging the stifling forces elicited in group interaction. By communicating via a computer network, processes such as evaluation apprehension and production blocking are hypothesized to be reduced and thus, electronic groups are predicted to outproduce face-toface groups, at least in respect of ideational fluency. However, the empirical data of this study reveals no such pattern. The design of the task is assumed to have contributed to the deviation between previous theory and present findings. Participants in this experiment were not explicitly told to generate as many ideas as possible, but to create a wellmerited solution. However, the subjects should then have put more emphasis on qualitative aspects of creative performance, which would result in more flexible, original and elaborated products. Concerning both flexibility and elaboration, for example, theory yields that groups that are not socially inhibited or suffering from production loss should more easily reach a wider span of ideas and a higher level of detail in their responses (Pinsonneault et al., 1999). Recognizing this, the failure to confirm theory may be interpreted thus, that the underlying information-processing pattern that represents creative thinking is to a larger extent affected by other external stimuli than communication arena, which, in this experiment, has been randomly distributed across the groups. In line with this suggestion, Ziegler, Diehl and Zijlstra (2000) have been questioning the validity of the suggested advantages for electronic brainstorming. According to them, the assumption that CMC can function as a means of enhancing idea production is optimistic and although several studies have been carried out, firm conclusions cannot be drawn. With respect to this, future research should concentrate on examining what kinds of process stimulate the more qualitative dimensions of creative production, e.g. originality and elaboration, as these aspects represent unique responses to yet unsolved questions. Regarding the assemblage condition in the present study, groups outperformed individuals on the flexibility dimension. Also, groups showed a significantly higher ability for incubation than individuals. A reasonable explanation for this is that groups have better opportunities to make connections between and within categories. According to Paulus
Volume 12
Number 1
March 2003
(2000), a unique potential benefit of face-toface or electronic sharing of ideas is mutual stimulation of associations. For example, when students problem solve on ways to improve their university, they tend to generate ideas within different conceptual categories. When members are exposed to an idea within a particular category, it will tend to stimulate more ideas within this category since associations commonly follow the rule of similarity. Given enough time a group will have depleted the most obvious associations within a category and then the group members will attempt to shift to more cognitively remote categories. A significant benefit of problem solving in groups, therefore, is the increased probability that the group will come across categories of ideas that one would not have thought of single-handed (Paulus, 2000). Future research should address several limitations of present research. First of all it is necessary to consider the question of external validity as in all laboratory research. The question of task motivation, furthermore, could well be reasoned to have differed across participants, thus threatening internal validity. However, in a random experiment like this, any differences are most likely to be distributed equally across the conditions. What perhaps should differentiate then is the level of task motivation of students compared to product and service developers who, in contrast to the former, are daily dependent on creative performance (Amabile, 1996). Recognizing this nevertheless, it is judged that the level of motivation was at least satisfactory since it concerned a matter that is often debated by university students. If participants did have any lack of interest, this ought to have had the largest consequences in the diary phase, which was difficult to control experimentally. Another limitation concerns the lack of control of gender aspects. Phenomena such as communication, computer usage and knowledge are most likely to differentiate among sexes, and it is therefore unsatisfactory that this variable is left outside control in the present study. Despite these limitations, this study is valuable because it involves an advance toward a fuller understanding of the effects of electronic communication arenas on creativity. The complexity of the notion of creativity particularly has emerged as central in our study. As noted in the introduction, creating new services may become increasingly dependent on the ability to use information technology as a medium for innovation. A managerial implication of this study is that information technology may well serve as an important
© Blackwell Publishing Ltd, 2003
ELECTRONIC INTERACTION AND CREATIVITY
arena for exchanging and developing ideas, especially under circumstances where the initial problem has been carefully prepared, preferably under face-to-face conditions, and when regular face-to-face meetings are made impossible by time constraints and geographical obstacles.
Acknowledgements This study was supported by grants to the Service Research Center from The Swedish National Board for Industrial and Technical Development – NUTEK.
References Amabile, T.M. (1996) Creativity in Context: Update to the Social Psychology of Creativity. Westview, Boulder, USA. Bordia, P. (1997) Face-to-face versus computermediated communication: a synthesis of the experimental literature. Journal of Business Communication, 34(1), 99–120. Brown, S.L. & Eisenhardt, K.M. (1995) Product development: Past research, present findings, and future directions. Academy of Management Review, 20(2), 343–378. Clark, K.B. & Wheelwright, S.C. (eds.). (1994) The Product Development Challenge. Competing through Speed, Quality and Creativity. Harvard Business Review, Boston. Connolly, T., Jessup, L.M. & Valacich, J.S. (1990) Effects of anonymity and evaluative tone on idea generation in computer-mediated groups. Management Science, 36(6), 689–703. Daft, R.L. & Lengel, R.H. (1986) Organizational information requirements, media richness and structural design. Management Science, 32, 554–571. Dennis, A.R., Kinney, S.T. & Hung, Y.C. (1999) Gender differences in the effects of media richness. Small Group Research, 30(4), 405–437. Dennis, A.R. & Valacich, J.S. (1993) Computer brainstorms: More heads are better than one. Journal of Applied Psychology, 78(4), 531–537. Diehl, M. & Stroebe, W. (1987) Productivity loss in brainstorming groups: Toward the solution of a riddle. Journal of Personality and Social Psychology, 53(3), 497–509. Diehl, M. & Stroebe, W. (1991) Productivity loss in idea-generation groups: Tracking down the blocking effect. Journal of Personality & Social Psychology, 61(3), 392–403. Ekvall, G. (1988) Förnyelse och Friktion. Om organisation, Kreativitet och Innovation [Renewal and Friction. Organization, Creativity and Innovation]. Natur och Kultur, Stockholm. Guilford, J.P. (1967) The Nature of Human Intelligence. McGraw-Hill, New York. Hollingshead, A.B. (2001) Communication technologies, the Internet, and group research. In
© Blackwell Publishing Ltd, 2003
39
Hogg, M.A. & Tindale, R.S. (eds.), Group Processes. Blackwell, Malden, MA. Holmqvist, R. (1986) Manual till FS: Förändring och Stabilitet [Manual for the FS-test: Change and Stability]. Psykologiförlaget AB, Stockholm. Kay, S. (1994) A method for investigating the creative thought process. In Runco, M.A., Problem Finding, Problem Solving, and Creativity. Ablex, Norwood, NJ. Lamm, H. & Trommsdorff, G. (1973) Group versus individual performance on tasks requiring ideational proficiency (brainstorming): a review. European Journal of Social Psychology, 3, 361– 388. Nemiro, J.E. (2002) The creative process in virtual team. Creativity Research Journal, 14(1), 6–83. Norlander, T. & Gustafson, R. (1996) Effects of alcohol on scientific thought during the incubation phase of creative process. Journal of Creative Behavior, 30, 231–248. Nunamaker, J.F. Jr. (1997) Future research in group support systems: needs, some questions and possible directions. International Journal of HumanComputer Studies, 47, 657–385. O’Quin, K. & Besemer, S.P. (1989) The development, reliability and validity of the revised creative product semantic scale. Creativity Research Journal, 2, 267–278. Osborn, A.F. (1957) Applied imagination. Principles and procedures of creative problem-solving. The Creative Education Foundation Press, New York. Patrick, C. (1938) Scientific thought. Journal of Psychology, 5, 55–83. Paulus, P.B. (2000) Groups, teams and creativity: the creative potential of idea generating groups. Applied Psychology – An International Review, 49, 237–262. Pinsonneault, A., Barki, H., Gallupe, R.B. & Hoppen, N. (1999) Electronic brainstorming: the illusion of productivity. Information Systems Research, 10(2), 110–133. Rickards, T. (1999) Brainstorming. Encyclopedia of Creativity, vol 1, 21–227. Rickards, T. (1994) Electronic brainstorming: Asking the right question. Creativity and Innovation Management, 3(2), 110–114. Scheier, M.F. & Carver, C.S. (1985) Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychology, 4(3), 219–247. Sosik, J.J. (1997) Effects of transformational leadership and anonymity on idea generation in computer-mediated groups. Group & Organization Management, 22(4), 460–487. Stroebe, W. & Diehl, M. (1994) Why groups are less effective than their members: On productivity losses in idea-generating groups. European Review of Social Psychology, 5, 271–303. Taggar, S. (2001) Group composition, creative synergy, and group performance. Journal of Creative Behavior, 35(4), 261–282. Tidd, J., Bessant, J. & Pavitt, K. (2001) Managing Innovation. John Wiley, New York. Wallas, G. (1926) The art of thought. In Vernon, P.E. (ed.). Creativity. Penguin Books Ltd, Harmondsworth, UK.
Volume 12
Number 1
March 2003
40
CREATIVITY AND INNOVATION MANAGEMENT
Walther, J.B. (1996) Computer-mediated communication: impersonal, interpersonal and hyperpersonal interaction. Communication Research, 23(1), 3–43. Ziegler, R., Diehl, M. & Zijlstra, G. (2000) Idea production in nominal and virtual groups: Does computer-mediated communication improve group brainstorming?. Group Processes & Intergroup Relations, 3(2), 141–158.
Volume 12
Number 1
March 2003
Per Kristensson, Ph D Candidate, Service Research Center, Karlstad University. Torsten Norlander, Associate Professor, Dept of Psychology, Karlstad University. Correspondence should be addressed to Per Kristensson Service Research Center, Karlstad University, SE-651 88 Karlstad, Sweden, phone: +46 (0) 54 700 2128, e-mail
[email protected].
