Innovation Roles: From Souls of Fire to Devil's Advocates.
Keywords: Champions, Devil's Advocates, Health Communication, Innovation, Roles
Much of the existing research on the phenomenon of innovation has been summarized by Rogers in his review of the diffusion literature (Rogers, 1995). Rogers defined innovation as "an idea, practice, or object that is perceived as new by an individual or other unit of adoption" (1995, p. 11). More generally, innovation can be conceived of as "the adoption of an internally generated or purchased device, system, policy, program, process, product, or service that is new to the adopting organization" (Damanpour, 1991, p. 556). Whether innovations are actually new or just perceived to be new, they usually are geared to produce a desired end-state. As Kanter pointed out, "innovation is the process of bringing new ideas into productive use" (Kanter, Kao, & Wiersema, 1997, P. 20). However, because organizations exist in highly complex, uncertain environments (Weick, 1995), innovations frequently result in a number of unanticipated outcomes. Ultimately, the extent to which organizational members perceive advantages or disadvantages associated with innovation outcomes is determined by their subjective experiences, which are rooted in communication.
This study examines whether organizational members in various innovation roles differ significantly with respect to their perceptions about the pros and cons of innovation and their levels of innovation-related communication. This work begins with an overview of the roles that organizational members play in the innovation process. Subsequently, it tests the prediction that those who play central roles in the innovation process will be more likely to have favorable attitudes toward the innovation and higher levels of innovation-related communication than those who play peripheral roles in the innovation process. Finally, a discussion of the results provides insight about the nature of resisting innovation, elaborates on existing innovation role typologies, and explores the role of communication in taking ownership of the innovation process.
Several factors can lead to differences in perceptions of innovation among organizational members. For example, differences in the acceptance of innovations are often based upon organizational members' levels within the hierarchy, seniority, experience, and unit affiliations (Kossek, 1989). One reason for these observed differences may be that distinct groups of stakeholders have differing interests in the evaluation process (Roberts & Bradley, 1991; Weiss, 1983). In an effort to evaluate innovation outcomes from their own perspectives, various groups focus on different data in their assessment of outcomes (Ashmos, McDaniel, & Duchon, 1990; Brimm, 1988). As a result, different groups' perceptions of the same innovation may vary depending on their stake and role in the innovation process (King, 1990).
Organizational innovation requires the fulfillment of specific key roles that guide a new idea through the innovation process. These roles are carried out by members of the organization, and are commonly referred to as idea generators, sponsors, and orchestrators (Galbraith, 1984). In smaller organizations or those that are implementing a variety of innovations, however, there may be more innovation roles than there are adopters. This shortage of organizational members to fill innovation roles may be exacerbated by a lack of slack resources (Nohria & Gulati, 1996).
Idea generators create the innovative ideas that could be of potential use to the organization (Galbraith, 1984). Idea generators initiate innovation by reformulating a particular problem through a creative perspective that they are willing to promote within the organization (Brimm, 1988). In organizations with informally generated innovations (Johnson, 1993), idea generators are usually low-level staff who are close enough to the problem to create an innovative solution. Because of their low status, idea generators require sponsors to help promote the idea in the organization. In formally generated innovations, however, higher status idea generators may be their own chief advocates.
The sponsor, or idea champion (Galbraith, 1984; Howell & Higgins, 1990; Rogers, 1995), is usually a management-level person who recognizes the usefulness of the idea to the organization and lends authority and resources to the innovation throughout its development and implementation. The idea champion, or "soul-of-fire" (Stjernberg & Philips, 1993), of an innovation plays a significant role in gaining organizational acceptance of the innovation. In order to promote innovations, champions tend to initiate more influence attempts and employ a greater variety of influence tactics than do non-champions (Howell & Higgins, 1990). In addition, champions demonstrate their commitment to a particular innovation through a personal identification with the innovation and its outcomes (Brimm, 1988).
The third role needed in the innovation process is that of the orchestrator (Galbraith, 1984), who is likely to be a central player in innovation-related communication networks. Innovations are often controversial and may be perceived as impinging upon territorial rights and personal investments of others within the organization. Therefore, orchestrators are needed to maneuver the innovation through the organization's political process. Akin to change agents (Armenakis, Harris, & Mossholder, 1993; Rogers, 1995; Tichy, 1974), orchestrators create a climate for change and influence others to adopt the innovation. The orchestrator must protect the innovation process by supporting idea generators, finding sponsors for innovations, and promoting the trial period and testing of innovative ideas. Because the organization's political process is biased toward those who have power, prominent organizational members are likely to be effective orchestrators. Indeed, Mitchell, Agle, and Wood (1997) identify power as a sal ient characteristic of stakeholder influence that enables orchestrators to use their authority and resources to promote the innovation process.