© Blackwell Publishing Ltd, 2003
THE RISKS OF AUTONOMY
41
The Risks of Autonomy: Empirical Evidence for the Necessity of a Balance Management in Promoting Organizational Innovativeness Diether Gebert, Sabine Boerner and Ralf Lanwehr For the purpose of promoting innovativeness in organizations, the literature recommends more decentralization of power and more participative leadership as one dimension of empowerment and thus greater situation control for employees. In fact, however, increasing situation control involves specific risks (including co-ordination problems). Without concurrent integration to cushion these risks through orientation, consensus and trust, increasing situation control therefore leads not to a further increase in innovativeness but to a decline. This first empirical demonstration of the covert curvilinear relationship between situation control and innovativeness in n = 101 organizations reveals these risks. At the same time, it calls into question the widespread recommendations in the literature for action on innovation-friendly organization and leadership.
The Issue
T
here is broad agreement in the literature that innovative organizations tend to be decentralized, to grant employees more decision-making autonomy, and to be characterized by delegatory and participative leadership (Axtell et al., 2000; Damanpour, 1991 p. 581; Glynn, 1996, p. 1102). Innovative organizations thus exhibit attributes of empowerment (Hardy & Leiba-O’Sullivan, 1998, S463) that Burns and Stalker (1961) have identified as aspects of the “organic” organization. Common to the organizational and leadership principles which the literature claims to be important for innovativeness – decentralized self-regulation, the decentralization of decision-making powers, the promotion of decision-making autonomy, the delegation of complex tasks and corresponding decisionmaking rights, and participative leadership – is that situation control by the led increases. The degree to which employees can shape and influence matters increases. They have the © Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
opportunity to try out changes or innovations independently within their areas of responsibility. If they are led participatively, they can also influence their superiors and thus initiate projects for change. To this extent, the following theoretical reflections and empirical findings on the construct situation control are significant both for organizational and leadership studies. Despite repeated reports of inconsistencies in the empirical relationship between variants of increasing situation control and organizational innovativeness (Fiol, 1996, p. 1013f.; Wolfe, 1994, p. 424), greater situation control is generally recommended to increase innovativeness, thus taking the linearity premise for granted: the more situation control, the more innovativeness. We confront this recommendation with the thesis that an increase in situation control systematically causes specific unplanned negative secondary effects (Gebert & Boerner, 1999), which not only impede but frustrate any further growth in innovativeness unless the organization cushions the unplanned negative
Volume 12
Number 1
March 2003
42
CREATIVITY AND INNOVATION MANAGEMENT
secondary effects with an appropriate counterstrategy (integration). According to this thesis, the relationship between situation control and innovativeness is only apparently linear but in fact curvilinear. Curvilinearity, as opposed to linearity, means a specific inverted U relationship between situation control and innovativeness (in mathematical terms, a second-degree polynomial with negatively weighted quadratic term). It is obvious that proof of covert curvilinearity would have far-reaching consequences. From a practical point of view, the usual recommendations to increase situation control would be problematic, since they underestimate the risk of advising this procedure without specific proposals on how to absorb these risks systematically. From a methodological point of view, for example, bivariate analyses could misjudge how close the relationship between situation control and innovativeness is because of the failure to take into account that the organization concurrently uses strategies to absorb these risks. Furthermore, quadratic functions would be needed as the basis for determining inverted U relationships statistically. And from a theoretical point of view, any description of conditions for enhancing innovativeness would be incomplete and misleading if theory takes no explicit account of the risks of extending situation control. Against this background, we argue in detail the thesis of a covert curvilinear relationship between increasing situation control and innovativeness. On a theoretical basis we explain why innovation-related initiatives can be expected to multiply with an increase in situation control, but that specific unplanned negative secondary effects also arise. It is also explained why the postulated risks of increasing situation control can be cushioned by a specifically defined counter-strategy of integration. We go on to test empirically the assumption of a covert curvilinear relationship between situation control and innovativeness and the associated thesis of the buffer effect of integration. Conclusions are drawn for the promotion of organizational innovations.
of the organization to the full. We regard innovativeness, thus defined at the aggregate level of the organization, as a function of the innovation-related initiatives of employees implemented per unit of time in the organization. The implementation rate is, among other things, a function of the situation control of employees. Situation control means the degree of perceived susceptibility of a situation to change, i.e. the extent to which members of the organization believe they can contribute directly by their own action or indirectly by actualizing the resources of others to the innovative improvement of the situation. Situation control and innovativeness are linked theoretically in a model formulated by Gebert (1987) and empirically tested by Boerner (1998), which is based on Lazarus’ (1991) stress management theory. We examine the relationship between situation control and innovativeness in the first place at the disaggregated level of the individual. We assume that a double process of appraisal precedes any innovative initiative on the part of employees. Any action with the aim of innovative improvement presupposes both that a need for change in the situation is recognized and that a susceptibility to change is perceived (Gebert, 1987). The first evaluative process is concerned with the need for change. A situation is categorized as needing change when employees perceive a discrepancy between aspirations and actuality in the corporate situation. The measure of this discrepancy therefore determines the degree to which change is needed and, thus, the perceived need for innovative initiative. If no need for change is perceived, employees develop no innovative initiative.
Rise in level of aspiration
Primary appraisal*
Secondary appraisal*
Aversity?
Situation
Need for change?
Situation Control? yes
Susceptibility to change?
yes Innovative initiatives
Theory Innovation-Friendly Effects of Situation Control By innovativeness we mean the capacity of an organization to improve existing products and/or processes (Glynn, 1996, p. 1095), and the capacity to utilize the creativity resources
Volume 12
Number 1
March 2003
Decline in level of aspiration ‘Flight’ Resignation
subjective
no ‘Flight’ objective
* following R. S. Lazarus (1991)
Figure 1. Situation Control as Catalyst for Innovativeness
© Blackwell Publishing Ltd, 2003
THE RISKS OF AUTONOMY
Where a need for change is seen (actuality < aspirations), the controllability of the situation is examined in a second process of appraisal. If sufficient situation control is detected, the person concerned develops innovative initiative in the hope of improving the situation that needs to be changed. For this reason we assume a positive relationship between the degree of situation control and the number of innovative initiatives. If a need for change is established in the course of primary appraisal, but secondary appraisal judges the situation not susceptible to change, and if the person concerned cannot objectively withdraw from the situation, resignative adaptation occurs. The demands made of the situation by the employee are lowered, and/or the actual position is upgraded by embellishment. The original aspiration/ actuality discrepancy is thus subjectively reduced, so that, as a result of this reappraisal, the situation is no longer perceived as needing to be changed. In this constellation it is likely that no initiative for innovation will be forthcoming from employees. An additional risk of resignative adaptation processes lies in their generalization to fields where change is possible and desired. For many innovations, decisions on experiments and test runs in the context of organizational learning about, for example, the generation of new if-then rules (Cheng and Van de Ven, 1999, p. 606), cannot be decentralized but have to be made centrally. In this case it is crucial for employees to be able to communicate perceived needs for change up the hierarchy with subjective and objective prospects of success, so that, at least indirectly (by actualizing executive resources) they can categorize the situation as susceptible to change. This is one of the theoretical values of participative leadership: it raises the subjective probability of desired changes being feasible. The vertical communication thus encouraged, which makes tacit knowledge public and, in the sense of organizational learning, available for further processing by others, is repeatedly stressed as a significant factor for innovativeness (Cohen & Levinthal, 1990; Glynn, 1996, p. 1103; Lam, 2000, p. 491). It can thus be argued that increasing situation control for the members of an organization promotes its innovativeness because resignative adaptation processes are avoided and innovation-friendly initiatives on the part of organization members are enabled (Axtell et al., 2000, p. 274), which constitute the raw material for later innovation at the level of the organization.
© Blackwell Publishing Ltd, 2003
43
Negative Secondary Effects of Greater Situation Control In the wake of increased situation control, however, unplanned negative secondary effects are to be expected (Dougherty, 1996). These effects ultimately mean that, although there are more innovative initiatives, the implementation rate falls (see Figure 2) if the unplanned negative secondary effects are not appropriately cushioned. If the situation is increasingly perceived as susceptible to change, employee aspirations generally grow relative to existing corporate practices. Expectations are raised and the value of the current position lowered. As the perceived susceptibility to change increases, so does the potential for criticism (lowering the value of the current position) because characteristics of the existing situation are interpreted less and less as being subject to material constraints (Gebert, 1987). On the other hand, aspirations grow because, the situation being perceived as susceptible to change, the search for alternative aspirations (for example, by comparison with other situations in other companies) is activated. This process of raising the aspiration level can – with reference to Figure 1 – be interpreted as positive feedback from the process of secondary appraisal to that of primary appraisal (Lazarus, 1991). If a situation is perceived as susceptible to change, there is a greater tendency to perceive it as needing change. Raising the level of aspiration thus mirrors the decline in the level of aspiration described above (see Figure 1). However, as a result of this rise in the level of aspiration (lowering the value of the current position and raising the level of expectations), which are in principle to be seen as favourable from the point of view of activating initiative, the following three unplanned negative secondary effects occur: • Dysfunctional qualitative orientation of innovative initiatives. Reflecting a higher level of aspiration, the number of innovation-related initiatives that are not constructive and therefore not worth implementing increases with the rise in the absolute number of initiatives for change originated by employees. • Dysfunctional intensity of innovative initiatives. The time and energy spent seeking compromises on innovations is generally experienced as less fascinating than achieving “absolute, resolute” solutions. An increase in situation control in the sense of a expansion of aspirations therefore often generates “everything-or-nothing” and
Volume 12
Number 1
March 2003
44
CREATIVITY AND INNOVATION MANAGEMENT
Integration - Orientation - Consensus - Trust
Pre-selection Pre-coordination
Ex post-selection Ex post-coordination
Conflict avoidance
Conflict resolution
Increasing situation control
Negative secondary effects: -dysfunctional quality of initiatives
-
-dysfunctional intensity of initiatives
Conflicts
-
Initiative implementation rate
-
Innovativeness
-incompatibility of initiatives
Figure 2. Negative Secondary Effects of Greater Situation Control on Innovativeness and their Compensation by Integration “now-or-never” attitudes in practice, which tend to give rise to a more radical push for change (Gebert, 1987). • Dysfunctional incompatibility of innovation-related initiatives. Augmenting situation control involves the risk of inadequate co-ordination in substance and timing between various innovative initiatives being tested independently by decisionmakers in their respective functional areas. This risk is discussed in the literature under the heading “autonomy costs of decentralization”. All three effects (dysfunctional qualitative orientation, dysfunctional intensity, and dysfunctional incompatibility of innovationrelated initiatives) imply that with increasing situation control not only the number of constructive innovation-related initiatives increases but also the number of those that put the goal in jeopardy. If detrimental initiatives outweigh constructive initiatives, the implementation rate declines. The generation of this constellation can be explained as follows. The three negative secondary effects imply subsequent decisions on the modification or rejection of innovation-related initiatives. These decisions require clear selection criteria. Since the existence of such selection criteria cannot be taken for granted, decisions about modification or rejection can lead to conflict, not least of all because these decisions disappoint the people who have developed the
Volume 12
Number 1
March 2003
initiatives. Since, although substantive conflicts can be analytically distinguished from relational conflicts, they frequently re-inforce each other, disputes can be expected to escalate (Jehn, 1995, p. 268). As conflict escalates, even selection criteria that may so far have been respected become less acceptable and binding. The growing void is increasingly filled by the exercise of power and micropolicy (Burns & Stalker, 1961). This increases the danger of constructive innovation-related initiatives being rejected and detrimental initiatives being accepted: the implementation rate falls. Increasing situation control thus proves a Janus-faced strategy, since the unplanned negative “secondary effects” trigger processes that displace the positive “primary effects”. If such a development is to be avoided, innovation-friendly conflict avoidance and resolution are needed (see Figure 2). This is the purpose of the compensatory counter-strategy of integration.
Integration as Compensation of Negative Secondary Effects We assume that the negative secondary effects of situation control can be compensated by integration. We understand the construct integration to include cognitive and evaluativeaffective aspects. The cognitive aspect of integration (‘orientation’) is concerned with
© Blackwell Publishing Ltd, 2003
THE RISKS OF AUTONOMY
the clarity of the organization’s strategic course (clear objectives in basic issues and organizational policy). The evaluativeaffective aspect of integration is concerned with the degree to which the organization’s strategic course is shared by all (‘consensus’), and is thus approved. Second, the evaluativeaffective aspect is concerned with the degree to which mutual confidence (‘trust’) arises as an expression of a clear, shared strategic course (in the sense of a positive sum game), and people consequently feel themselves to be emotionally integrated. These aspects thus merely reflect the same construct (‘integration’) from different angles. We now explain the compensatory effects of integration as thus defined. In the first place, integration supports the processes of pre-selecting and pre-coordinating innovation-related initiatives, which serve conflict avoidance. (a) Orientation, defined in terms of the clarity of the strategic course set by the organization, is one aspect of integration. The clarity of course determines the direction and cognitive scope for potential innovation-related initiatives, thus setting cognitive limits to the process of developing initiatives. In the knowledge of the long-term organizational course, the pre-selection of innovative initiatives is more purposive, more moderate and better co-ordinated. Additionally needed interpersonal pre-coordination is also facilitated by the evaluativeaffective aspects of integration (‘consensus’ and ‘trust’). (b) We define consensus as the capacity to reach agreement within the organization on fundamental issues. Consensus obviates debates on principles when specific innovative initiatives are to be modified in the course of pre-coordination. (c) We define trust as mutual willingness to accept vulnerability (even without contractual protection). Trust reflects the experience of mutual fairness and reliability (Rousseau et al., 1998, p. 401). In the context of positive-sum games, this experience reduces the likelihood of micro-political intentions being mutually imputed in the course of pre-coordination when changes in innovative initiatives are under discussion. By fostering pre-selection and precoordination, integration thus contributes to the following effect: the number of dysfunctional innovation-related initiatives for change
© Blackwell Publishing Ltd, 2003
45
declines, so that the rate of successfully innovation implementation rises. But integration is needed for another reason, too. Even when orientation, consensus and trust are highly developed, the quantity of detrimental innovation-related initiatives is not reduced to zero. This is impossible if only because, although the cognitive scope for innovation-related initiatives defined by orientation is limited, it nevertheless remains subject to interpretation and, to a certain extent, ambiguous. For this reason, too, not only pre-selection and pre-coordination are needed in practice, but also ex-post-selection and ex-post-coordination as well as appropriate conflict resolution when specific innovation-related initiatives need to be rejected or changed. Keeping in mind the reasons we have given for why the rate of successfully implemented initiatives declines in the face of inevitable subsequent decisions on modifying or rejecting innovative initiatives, it is clear that the counter-strategy of integration favours the implementation rate because it promotes conflict settlement. As integration intensifies, the unplanned negative effects of increasing situation control are effectively cushioned by conflict avoidance and later conflict resolution. However, this means that without such compensation, an organization can find itself in difficulties that can substantially jeopardize innovativeness. This is obviously also realized by consultants and companies (Hardy & Leiba-O´Sullivan, 1998, p. 463). This explains why on the one hand they advocate empowerment and entrepreneurship (corresponding to increasing situation control) while nevertheless stressing visionary leadership and the importance of trust in the organization (corresponding to increasing integration). If renowned businesses embrace such practices, they are emulated by other companies (Greenwood & Hinings, 1996, p. 1028), so that the compensatory strategy of integration is more broadly diffused. For this reason we expect the degree of situation control in organizations to be positively linked with the degree of integration. Organizations that take no compensatory measures to counter increasing situation control would, according to this approach, run the risk of losing their innovativeness and thus their competitiveness. Finally, we assume that, because of the concurrent integration processes, the empirical observer gains the impression that the innovativeness of an organization grows as situation control increases (in terms of the linearity thesis). If, in contrast, the degree of integration were to be kept constant, it should be apparent that, in
Volume 12
Number 1
March 2003
46
CREATIVITY AND INNOVATION MANAGEMENT
keeping with the curvilinear thesis, innovativeness declines from a certain level of situation control. The hypothesis to be tested is as follows. If integration is held constant, situation control and innovativeness exhibit an (inverted U) curvilinear relationship; if integration is concurrent, the relationship is linear.
Empirical Study Measurement The construct situation control embedded in the theory shown in Figure 1 was operationalized with the following items (the data were collected in standardized form on a five-point scale: “strongly agree”, “agree”, “uncertain”, “disagree”, “strongly disagree”). • In this company employees have the feeling they can discuss even fundamental company practices with management (S1). • We have learned that you can elude material constraints (S2). • Employees here think a lot needs to be changed, but that change is impossible (recoded) (S3). • We’ve come to the conclusion that practically everything can be done otherwise (S4). • The success of this company is attributed mainly to the ability to get things moving and to bring about change actively (S5). • In our experience, you can generally achieve more in this organization that it would seem at first (S6). The construct integration (consensus, orientation, trust) as defined above was operationalized with three items: (a) Orientation (O) • Many feel a lack of reliable orientation on fundamental issues (recoded) (O1). • The top-management in this organization always succeed in giving employees a clear basic orientation (O2). • Even if a person is sometimes at a loss on details, the general direction of the task is clear (O3). (b) Consensus (C) • In our organization, everyone pulls together. There is no quarrelling that could seriously endanger unity (C1). • Conflicts about the matter itself seldom end in protracted debates on principles (C2). (c) Trust (T) • In our organization people can show their weaknesses without it being used against them (T1).
Volume 12
Number 1
March 2003
• We’re very far from being a trusting organization (recoded) (T2). • Without backing from your superior noone here takes a risk (recoded) (T3). In operationalizing innovativeness we set the following seven questions concerning comparison with competitors, comparison with own past innovativeness and comparison between the innovativeness attained and own innovative potential. As we see it, only the combination of these three aspects permits a valid measurement of innovativeness; taking only one of the three measures into account gives a one-sided and thus less valid picture. • With regard to innovativeness (e.g. creativity, new products, new processes), is your company or the division you are assessing better, just as good or worse than its competitors? (I1) • In your opinion, is the available potential for innovativeness/creativity fully exploited in your company (to a low degree, partly, to a high degree)? (I2) • Over the past three years, has your company been better, just as good or worse than before as regards innovativeness (e.g. creativity, new products, new processes)? (I3) • If we want to be as innovative as our best competitor, we still have a long way to go (recoded) (I4). • The last innovation in our organization that was really important for the success of the company was years ago (recoded) (I5). • Our company has a reputation among internal/external customers for creative, innovative problem-solving (I6). • Crucial innovation/creativity resources lies fallow in this organization (recoded) (I7). With regard to the reliability of measurement of the three constructs situation control, integration and innovativeness the following results emerged, as shown in Table 1. On the basis of these three reliable scales, the six items on situation control are summated under the heading “situation control”, the eight items on integration under “integration” and the seven items on innovativeness under “innovativeness”.
Sample In this study the number of respondents corresponds to the number of organizations or organizations units under investigation (n = 101). In contrast to studies in which respondents are recruited from only a few organizations, variance in this investigation is organizational rather than personal. Only
© Blackwell Publishing Ltd, 2003
THE RISKS OF AUTONOMY
47
Table 1. Reliabilities (Cronbach’s a) of the Measured Constructs Construct
Situation Control Integration Innovativeness
No of items
Reliability (a)
6 8 7
0.758 0.808 0.901
Table 2. Correlation Coefficients between Situation Control, Integration and Innovativeness Innovativeness
Integration
0.639*** (n = 95) 0.634*** (n = 93)
0.591*** (n = 97)
Situation Control Integration
*** : significant for p < 0.001.
executives were interviewed (mostly departmental managers). We approached the randomly selected respondents personally; presumably as a result of this personal contact, the response rate was 65%. The 101 organizations or organizational sub-systems were recruited in the service sector (insurance, banking, consultancy, media) and manufacturing (motor vehicles, chemicals, electric and electronic equipment, mechanical engineering).