Although innovation roles are informal positions that may be assumed by individuals throughout the organization, people in certain functional positions may be more likely candidates for key innovation roles than may others. Indeed, despite prescriptions for considering innovation processes separately from standard organizational operating procedures (Galbraith, 1984), it generally has been argued that communication patterns and perceptions are dependent on hierarchical positions or roles in organizations (Jablin, 1987). Thus, innovation roles and functional positions are likely to overlap, particularly in a milieu characterized by a lack of slack resources and a culture where commitment to service takes precedence over innovation (Meyer, 1996).
In sum, formal or informal groups may perform different roles in the innovation process. Because of varying experiences related to an innovation, groups may form distinctly different perceptions of innovation characteristics. Because various groups perform discrete innovation-related functions, they may also exhibit different communication patterns.
Historically, researchers have described innovations in terms of their attributes, or perceived characteristics. Rogers (1995) reviewed extant research that focused on the effects of innovation attributes (e.g., relative advantage, compatibility, complexity, observability, and trialability) on innovation adoption. However, much of the empirical research on innovation attributes is characterized by inconsistent conceptual and operational definitions, as well as substantial measurement limitations (see Meyer, 1996, for a critique). In response to these shortcomings, this study uses an alternate conceptual framework, that of Meyer's (1996) pros and cons of innovation. The pros and cons of innovation are overarching constructs that subsume several innovation attributes, which share common features to the degree that they may not be mutually exclusive. Unlike traditional innovation attributes, then, the pros and cons of innovation are innovation characteristics that demonstrate construct validity as well as face validity (see Meyer, 1996 for a detailed discussion of validity issues).
The pros of innovation indicate the positive aspects or advantages that organizational members associate with an innovation (Lewis & Seibold, 1996; Meyer, 1996; Rakowski, Dube, Marcus, Prochaska, Velicer, & Abrams, 1992; Weenig, 1999). Essentially the extent to which organizational members think that an innovation is a good idea, pros are linked closely to the traditional innovation attribute of relative advantage. Pros may encompass other attributes that are positively related to innovation adoption, such as observability, adaptability, compatibility, and trialability. Pros may also gauge the level of acceptance for an innovation: Organizational members who report high levels of pros associated with an intervention are likely to support or "buy in" to that innovation.
The cons of an innovation can best be understood in contrast to its pros. Cons are negative aspects or disadvantages that organizational members associate with an innovation (Lewis & Seibold, 1996; Meyer, 1996; Rakowski et al., 1992; Weenig, 1999). Closely related to the innovation attributes of complexity and risk, cons tap the extent to which organizational members perceive that there are drawbacks or negative consequences associated with an innovation. Cons also may gauge the level of resistance to change: Organizational members who report high levels of cons associated with an intervention may fail to support the innovation, or even sabotage it.
Intuitively, the strength of a social tie is a function of time, emotional intensity, intimacy, and reciprocity (Granovetter, 1973). Weak ties refer to our less developed relationships which are more limited in space, place, time, and depth of emotional bonds (Adelman, Parks, & Albrecht, 1987; Weimann, 1983). The "strength of weak ties," derived from the work of Granovetter (1973) is perhaps the most well-known concept related to network analysis (Granovetter, 1982). According to Granovetter, the most useful information comes from individuals in a person's extended network: casual acquaintances and friends of friends. This information is the most useful precisely because it comes from infrequent or weak contacts. Contemporary network theory extends classic work on the informational benefits of weak ties (Granovetter, 1973, 1982) to delineate the control benefits of structural holes (Burt, 1992, 1997). As it is argued here, the informational and control benefits of informal communication structure have a numb er of important implications for organizational innovation.
Weak ties are akin to structural holes, or "the relationship of nonredundancy between two contacts" (Burt, 1992, p. 65). In essence, a weak tie spans a structural hole in a communication network. In addition to the well-noted information benefits of weak ties or structural holes, Burt suggests that these special ties also offer control benefits which give certain players strategic advantages, such as early promotion and higher bonuses (Burt, 1997). In the context of innovation, control benefits mean that individuals who have many weak ties can capitalize on the existence of structural holes around them: As they are intimately aware of innovation efforts, they have the opportunity to become strategically involved in innovation processes. Subsequently, this ability enables them, as change agents (Rogers, 1995), to influence others in the innovation process through their opinions and actions.
The literature suggests that organizational members in different innovation roles would form distinctly different perceptions of the pros and cons of innovation and varying patterns of weak ties within the innovation-related communication network. In general, those who play central roles in the innovation process (i.e., idea generators, idea champions, and orchestrators) will be more likely than those who play peripheral roles to have favorable attitudes toward the innovation and high levels of innovation-related communication. The focus of this investigation is to determine whether organizational members in various innovation roles differ significantly with respect to their perceptions about the pros and cons of innovation and their levels of innovation-related communication.
H1: Idea generators and idea champions will report higher levels of pros and lower levels of cons associated with innovation than will other organizational members.
H2: Orchestrators will report higher levels of communication contacts (e.g., more weak ties) associated with innovation than will other organizational members.
This study takes place in the Cancer Information Service (CIS), a federal government health information and education network that is piloting innovations through a new organizational form to provide cancer information to the public. The CIS was implemented in 1975 by the National Cancer Institute (NCI) to disseminate accurate, up-to-date information about cancer to the U.S. public (Morra, Van Nevel, Nealon, Mazan, & Thomsen, 1993). Over time, the CIS has become a laboratory for state-of-the-science communication research (Marcus, Woodworth, & Strick-land, 1993). The Cancer Information Service Research Consortium (CISRC) is currently involved in the process of generating, piloting, and evaluating communication and health behavior innovations designed to provide cancer information to the U.S. public (Marcus, 1998). In these endeavors, the CIS must be creative in its attempts to manage innovation to generate organizational members' acceptance of change that at times may be challenged by geographic and institut ional barriers.
At the time of this study, the CIS was in the process of piloting three new intervention strategies (i.e., innovations) designed to disseminate cancer information to the public. See Marcus et al. (1998) for a complete description of the three program projects and Johnson and Chang (2000) for another study based on data from the larger study. My study focuses on perceptions of Project 2, the only innovation that was piloted from start to finish during the course of the data collection whose topic remained constant. (See Figure 1 for Project 2 timeline.) To encourage women to receive regular mammograms, Project 2 made cold calls from the CIS to low income and minority women in targeted communities in Colorado. This activity was substantially different from traditional service delivery, in which callers reach CIS telephone information specialists by dialing 1-800-4-CANCER. Because this innovation involved a considerable departure from the network's historic mission (Meyer, Johnson, & Ethington, 1997), the CIS d ecided to pilot it in only one regional office. During the time of this study, CIS members were aware that the decision whether to adopt and implement the innovation system-wide would be made only after all program evaluation data were evaluated.
The participants in this study (N = 90) were highly educated. At baseline, 94 percent of respondents had earned college degrees, and 62 held graduate degrees. The majority of respondents were low in tenure: Nearly two-thirds had worked for the CIS for fewer than five years. This study focused on key stakeholders within the CIS: the Office of Cancer Information, Communication, and Education (OCICE) staff at NCI (n = 12), Principal Investigators (PIs; n = 9), Project Directors (PDs; n = 20), Telephone Service Managers (TSMs; n = 20), and Outreach Program Managers (OPMs; n = 22) at the regional CIS offices; and CISRC Program Project staff (PP; n = 7), who conduct research and evaluation related to new intervention strategies.
Members of the stakeholder groups could be identified as holding a variety of innovation roles. For example, Program Project staff simultaneously enact dual roles: First, as idea generators, they conduct research and evaluation related to new intervention strategies; second, as orchestrators, they play an important role in building support for innovation by developing and maintaining an innovation-related communication structure across the network. Next, the OCICE staff and Project Directors can be seen as idea champions for innovation: The Office of Cancer Information, Communication, and Education (OCICE) staff is comprised of officials at the policy level who are most involved in centralized decision-making processes related to innovation adoption and implementation; Project Directors are officials at the regional level who have day-to-day responsibility for managing the CIS as it pilots new intervention strategies. Although people in other functional roles (e.g., program officials at the local level, incl uding Telephone Service Managers, Outreach Program Managers, and Principal Investigators) do not hold central innovation roles in this study, they are nevertheless important stakeholders in the innovation process: They are concerned about how innovation implementation will affect their organization's day-to-day operation, especially the effectiveness with which they deliver the services that they provide to the public. 