Results Table 2 shows correlations between the three central variables situation control, integration, and innovativeness. All three variables show significant positive inter-correlation. In substance, our central thesis is that the relationship between situation control and innovativeness takes the covert curvilinear form of an inverted U, while appearing to be linear because concurrent integration cushions the unplanned negative secondary effects of a higher measure of situation control. If one wished to reveal the risks of a marked and one-sided increase in situation control (i.e. the potential inverted U form of the relationship between situation control and innovativeness), integration processes covariant with situation control that cushion the negative secondary effects of situation control must be controlled for and their effect eliminated in computing the correlation between situation control and innovativeness. If the level of integration is kept constant, the risks of increasing situation control (and thus the posited covert curvilinearity) would have to be apparent. The results of this test are shown in Table 3. If, after controlling for the effect of integration, we take the inverted U function – a quadratic function – as the basis, the quadratic term is significant, as Table 3 shows. This means nothing other than that, after controlling for integration, the function between situation control and innovativeness does not rise in
© Blackwell Publishing Ltd, 2003
Table 3. Regression Coefficients (b) of Situation Control on Innovativeness after Controlling for the Impact of Integration R 0.409***
bLinear
bQuadratic
0.358***
-0.244*
* : significant for p < 0.05; *** : significant for p < 0.001
constant linearity but reaches a maximum beyond which it falls again. Precisely this finding confirms our central hypothesis (see Figure 3).
Conclusions The successful demonstration of covert curvilinearity has a number of implications. Since the frequently recommended procedures of decentralized self-regulation, decentralization of decision-making powers, the promotion of decision-making autonomy and empowerment as well as participative leadership imply an increase in situation control, they are associated not only with opportunities but also with identifiable risks (hitherto little discussed in the literature). With regard to innovativeness, there can be not only too little situation control but, where not absorbed, too much. Burns and Stalker (1961) have warned against absolutizing the strategies of decentralized self-regulation, the promotion of decision-making autonomy and so on, although their terminology differs. At the practical level, solving this problem by dosing situation control proves very difficult. Since there is no clear yardstick for situation control and the optimum is unknown ex ante, practitioners can often act only ex post, when
Volume 12
Number 1
March 2003
48
CREATIVITY AND INNOVATION MANAGEMENT
0,8 0,6 0,4
Innovativeness (standardized values)
0,2 0,0 -0,2 -0,4 -0,6 -0,8 -1,0 -1,2
quadratic fit
-1,4 -1,6 -3
-2
-1
0
1
2
3
Situation Control (standardized values) Figure 3. Relationship between Situation Control and Innovativeness (Controlling for the Effects of Integration) innovative initiatives either peter out or develop into prospectively uncontrollable destabilization of the organization. Not infrequently, an everything-or-nothing attitude and now-or-never thinking arise as well, especially among younger employees, with the result that members of the organizations are not to be satisfied with gradual increases in situation control. To this extent a strategy of compensation is preferable to one of dosing. The promotion of orientation, consensus and trust seems to be the counterweight that prevents the actuation of initiatives and change from veering into conflict escalation and paralysis. Under these conditions the relationship between situation control and innovativeness is in fact linear and does not merely appear to be linear, so that innovativeness does not begin to decline with increasing situation control but continues to rise. From a practical point of view, transformational leadership, operationalized in accordance with Avolio, Bass and Jung (1999), is relevant in this context (Keller, 1992). The trust bonus of leaders vis-à-vis the led allows a potential for trust to develop; an inspiring common vision enables orientation and builds a potential for consensus. In this way the risk of increasing situation control through delegation and participation can be cushioned. As we have seen, companies tend to favour this buffer strategy. Nevertheless, the need for
Volume 12
Number 1
March 2003
it must be stressed: as our empirical analysis has shown, there are companies that do not consistently practice such compensation. Moreover, it is important for practical reasons to describe and justify the content of the strategy more precisely to permit more effective action.
References Avolio, B.J., Bass, B.M. & Jung, D.I. (1999) Reexamining the components of transformational and transactional leadership using the multifactor leadership questionnaire. Journal of Occupational and Organizational Psychology, 72, 441–462. Axtell, C.M., Holman, D.J., Unsworth, K.L., Wall, T.D. & Waterson, P.E. (2000) Shopfloor innovation: Facilitating the suggestion and implementation of ideas. Journal of Occupational and Organizational Psychology, 73, 265–285. Boerner, S. (1998) Work alienation and continuous improvement. In Boer, H. & Gieskes, J. (eds.), Continuous improvement – From idea to reality, Twente University Press, Enschede, pp. 63–75. Burns, T. & Stalker G.M. (1961) The management of innovation. Tavistock, London. Cheng, Y.T. & Van de Ven, A.H. (1996) Learning the innovation journey: Order out of chaos?. Organization Science, 7(6), 593–614. Cohen, W.M. & Levinthal, D.A. (1990) Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35, 128–152. Damanpour, F. (1991) Organizational innovation: A meta-analysis of effect of determinants and
© Blackwell Publishing Ltd, 2003
THE RISKS OF AUTONOMY
moderators. Academy of Management Review, 34(3), 555–590. Dougherty, D. (1996) Organizing for innovation. In Clegg, S. R., Hardy, C. & Nord, W.R. (eds.), Handbook of organization studies. Sage Publications, London, pp. 425–439. Fiol, C.M. (1996) Squeezing harder doesn’t always work: Continuing the search for consistency in innovation research. The Academy of Management Review, 21(4), 1012–1021. Gebert, D. (1987) Führung und Innovation [Leadership and Innovation]. Zeitschrift für betriebswirtschaftliche Forschung, 39(10), 941– 951. Gebert, D. & Boerner, S. (1999) The open and the closed corporation as conflicting forms of organization. The Journal of Applied Behavioral Science, 35(3), 341–359. Glynn, M.A. (1996) Innovative genius: A framework for relating individual and organizational intelligences to innovation. The Academy of Management Review, 21(4), 1081–1111. Greenwood, R. & Hinings, C.R. (1996) Understanding radical organizational change: Bringing together the old and new institutionalism. The Academy of Management Review, 21(4), 1022– 1054. Hardy, C. & Leiba-O´Sullivan, S. (1998) The power behind empowerment: Implications for research practice. Human Relations, 51(4), 451–483. Jehn, K.A. (1995) A Multimethod Examination of the Benefits and Detriments of Intragroup Conflict. Administrative Science Quarterly, 40(2), 256–282. Keller, R.T. (1992) Transformational Leadership and the Performance of Research and development project groups. Journal of Management, 18(3), 489–501.
© Blackwell Publishing Ltd, 2003
49
Lam, A. (2000) Tacit knowledge, organizational learning and societal institutions: An integrated framework. Organization Studies, 21(3), 487–514. Lazarus, R.S. (1991) Emotion and adaptation. Oxford University Press, New York. Rousseau, D., Sitkin, S.B., Burt, R.S. & Camerer, C. (1998) Not so Different After All: A crossdiscipline view of trust. The Academy of Management Review, 23(3), 393–404. Van de Ven, A.H. (1999) Central Problems in the Management of Innovation. Management Science, 32(5), 590–617. Wolfe, B. (1994) Organizational innovation: Review, critique and suggested research directions. Journal of Management Studies, 31, 405–431.
Prof. Dr. Diether Gebert is at the Berlin Technical University, Faculty of Economics, and Management, Department of Human Resource Management, 10585 Berlin Wilmersdorfer Str. 148, Germany, phone (030) 314 23 234. Prof. Dr. Sabine Boerner is at the Lahr School of Economics, Department of Human Resource Management and Intercultural Management, Hohbergweg 15–17, 77933 Lahr Germany, phone (07821) 9238–59. e-mail:
[email protected] Ralf Lanwehr is at the Berlin Technical University, Faculty of Economics and Management, Department of Human Resource Management, 10585 Berlin, Wilmersdorfer Str. 148 Germany, phone (030) 314 23 234.
Volume 12
Number 1
March 2003
50
CREATIVITY AND INNOVATION MANAGEMENT
Supporting Innovation through HR Policy: Evidence from the UK Rosalind H. Searle and Kirstie S. Ball This paper focuses on the relationship between the importance of innovation for organizations and their human resources policy. Drawing on survey findings, we examine the coherence of organizations’ utilization of HR recruitment, training and performance management policies to support and enhance firms’ innovation performance. Through a social–psychological perspective, we situate our findings in two diverse areas: the psychological literature, exploring the measurement of innovation, and second, with regard to the internal (with each other) and external (with broader organizational objectives) integration of distinct HR policy elements. Our surveyed organizations indicate that, whilst attaching importance to innovation, they fail to consistently translate this importance into coherent HR policies. Typically, HR policy rewarded non-managerial employees for innovation, whilst managerial staff were expected to do so as a matter of course. This inconsistency is one source of resistance which blocks the generation of new ideas, and their implementation, organization-wide.
Introduction
T
his paper addresses the question of whether organizations coherently use HR policy in promoting innovative behaviours to enhance competitive performance. Drawing on survey findings, we examine the extent to which organizations identify, utilize and support innovation in their recruitment, training and performance management processes. We find that our surveyed organizations, whilst attaching importance to innovation at a conceptual level, fail to consistently translate and support this in HR policy. Furthermore, we find that the application of HR policies to encourage innovative behaviour is targeted at specific groups of employees, typically at lower organziational levels. Thus, whilst organizations are encouraging their non-managerial employees to innovate, the lack of attention paid to middle and senior managers (perhaps with an implicit expectation that they should innovate anyway) will sustain an HR-policy approach fraught with resistance and misunderstanding about the implementation of new ideas. Whilst we are writing from a social–psychological perspective, we situate our findings in two diverse areas. First, in psychological literature, which addresses the development of
Volume 12
Number 1
March 2003
an innovation–specific behavioural category, and the application of innovation assessment and development methods; and second, in recent developments of the ongoing debate surrounding the integration of different HR policy elements, both internally (with each other) and externally (with broader organizational objectives). Within the latter, innovation1 has been discussed from early strategic fit models of HR (Schuler & Jackson, 1987) and in more recent work which discusses the HR implications of the implementation of new work practices. Thus, our paper proceeds as follows. First, we outline the two literatures upon which we draw, and then present our survey findings. In the light of the findings, we then discuss the theoretical and practical implications of the development, identification and integration of an HR policy designed to support innovation in the organization at large. 1
We note that ‘innovation’, as a sub-discipline of management studies, is extremely heterogeneous in nature, with studies utilising perspectives from micro psychology to structuralist sociology and neo – classical economics. In this paper, and in the literature upon which we draw, we refer to micro conceptions of innovation as a social – psychological process.
© Blackwell Publishing Ltd, 2003. 9600 Garsington Road, Oxford OX4 2DQ and 350 Main St, Malden, MA 02148, USA.
SUPPORTING INNOVATION THROUGH HR POLICY
Managing an Innovative Human Resource As academic and practitioner commentators continue to note that the ‘rapid’ change in business environments and ‘incessant’ technological development requires organizations to be constantly looking to revise, reflect and develop anew, the question of how innovative process might be managed from a human resources perspective is indeed a timely one. Innovation is important in two key ways for organizations, playing a role in both radical developments (Colewell, 1996) and in smaller continuous changes (Bessant, 1992). Findings suggest that the ongoing survival of existing firms is dependent upon their ability to innovate in terms of their products, processes and systems (Oldham & Cummings, 1996). Whilst the observation concerning an organization’s need to innovate is hardly new (Schumpeter, 1934), we see rapid changes in the speed with which organizations are required to develop novel services and products in order for them to maintain and enhance their position (Jackson, 1976). One way for organizations to become more innovative lies in their ability to foster, develop and utilize the talents, in particular the innovative potential, of their employees (Amabile, 1988; Oldham & Cummings, 1996; Wolf, 1994). As Axtell et al. (2000) have indicated, innovation is not confined to individual creativity per se but to the novel application and implementation of ideas by groups within the organization’s context. The issue for organizations who value innovation is, therefore, how to select, develop and motivate individuals capable of formulating ideas in the first place, and also to create the supportive environment in which groups can productively and swiftly implement them. These notions are reflected in recent social–psychological models of organizational innovation. Guzzo and Shea (1992) highlight the role of the wider organizational context in shaping such novelty. They demonstrate that recruitment and selection, training and development, feedback, reward and recognition of employees are vital inputs in this respect. Furthermore, Axtell et al.’s (2000) study of shopfloor workers provides some specific evidence concerning the role of some of these factors in creating innovation. They confirmed an important link between individual (predispositional and job) factors, and the generation of ideas, whilst organizational and team factors emerged as of more significance during their implementation. Thus, the need to select, develop and organize with these factors in mind is confirmed.
© Blackwell Publishing Ltd, 2003
51
In this respect some more wide-ranging personality measures, such as the Occupational Personality Questionnaire (Saville, 1973), have included a factor called innovation. Other researchers have chosen to produce personality tools specifically concerned with innovation. The most well-known attempt to measure individual innovation levels is Kirton’s (1976) Adaptation-Innovation inventory (KAI), which distinguishes two types: ‘adaptors’, who are ‘doing things better’ and ‘innovators’, who are ‘doing things differently’. Concurrent with this Jackson’s (1976) Personality Inventory (JPI) includes an innovation sub-scale (JI) which focuses on identifying both individual creativity and innovation. However, these measures treat innovation as a simple uni-dimensional construct and ignore any motivational or organizational component. The importance of personality factors and behaviour in context in securing innovative performance was empirically confirmed by Campbell, Gasser and Oswald (1996). Adopting a process–based view of innovation, they examined job performance and found a link between the incidence of innovation-specific behaviour and organizational performance. Thus, as organizations seek to become more innovative, so the behaviours of their employees’ would also have to change. It also follows that if organizations claim to have an ‘innovative’ strategy (Schuler & Jackson, 1987), and promote this as a strategic requirement, we would expect to find such behaviours in evidence, reinforced through various recruitment, training, reward and development activities. As we shall see, however, this logic is rarely in evidence. In the next section, we proceed to discuss the nature of the linkage between employee behaviours, HR policies and organizational performance more broadly.
Innovation, HR Policy and ‘Organizational Fit’ We begin with the question of how organizations might go about re-inforcing innovative behaviours through HR policy. Earlier fieldwork conducted by Forrester (1994) has revealed a number of different techniques deployed by organizations to either select or promote innovation amongst their workforces. Field studies of several large organizations indicated the application of ‘innovation’ as a competence for use in assessment centres either from individual or group exercises or criterion-based interviews, whilst others utilized scales in existing personality tools (e.g.
Volume 12
Number 1
March 2003
52
CREATIVITY AND INNOVATION MANAGEMENT
SHL’s OPQ measure). In addition it was also found in the application of performance management techniques. These involved the application of either individual or group-based targets which included either specific action (e.g. to create ten new ideas) or more behavioural approaches (e.g. introducing ideas from outside the organization). It was also observed that many organizations used different intrinsic or extrinsic incentive systems (Amabile, 1988). These might include for example, financial rewards or prizes for ideas, whilst others had an implicit understanding that by selecting innovative people, they had to do nothing further other than wait for the ideas to emerge. Links between employee behaviour and organizational performance have not been so evident in broader studies of HR policy and organizational performance. Despite this, the idea that different elements of HR practice should be integrated with each other in a ‘bundle’ (internal fit) and that HR policy should dovetail with business objectives (external fit) has been intrinsic to the HRM ethos since its earliest models in the 1980s. To elaborate, ‘internal fit’ refers to the extent to which a set of HR policies, pertaining to recruitment, training, performance management and reward exhibit internal coherence and consistency. For example, that they all be based on a given competency framework, that performance data inform recruitment, training and reward decisions, and so on. ‘External fit’ refers to the extent to which a given set of HR policies cohere and integrate with overarching business objectives, and that HR decisions both inform and are informed by organizational performance, at both strategic and operational level. Whilst internal fit is a topic which has received minor attention (see Storey, 1992), external fit has been an area of continued debate between academics. From early models of HRM, such as that of Beer et al. (1984) and Guest (1987), external integration and fit has been constructed as a desirable objective for the competent HR department and the business enterprise which sees people as ‘the key to success’. Since these early models, research has attempted to posit the nature of this link. The linear integration of HR policy and strategy has been critiqued by writers such as Mueller (1996) who claims that the organization’s social fabric, or ‘social architecture’ is responsible for codified, yet disparate, HR practices throughout the organization. These evolve alongside strategic policy, and relate to it in a recursive way. Some writers claim that HR practices make a positive contribution (Boxall, 1996) to organizational performance, but in
Volume 12
Number 1
March 2003
order to sustain a competitive advantage, would have to be somehow idiosyncratic to the organization. Others argue that coherent HR practices only emerge when the organization is performing successfully (Paauwe & Richardson, 1997) whilst Guest and Hoque (1994) conclude that any level of fit is the exception rather than the norm. The extent to which we see any clear evidence that HR policy and strategy can be coherently integrated is thus limited. Furthermore, Hope-Hailey et al.’s (1997) review of the state of HRM in the 1990s suggests that the integrative models of the 1980s are not in evidence. They posit the importance of rich contextual variables of the type alluded to by Mueller (1996) in shaping HR practice, citing incidences of a ‘nuts and bolts’ approach to performance management, a focus on hard HRM, team-based discipline and self-management. Problems with external fit and devolution of people management practice to line management were also identified. The latter is something that is not altogether unanticipated, as Hall and Torrington (1998) reported difficulties experienced by line managers in adopting the responsibilities associated with HRM, such as welfare and counselling roles. Similarly Tyson (1999) concurs with HopeHailey et al.’s emphasis on the ‘nuts and bolts’ approach, arguing that greater weight was given to process knowledge (e.g. management of change) and diagnostic skills, relating to HR issues, rather than technical expertise, on the part of line managers. A ‘nuts and bolts’ response is also suggested in work examining grassroots innovative work practices. MacDuffie (1995) asserts that innovative work systems themselves produce skills requirements (for example, problem solving). This is similarly observed by Felstead and Ashton (2000) who, following empirical investigation, argue that these emergent operational process and diagnostic skills can then be articulated in HR policy and practice, from recruitment through to training and development, appraisal and reward (Felstead and Ashton, 2000). Furthermore, the latter observe that new skills development is not associated with training in that particular skill. However, when a pro innovation strategy is in evidence, companies carry out more training actions in product or process techniques for both managerial and non-managerial jobs (Sanz–Valle et al., 1999). From current evidence, it would seem that we can tenuously conclude that HR policy has a greater role to play in supporting and rewarding desired behaviours, which are at the locus of action between line managers and employees, than on any other, more grand, strategic scale,
© Blackwell Publishing Ltd, 2003
SUPPORTING INNOVATION THROUGH HR POLICY
although this level is not to be completely overlooked. Similarly, while there is little evidence as to the internal fit of HR policy to support innovation, current evidence suggests that any external fit of HR policy in terms of innovation is more likely to be at lower organizational levels. Based on recent evidence, the paper interrogates the following four propositions: 1. Innovation will be reported as important to the majority of organizations. 2. Human resource policies supporting innovation will be reported as applied at the locus of everyday organizational action 3. The external fit of HR policy that supports innovation will have an inverse relationship with the hierarchical level at which it is applied. 4. There will be little evidence that the different HR policy elements relating to innovation will show any degree of internal fit. Thus, this paper examines the above in a larger number of organizations, to explore the link between the importance assigned to innovation, the HR practices used to develop such activities and to assess what type and level of employee was being targeted in these policies. We shall now assess data from a survey of different types of organization regarding their HR practices and procedures, and their relationship to an innovative focus.