Respondents received a mailing that consisted of a network questionnaire and a battery of self-report questions relating to innovation. Prior to completing the surveys, participants gave informed consent in a manner consistent with APA guidelines (APA, 1992). To make completion more likely, the surveys were sent to the respondents approximately ten days before the sampling period along with a self-addressed, stamped envelope. Many follow-up steps (e.g., letters, faxes, e-mails) recommended in the literature (e.g., Dillman, 1978, 1991) were also taken to increase response levels. These extensive follow-up efforts resulted in very high response rates (96% and 85%).
The variables in the model were measured in two ways: self-report questionnaires focusing on innovation and classic self-report network analysis. Collaboration with members of the network led to the decision to gather self-report data on perceptions of innovation characteristics over a 15-month period. Baseline measures of respondents' perceptions of innovation characteristics were gathered soon after the main pilot study for Project 2 was implemented. Follow-up measures of perceptions of innovation characteristics were gathered after the main pilot study was completed, at the time of the six-month follow-up (see Figure 1).
Network data were gathered during a three-day data collection period that coincided with the collection of self-report survey data. Network data were gathered via self-report communication logs. Only innovation-related interpersonal contacts, either face-to-face or by telephone, that were initiated or received during the three day sampling interval were recorded in the communication log.  Using predated forms and a directory of all members of the network, respondents were asked to record the complete names of individuals with whom they had contact during the data collection period.
The procedure by which self-report questionnaires were developed followed the stages suggested by Devellis (1991). The measurement model was evaluated with the confirmatory factor analysis subroutine of the PACKAGE computer program (Hunter & Lim, 1987).  For a complete discussion of instrument design and evaluation of the measurement model, see Meyer, 1996.
The pros and cons of innovation were measured by eleven-point Likert type scales designed to tap perceptions of innovation attributes salient to Project 2. For example, in the pro scale, respondents were asked to indicate how much they agreed or disagreed with the statement, "This intervention strategy is a positive way to reach members of a target audience," whereas in the con scale, respondents indicated how much they agreed or disagreed with the statement, "This intervention strategy means that information specialists can't meet other job responsibilities as well as they used to." Confirmatory factor analysis (Hunter & Lim, 1987) results indicate that the 8-item pro and 3-item con scales demonstrated internal consistency and parallelism ([x.sup.2] = 74.5, df = 79). At Ti and T2, standardized alphas were .96 and .95 for pros, and .88 and .77 for cons, respectively.
Weak ties were measured with STRUCTURE's (Burt, 1991) indices of range and prominence.  The range measures of contacts and nonredundant contacts reflect the breadth of an individual's network ties, whereas the prominence measure of choice status reflects the proportion of an actor's contacts from others in the network to the total number of contacts possible. In other words, those organizational members who are strategically positioned in the network (in terms of range and prominence) demonstrate higher levels of weak ties than will those who are less advantageously situated. Because weak ties generally have been found in work-related, single-content networks that exist outside one's primary group (Albrecht & Ropp, 1984; Granovetter, 1982) and because the innovation-related contacts in this organization occurred primarily among individuals who were located in 19 geographically-dispersed regional offices across the country, range and prominence indices demonstrated face validity for measuring weak ties in this study. In addition, confirmatory factor analysis (Hunter & Lim, 1987) results indicate that the weak ties measure demonstrated internal consistency and parallelism ([x.sup.2] 8.2, df = 9). Standardized alphas for the 3-item Weak Ties scale were .96 and .90, respectively.
Preliminary analyses of the scales' mean, standard deviation, and Pearson correlations are reported as descriptive statistics in Table 1. The factor with the lowest average item mean was weak ties at T2 (.57), a sizable decrease from the baseline level of 4.85. An exploratory paired t-test  indicated that the decline in weak ties from 4.85 at T1 to .57 at T2 was significant (t = 5.37, df 81, p =.000).  The pro scale at T1 had the highest average item mean (6.69). Over time, there was a nonsignificant decrease in the pros associated with Making Outcalls from T1 (6.69) to T2 (6.01) and a concurrent nonsignificant increase in cons (4.72) to (5.67). Three significant correlations existed between variables: pros and cons had a strong negative correlation with one another at T2 (r = - .42); pros T1 and cons T2 were negatively correlated (r = -.39); and weak ties T1 was negatively correlated with pros T2 (- .35). The scales had high standard deviations, an indication of between-groups variance.