Method Data were collected via a survey of the top 300 organizations identified from the FTSE 500, employing over 100 people. In order to increase the response rate, the questionnaire was sent to named individuals and a selfaddressed stamped envelope included. The individual was the company’s most senior HR professional who would have the greatest knowledge of the company’s strategy and HR strategy, policies and practices. The survey concerned the importance of innovation in each organization and the HR policies and practices they utilized. The first section gathered demographic details of the organization: sector, size, location of head office and financial turnover. It also included a five-point Likert rating scale to indicate how important innovation was for the organization. This ranged from (one) not important at all, through to (five) critical. The survey then focused on the organizations’ practices in five aspects of human resource management. Drawing on Forrester (1994), these included external recruitment, internal selection, performance management,
© Blackwell Publishing Ltd, 2003
53
training and development, and ideas management schemes. For each of these practice areas, the respondent was asked to identify differences across six different staff levels: senior managers, middle managers, new graduate, clerical, skilled and semi-skilled. Although the survey drew from the techniques and tools mentioned in the literature, the inclusion of the open response box after every section enabled the potential capture of emergent policies and approaches. All of the responses were entered into SPSS. Where an open response had been used, common themes were identified in order that a new category could be recorded. The second section focused on selection and recruitment, with respondents identifying whether innovation was a criteria used for a range of practices employed. These practices included personality tests, interviews, individual and group exercises. The survey also distinguished between internal selection and the external recruitment processes. The third section considered training and development. Here, commonly utilized innovation training techniques had been identified from the literature. These included brainstorming, lateral thinking and problem solving. Respondents indicated training provision for the aforementioned six levels of staff. Performance management policies formed the fourth aspect in the survey. The literature has revealed two distinct approaches to performance management, in terms of ‘what’ or ‘how’ performance is achieved. Respondent were asked first to indicate the role of innovation in terms of ‘what’ staff achieved, by focusing on innovation in terms of a performance target. They then indicated whether innovation was an important feature of ‘how’ these targets were met. This was more concerned with innovation as a behavioural competency. The respondents indicated any differentiation between individual and team-based approaches to performance management. The final section focused on ideas management systems. Five most commonly used policies were identified from the literature; including expectations that innovation was part of the individual’s role, through to more formal organizational processes like the use of ideas champions, tangible reward systems (differentiating between financial rewards versus opportunities to win prizes), through to no formal policy. As with performance management systems, a distinction was made between individual and team-based applications. The data were analysed using one-way ANOVAs with importance of innovation as
Volume 12
Number 1
March 2003
54
CREATIVITY AND INNOVATION MANAGEMENT
the dependent variable, and each of the different sections on HR policies entered as independent variables in order to examine how the organization’s attention to innovation was translated into HR policies and procedures.
Results Eighty-eight organizations responded to the survey, representing a response rate of 30 per cent. They came from a range of industrial sectors operating within the UK, as can be clearly seen in Table 1. Figure one indicates the level of importance assigned by the organization to innovation. Over 84 per cent of these organizations regarded innovation as either critical or important, regardless of their size (in terms of both staff number and turnover) or the sector they occupied. Furthermore, the results, show, first, that the organizations surveyed are using a range of different HR policies to support innovation at different levels, and second, that this varies further when we come to consider the relationship between the reported importance of innovation to the organization, and HR policies.
Recruiting for Innovation Different approaches to the application of innovation within external selection at the various grades of staff are evident (see Figure 2). Although the most commonly utilized tool to assess innovation is interviewing, it is used progressively less frequently at lower organizational levels. It is striking that at graduate level and above there is a wider range of tools
used to identify innovation. At clerical level and below, personality profiles and individual exercises are also used less frequently. One–way ANOVAs revealed that there is a differentiation between the use of external recruitment tools by those who regard innovation as critical and those for whom it was a lower priority. We found significant differences (p > 0.05) in the range and level of tools used by firms who regard innovation as important. Furthermore, whilst the recruitment practices for graduate, skilled and semiskilled staff did not alter according to the importance of innovation, this was not the case if senior managers, middle managers and clerical staff were being recruited. Different recruitment processes were deployed by
Table 1. Characteristics of the organizations Characteristics of the organizations Sector
Number of organizations
% of survey
Manufacturing Service Financial Utility Engineering (inc. building) Misc
30 17 14 10 5
34 19 16 11 6
12
14
Total
88
100
critical important one of a no. of initiatives marginal importance not significant to the firm
Figure 1. Importance of Innovation in the Organizations
Volume 12
Number 1
March 2003
© Blackwell Publishing Ltd, 2003
SUPPORTING INNOVATION THROUGH HR POLICY
55
100 90
Frequency of use (%)
80 70 Personality tool
60
Interview
50
Individual exercises
40
Group exercises
30 20 10
ki
lle
ille
d
d
ric al
m
i-s
le
Sk
ra
C
du
at
es
g. an a M
Se
G
Se
n.
M
an a
g.
0
Level & Tool preference
Figure 2. Preferred Selection Tools – External Recruitment Table 2. One-way ANOVA for external and internal selection approaches favoured by those who regard innovation as important Organization level
Senior management (external) Management (external) Clerical staff (external) Senior management (internal) Management (internal) Clerical staff (internal) Skilled staff (internal)
External Selection process
Degrees of freedom
F
Group exercise Individual group exercise Personality profile Group exercise Individual exercise Personality profile Personality profile
1.87 1.87 1.87 1.87 1.87 1.87 1.87
5.66 8.79 7.42 4.09 6.39 4.83 4.96
organizations that regarded innovation as important, in relation to these groups. For senior management positions, group exercises were preferred (F(1,87) = 5.65, p < 0.05), whilst for management grades, individual exercises were the differentiator (F(1,87) = 8.78, p < 0.01), and personality tools were deployed for clerical staff (F(1,87) = 7.42, p < 0.01). Of these approaches, the most common method was individual exercises at the management grade (78 per cent of those who used this approach regarded innovation as critical to their success), whilst at the clerical grade, clerical personality profiling was used in 73 per cent of firms who held innovation as important.
© Blackwell Publishing Ltd, 2003
Significance
p p p p p p p
> 0.05 > 0.01 > 0.01 > 0.05 > 0.01 > 0.05 > 0.05
When making internal selection decisions, specific approaches emerged in relation to all grades except for non-skilled staff, which bore relation to the stated importance of innovation. For example, personality profiles are used in firms where innovation is important, for skilled staff. This was also revealed in the previous analysis, but not to a statistically significant level.
Training for Innovation In examining training and development methods, problem solving is the dominant technique in firms that rate innovation as
Volume 12
Number 1
March 2003
56
CREATIVITY AND INNOVATION MANAGEMENT
important. Lateral thinking tools are utilized from graduate to senior levels, and brainstorming is common at all levels. Most striking is the difference between senior and other management, with senior staff receiving less development opportunities in these area compared to graduate and managerial staff. One-way ANOVAs revealed that a very narrow area of skills were being targeted. Problem solving and brainstorming opportunities were being offered to clerical staff (F(1,87) = 5.50, p < 0.05) and (F(1,87) = 5.62, p < 0.05, respectively) and was used by 56 per cent of firms, 81 per cent of which regarded innovation as at least important. Non-skilled staff were also being trained
in problem-solving techniques (F(1,87) = 4.58, p < 0.05).
Managing Performance for Innovation In turning our attention towards recognition and reward of innovative behaviour an individualized approach both in terms of targets and behaviours is the favoured approach, regardless of organizational level. Team-level targets and behaviours emerge at management level, but are far less common at other levels (see Figure 4). This is the area where, perhaps unsurprisingly, we find the most variation between importance of innovation and HR–specific
100 90
Frequency of use (%)
80 70 60
Lateral thinking
50
Problem solving
40
Brain storming
30 20 10
d
d
ki
lle
ille Se
m
i-s
Sk
G ra du at e C le ric al
M an g.
Se n. M an g.
0
Level of employee
100 80 60 40 20 0
Specific targets - individual level
d se
m
i/
...
ille sk
ic er cl
a. an m
al
Specific targets - team level
..
Frequency (%)
Figure 3. Innovation Training and Development by Grade
Behaviour competencies individual level Behaviour competencies team level
Grade Figure 4. Performance Management Processes of Organization by Grade
Volume 12
Number 1
March 2003
© Blackwell Publishing Ltd, 2003
SUPPORTING INNOVATION THROUGH HR POLICY
57
Table 3. One-way ANOVA for performance management approaches favoured by those who regard innovation as at least important Organizational level
Performance management approach
Degrees of freedom
F
Significance
Senior management
Individual performance targets Individual performance targets Individual performance targets Individual performance targets Team level behavioural competencies
1.87
7.62
p > 0.01
1.87
12.55
p > 0.01
1.87
6.34
p > 0.05
1.87
7.04
p > 0.01
1.87
4.17
p > 0.05
Clerical staff Skilled staff Non-skilled staff Non-skilled staff
Table 4. Ideas management schemes for organizations by grade Idea management tool/grade Expected as part of normal role – individual Financial reward – individual Points & Prizes – individual Idea champions – individual Suggestion scheme – individual No policy – individual Expected as part of normal role – team Financial reward – team Points & Prizes – team Idea champions – team Suggestion scheme – team No policy – team
approaches. One-way ANOVAs indicated that there was a significant difference between performance management practices for those organizations that regarded innovation as significant, and for those that regarded it as less so. The most commonly used technique was individually focused performance targets for senior managers, which were found in 77 per cent of the cases, with 61 per cent of these organizations rating innovation as important. Further down the organization, the deployment of this approach within firms who saw innovation as similarly important rose to 91 per cent for clerical, skilled and 92 per cent for non-skilled staff. Furthermore, the previously discussed action–based nature of innovation is clearly reflected in the deployment of individual performance targets at the four distinct grades, shown in Table 4. However, the only
© Blackwell Publishing Ltd, 2003
Management
Clerical
Skilled
81.8 14.8 3.4 10.2 2.3 4.5 35.2 3.4 1.1 6.8 58.8 3.4
58 21.6 5.7 6.8 3.4 10.2 23.1 6.8 3.4 6.8 19.3 5.7
48.9 19.3 4.5 5.7 2.3 10.2 22.7 6.8 3.4 5.7 0 4.5
performance management approach to focus at the team, rather than the individual level, was for non-skilled employees.
Innovation and Idea Management Surprisingly, the application of specific ideas management schemes to support innovation is not common. It is noteworthy that it is expected of management grades that they innovate as part of their job. There is far more limited deployment of other encouragement schemes for this level in comparison with the others. In comparison, clerical staff are subject to significantly more team and individualbased idea management schemes. Similar to the approaches deployed for performance management, one-way ANOVAs revealed that the processes which were
Volume 12
Number 1
March 2003
58
CREATIVITY AND INNOVATION MANAGEMENT
designed to actively reward new ideas were all applied at an individual level. Organizations in which innovation was at least important differed from the others through the use of two distinct approaches to ideas management for the clerical grades. These employees were more likely either to be expected to be innovative as part of their role (F(1,87) = 4.11, p < 0.05), or encouraged to be so through an ideas management scheme (F(1,87) = 5.33, p < 0.01).
Discussion: Policies in Practice and Levels of Fit In this section we use the findings to interrogate our earlier research propositions derived from the literature, and discuss the implications of our study for organizations. Our data show varying degrees of external and internal fit, between HR policy elements and espoused importance of innovation, an overall emphasis on recruiting for innovation over other elements of the HR process, a de-emphasis on training and development for innovation and ideas management, a performance management style which focuses on the individual, rather than team for developing innovative behaviour, and a de-emphasis on ideas management for sustaining the behaviour.
Proposition One: Innovation Will be Reported as Important to the Majority of Organizations The findings indicate that innovation was an important goal for 84 per cent of the organizations surveyed. This confirms Nadler and Tushman’s (1997, p. 135) observation on the imperative nature of innovation for businesses, as ‘there is no executive task more vital and demanding’.
Proposition Two: Human Resource Polices Supporting Innovation will be Reported as Applied at the Locus of Everyday Organizational Action Our findings challenge this proposition, largely because of the disparate and unconnected nature of the different HR policy areas. Furthermore, there is still some clarification required as to where the locus of everyday organizational action is: whether it is at the individual or team level. The proposition is first reflected in recruitment and performance management practices at senior levels of the organizational hierarchy. For senior and other managerial grades, the attention, through use of assessment centre, group or individual
Volume 12
Number 1
March 2003
exercises, respectively, is on selecting someone who shows the right innovative behavioural competencies, using work sampling. On joining the firm these behaviours are concentrated through the use of individual performance targets. In terms of preferred recruitment methods at other levels of the organization, the picture is a little more inconsistent. Interviews and then personality tools are favoured across the different grades, with the exception of graduate recruitment. It is unclear why graduate recruitment should be something of an anomaly and may reflect the increasing application of assessment centre approaches (Anderson & Shackleton, 1993). In terms of other lower grades we see more differentiation between firms, based on their innovative focus and the HR tools they use. This may be linked to the greater proportion of staff at this level, or it may reflect the need to recognize and reinforce such behaviour far more consistently. In performance management, locus of action is clearly relevant, through the application of individual targets. This approach concentrates on extrinsic reward systems. However, there is some debate as towards which level of action efforts should be directed. Our data show a concentration of performance management at the individual level, which corroborates recent research findings. Armstrong (1996) laments the failure of organizations in the UK to use team-based reward systems as a means of encouraging and supporting innovation, indeed this survey showed that team-based approaches to idea reward and recognition were uncommon. This suggests a potential lack of coherence between the use of teams and the rewarding of such group endeavours. Axtell et al. (2000) claim that teams have an important role to play in enabling, particularly the implementation of, novel applications and ideas. This survey indicates that there is a similar paucity of integration in HR processes aimed at supporting ideas once they are generated. It seems that most regard ideas management as an integral part of an employee’s role and, therefore is not to be supported or rewarded in any particular way, as part of everyday performance management. Even those organizations for whom innovation is a priority do not recognize, outside of an appraisal, their staff’s endeavours. It is noteworthy that our observations concerning the lower relevance of training and ideas management to importance of innovation confirms the results of Felstead and Ashton (2000) and enriches from those of Sanz-Valle et al. (1999). The former argued that the fostering of new skills is not always associated with training, whilst the latter argued
© Blackwell Publishing Ltd, 2003
SUPPORTING INNOVATION THROUGH HR POLICY
that innovative organizations performed more product and process-based training actions. We see evidence of this in relation to the training of lower-grade staff in problem solving and brainstorming techniques, where innovation is stated as important.
Proposition Three: The External Fit of HR Policy Which Supports Innovation will have an Inverse Relationship with the Hierarchical Level at Which it is Applied Overall, this finding was confirmed, but, as with proposition two, the nature of the external fit referred to individual elements of HR policy rather than HR policy as a coherent, integrated unit. The proposition reflects the idea that HR policies supporting innovation when it is an articulated organizational strategy will be less common than HR policies which support innovation in the organization as a response to operational issues. This is reflected in the overwhelming focus on clerical staff who are the most likely recipients of training, and performance management in respect of innovation. Furthermore, cohesion or fit between HR policy and reported importance of innovation, although inconsistent, was noted at disparate points in the HR process, with selection and performance management, most frequently. Given recent research which addresses notions of integration, this is not altogether surprising. The inconsistent nature of these links between articulated innovation importance, HR policy and behaviour resonates strongly with Mueller’s (1996) argument that the organization’s social fabric, or ‘social architecture’ is responsible for differences in HR practices throughout the organization which evolve alongside strategic policy, relating to it recursively. It also strongly reflects HopeHailey et al.’s (1997) observation that HR practitioners are increasingly adopting a pragmatic ‘nuts and bolts’ approach to their organizational role. Furthermore, if we follow Campbell et al. (1996) who regard innovation as action, and invoke Felstead and Ashton’s (2000) arguments which hint at a time lag between the occurrence of innovative behaviour in the organization, and its subsequent codification and re-inforcement by HR departments, this patchy picture of coherent HR policy is not altogether unexpected. A further note should be made about the use of personality-based selection tools to promote the identification of innovation-related skills. Personality tools are found to be a differentiator between those who regard innovation as an important strategic endeavour and those who do not, and there is evidence, on closer
© Blackwell Publishing Ltd, 2003
59
examination, of a difficulty with this result. The organizations were asked to indicate which tools they used in their selection and recruitment programmes at each of the different levels. In analysing the tools deployed, most organizations use either the ‘16PF’ (Catell, Eber and Tatsuoka, 1970) or the ‘OPQ’ (Saville, 1973). Of these, only the latter purports to measure innovation potential, but only focuses on the quantity, rather than the quality of ideas. There is a wide range of tools available which identify the underlying motivational aspects (e.g. Kirton, 1976; Jackson, 1976; Patterson, 1999), crucial to the generation of radical ideas (Patterson, 1999), and none are mentioned in the sample. The survey suggests at best weak deployment, and at worst misapplication of personality tools for selection purposes.
Proposition Four: There will be Little Evidence that the Different HR Policy Elements will Show any Degree of Internal Fit It is clear from the previous discussion that there is little internal fit between different HR policy areas, and this is further complicated by different clusters of connected policy areas at different hierarchical levels. The greatest degree of fit is evident at the clerical level of innovation–conscious organizations. Here, most attention was paid to innovation in terms of selection, training and performance management. For managerial grades, attention is far more focused on selection and performance management, and there is a marked difference in the HR processes that are deployed. At senior-management level, only recruitment stands out as the most significant reflector of an innovation–focused HR policy.
Organizational Implications In response to the survey there are three main implications for organizations. First, the research reveals a hierarchical stratification of HR policy support for innovation. Innovation is important for most organizations surveyed, but it is revealed that attention tends to be focused mainly at the lower levels. In order to remove barriers to potential radical ideas, and to gain a more supportive climate for innovation, this stratification of policy needs to be removed so that all levels are involved in both the creation and the implementation of new ideas. Second, and related to the hierarchical nature of HR policy, we must question the potential role of innovation policy as a means of empowering employees. The survey indi-
Volume 12
Number 1
March 2003
60
CREATIVITY AND INNOVATION MANAGEMENT
cated that training and performance management are focused at the clerical level. Potentially this might enhance the generation of new ideas. However, without awareness training for those higher in the organization innovation is likely to fail. Firms where innovation is important need to ensure all staff understand the techniques and tools which support innovation processes in order that employees may be supported and assisted in their efforts. This requires the integration of innovation policy across the organization so that responsibility being imparted to one group of employees, does not result in its abdication higher up. Related to this, HR professionals should examine how far the policies designed to enhance innovation are integrated and not merely focused at just one group in the organization. Finally, both this research and others (for example Axtell et al., 2000) reveal that innovation requires teamwork in order at the very least to implement ideas. The survey revealed an over-reliance on individually focused procedures, which at best, do not enhance group working, and at worst, substantially undermine it. Individuals maybe best placed to generate new ideas, but it requires the support of others to have these ideas adopted and accommodated in any organization. Through implementing an integrated HR policy which creates understanding, acceptance and active support at all levels within an organization, a climate of sufficient trust and support will be fostered which ensures the emergence of a range of ideas-incorporating incremental to radical changes. Without these conditions, organizations will produce, at best, moderate improvements which will enable the status quo to be maintained. If organizations are really only seeking to maintain a stable state, then perhaps this is the reason why such a paucity of evidence exists which supports the deployment of integrated, non-hierarchical, team-based HR practices designed to promote innovation. Given our findings and their implications, our study has a limitation, which concerns the extent to which the survey instrument can seek to measure with any degree of accuracy the firms’ most senior HR professionals’ conceptions of innovation importance in the organization. Whilst an extensive discussion of the pros and cons of survey-based research is inappropriate here, the meaning systems proper to HR management and the extent to which HR managers over-report their involvement in organizational matters has been referred to in other work. In particular, work which specifically examines the rhetoric of HRM, the realities that this creates for HR
Volume 12
Number 1
March 2003
practitioners, and the congruity, or otherwise, of these realities with other organizational worlds, is of particular relevance. Examples of such work to date include inter alia an examination of performance management (Bowles and Coates, 1993), the psychological contract (Grant, 1999), and involvement in strategic decision-making (see Legge, 1995). It is also quite possible that the meanings attached by HR practitioners to innovation will vary – indeed we noted the differing organizational roles for innovation at the beginning of the paper – and this may have been reflected in our findings too. Whilst these debates are of clear importance in formulating research strategies for studying HR practitioners, we argue that in this instance, in order to get a broad sample of HR policy in respect of innovation, a survey-based approach was the most appropriate. Similarly, in terms of HR policy, we acknowledge that, in most cases, HR managers have final dictum over the formulation and implementation thereof, and as such would be the most appropriate organizational members to systematically interrogate as regards the characteristics of the whole range of HR policies and practices. Furthermore, if we reflect on (much debated) models which link, in however a piecemeal fashion, HR policy to strategy, HR managers are the lynchpin of policy formulation in terms of translating an espoused innovation strategy into policy objectives for the HR department. It may well be that our survey participants’ focus upon innovation measurement and development at the clerical level and lower were responding to organizational requests for low cost, systematic investment in this area, as opposed to the comparatively expensive development needs of middle and senior managers. More generally, our research shows that piecemeal translations and implementations of this kind are certainly in evidence, and that this is also hierarchically stratified. This is problematic in terms of the success of innovation, as Bouwen, de Visch and Steyaert (1992) found that part of the reason innovations fail to be implemented in organizations is through a lack of integration of understanding across different levels, so by concentrating efforts at one level of hierarchy, organizations create potential barriers at others. In terms of further research, we suggest that there is a need for survey work to be conducted in conjunction with in-depth case studies to establish meanings attributed to ‘innovation’ by HR practitioners, and the translation over time of this through policy and action. Such case studies would be multi level, examining strategic-level policy and
© Blackwell Publishing Ltd, 2003
SUPPORTING INNOVATION THROUGH HR POLICY
meaning formulation, observing HR policy in action, employee behaviours, their development and wider intrinsic and extrinsic reward systems (Amabile, 1988). Each stage of the research would have a focus upon the meaning attached to innovation and its translation through the organization. This research would be particularly reflexive upon the way in which individual and team approaches to innovation differed (as highlighted by Axtell et al., 2000). It would similarly have theoretical objectives as to how ‘integration’ of strategy, policy and practice might be addressed, taking account of the espoused need to show links between individual and organizational performance, but also the complexities of organizations’ social architectures, as highlighted by Mueller (1996). A further strand of future research would address the development and application of an innovation-specific competency. Its constitution, context-specificity, components and assessment all need to be examined.
Conclusion In this paper, using survey methodology, we have considered whether organisations coherently use HR policy to promote innovation. Our findings have established that organizations use a limited range of HR tools to identify and support innovation, through recruitment, training and performance management. We have found that our surveyed organizations, whilst attaching importance to innovation, fail to consistently translate and apply this in HR policy. This was particularly apparent in the hierarchically stratified application of the aforementioned tools at lower organizational levels. Thus, whilst organizations are encouraging their non-managerial employees to innovate, the lack of attention paid to middle and senior managers, (perhaps with an implicit expectation that they should innovate anyway) will sustain an HR policy approach fraught with resistance and misunderstanding about the implementation of new ideas.
References Amabile, T.M. (1988). A model of creativity and innovation in organisations. In Staw, B.M. and Cummings, L.L. (eds.), Research in Organisational Behaviour, 10. Jai Press, Greenwich, CT, pp. 123–167. Anderson, N. & Shackleton, V. (1993) Successful selection interviewing. Blackwell, Oxford. Armstrong, M. (1996) How group efforts can pay dividends. Personnel Management, 25 January.
© Blackwell Publishing Ltd, 2003
61
Axtell, C.M., Holman, D.J., Unsworth, K.L., Wall, T.D., Waterson, P.E. & Harrington, E. (2000) Shopfloor innovation: Facilitating the suggestion and implementation of ideas. Journal of Occupational and Organisational Psychology, 73(3), 265–286. Beer, M., Spector, B., Lawrence, P., Mills, D., and Walton, R. (1984) Managing Human Assets. Free Press, New York. Bessant, J. (1992) Big bang or continuous evolution ?: why incremental innovation is gaining attention in successful organisations. Creativity and innovation management, 1(2), 59–62. Boxall, P. (1996) The strategic HRM debate and the resource-based view of the firm. Human Resource Management Journal, 6(3) 59–75. Bowles, M.L. and Coates, G. (1993) Image and substance: The management of performance as rhetoric or reality? Personnel Review 22(2), 3–21. Bouwen, R., De Visch, J. & Steyaert, C. (1992) Innovation projects in organisations: complementing the dominant logic by organisational learning. In Hoskin, D.M. & Anderson, N. (eds), Organisational change & innovation: Psychological perspectives & practices in Europe. Routledge, London, pp. 123–148. Campbell, J.P., Gasser, M.B., and Oswald, F.L. (1996) The substantive nature of job performance variability. In Murphy, K.R. (ed.), Individual difference and behaviours in organisations. Jossey Bass, San Francisco, CA, pp. 258–299. Cattell, R.B., Eber, H.W. and Tatsuoka, M.M. (1970) The Sixteen Personality Factor Questionnaire. IPAT, Champaign. Colewell, J.B. (1996) Quite change – Big bang or catastrophic shift: At what point does continuous improvement become innovative. Creativity and Innovation Management, 5(1), 67–73. Felstead, A. & Ashton, D. (2000) Tracing the link: organizational structures and skill demands. Human Resource Management Journal, 10(3), 5–21. Forrester, R.H. (1994) Implications of lean manufacturing for human resource strategy. Work Study, 33(3), 20–24. Grant , D. (1999) HRM, rhetoric and the psychological contract: a case of ‘easier said than done’. The International Journal of Human Resource Management, 10(2), 327–350. Guest, D. (1987) Human Resource Management and industrial relations. Journal of management Studies, 24(5), 503–521 Guest, D. & Hoque, K. (1994) Yes, personnel does make a difference. Personnel Management, 26(11), 40–44. Guzzo, R.A. & Shea, G.P. (1992). Group performance and intergroup relations in organisations. In Dunnette, M.D. & Hough, L.M. (eds), Handbook Of Industrial And Organisational Psychology, 3. Consulting Psychologists Press, Palo Alto, CA, pp. 269–313. Hall, L. and Torrington, D. (1998) Letting go or holding on – the devolution of operational personnel activities. Human Resource Management Journal, 8(1) 41–55. Hope-Hailey, V., Gratton, L., McGovern, P., Stiles, P. and Truss, C. (1997) A chameleon function? HRM
Volume 12
Number 1
March 2003
62
CREATIVITY AND INNOVATION MANAGEMENT
in the ‘90s. Human Resource Management Journal, 7(3), 5–18. Jackson, M.J. (1976) Jackson personality inventory manual. Research Psychologist Press, Goshen NY. Kirton, M.J. (1967). Some notes on the dynamics of resistance to change. The defender role. In Watson, G. (ed.), Concepts for Social Change. National Training Laboratories, Washington DC. Legge, K. (1995) Human Resource Management: Rhetoric and Realities. Macmillan, Basingstoke. MacDuffie, J.P. (1995) Human Resource Bundles and Manufacturing Performance: Organizational Logic and Flexible Production Systems in the World Auto Industry. Industrial & Labor Relations Review, 48(2), 197. Mueller, F. (1996) Human Resources as Strategic Assets: An Evolutionary Resource – Based View. Journal of Management Studies, 33(6). Nadler, D.A. & Tushman, M.L. (1997) Implementing new designs: Managing organisational change. In Anderson, P. (ed.), Managing strategic innovation and change: A collection of readings. Open University Press, London. Oldham, G.R. & Cummings, A. (1996) Employee Creativity: Personal And Contextual Factors At Work. Academy Of Management Journal, 39, 607–634. Patterson, F.C. (1999) The Team potential Indicator Manual. Oxford Psychologist Press, Oxford. Paauwe, J. & Richardson, R. (1997) Introduction: special issue: Strategic human resource management and performance. The International Journal of Human Resource Management, 8(3), 257–262.
Volume 12
Number 1
March 2003
Sanz-Valle, R., Sabater-Sánchez, R. & AragónSánchez, A. (1999) Human resource management and business strategy links: an empirical study. The International Journal of Human Resource Management, 10(4), 655–671. Saville, P. (1973) Occupational Personality Questionnaire. SHL, London. Schuler, R.S. & Jackson, S.E. (1987) Linking competition strategies with human resource management practices, Academy of Management Executive, 1, 207–219. Schumpeter, J. (1934) The theory of economic development Harvard University Press, Boston. Storey, J. (1992) Developments in the management of human resources. Blackwell, Oxford. Tyson, S. (1999) How HR knowledge contributes to organizational performance. Human Resource Management Journal, 9(3), 42–52. Wolf, R.A. (1994) Organisational innovation: Review, critique and suggested research directions. Journal of Management Studies, 31(3), 405–431.
Rosalind H. Searle is in the Department of Psychology, Open University, Milton Keynes, UK. e-mail:
[email protected] Kirstie S. Ball is at the Department of Commerce, Birmingham Business School, University of Birmingham, Birmingham, B15 2TT, UK.
© Blackwell Publishing Ltd, 2003