Analysis of variance was employed to determine if there were significant differences in perceptions about the pros and cons of innovation and levels of weak ties among organizational members in different innovation roles. Results indicated that there were several significant differences in how people in different innovation roles perceived the innovation, providing partial support for H1: Idea generators (Program Project staff) reported higher levels of pros at T1 and T2 and lower levels of cons at T2 than did other organizational members. Unexpectedly, however, idea champions (Project Directors and OCICE staff) exhibited significantly higher levels of cons and lower levels of pros than did idea generators (Program Project staff). With respect to H2, the data failed to support predicted differences in levels of innovation-related communication contacts. Although results indicated that there were significant group differences in informal communication structure, the differences were not in the expected direct ion: Orchestrators of innovation (Program Project staff) did not report higher levels of weak ties than did other organizational members. Instead, idea champions (Project Directors) and Outreach Program Managers generally reported more weak ties than did others. These functional role differences provide evidence for the existence of particular innovation roles associated with formal organizational structures and have a number of interesting implications for the study of innovation.
The results of this study suggest that organizational members play a variety of roles in the innovation process. When compared to other organizational members, idea generators (Program Project staff) reported higher levels of pros at T1 and T2 and lower levels of cons at T2. This indicates that perhaps idea generators have "bought in" to the innovation process more than have other organizational members. Counter-intuitively, idea champions (Project Directors and OCICE staff) did not report significantly higher levels of pros and lower levels of cons than did other organizational members. In fact, at two points in time, they reported considerably different views about the pros and cons of innovation than did idea generators. What might contribute to the discrepancy among idea champions' and orchestrators' perceptions of innovation?
Historical events may shed some light on the reasons behind idea champions' less favorable evaluations of the innovation (see Meyer, 1996, for a chronology). Process evaluation data released in national meetings during the summer and fall of 1995 cited evidence of low job satisfaction among Telephone Information Specialists and low cost effectiveness associated with Making Outcalls. This preliminary evaluation information may have highlighted negative unintended consequences of innovation for Project Directors, who may have been concerned with the negative impact of this innovation on staff and service delivery, and for OCICE staff, who may have questioned the feasibility of demonstrating significant outcomes associated with the intervention at a reasonable price.
Whatever the reason for the idea champions' unfavorable response to this innovation, the discrepancy among idea champions' and orchestrators' perceptions of innovation is thought-provoking. Interestingly, a document released by the CISRC leadership does not reflect such attitudinal dissonance:
From the outset, the plan was to mobilize and recruit CIS Project Directors to serve as "idea champions." ... Fortunately for the CISRC, the CIS Project Directors embraced this challenge and became highly effective idea champions within their organizations. (Marcus et al., 1998, p. S13)
This claim, when coupled with the fact that Program Project staff did not report significantly more weak ties than other organizational members, may indicate that the orchestrators were initially somewhat unaware that practitioners were appropriating their innovation--a common theme in organizational innovation research (Albrecht & Bach, 1997; Kanter, 1983). Alternatively, this finding could indicate that Program Project staff are primarily idea generators and not orchestrators. Finally, it is possible that diminished levels of weak ties reflect the paucity of innovation-related communication in general (Albrecht & Ropp, 1984; Farace & Johnson, 1974; Monge, Cozzens, & Contractor, 1992; Johnson, 1993) or, more specifically, the natural life cycle of this innovation's pilot phase, which came to a close toward the end of the data collection period (Meyer, 1996).
Kanter (1988) noted that most research about organizational innovation is characterized by an implicit assumption that innovation is a good thing. Similarly, Van de Ven observed,
innovation is often viewed as a good thing because the new idea must be useful--profitable, constructive, or solve a problem. New ideas that are not perceived as useful are not normally called innovations: They are usually called mistakes. (1986, p. 592)
Kanter found that it was extremely difficult to identify innovations that failed (1988).
Kanter (1988) concluded that the organizational context has a major impact on the conceptualization of innovation within an organization. According to Kanter, the difficulty inherent in identifying the failure to innovate can be attributed to one or more of the following reasons: the taboo nature of mentioning failure; the threatening nature of failure that discourages risky ventures; the multiple goals of many projects that prevent them almost by definition from being total failures; and the strategies by which clever innovators convert potential "failures" into minor successes. In the case of the CISRC, perhaps the leadership's hesitancy to acknowledge the discrepancy among idea champions' and orchestrators' perceptions of innovation reflects a natural tendency to avoid labeling an innovation as a mistake. This tendency may be especially pronounced in idea generators who have a vested interest in the successful implementation and dissemination of their innovations.
Likewise, resistance to organizational innovation usually carries negative connotations (Albrecht & Bach, 1997). However, resistance to change is not bad in and of itself--in fact, individuals may have very sound reasons for resisting an innovation. When a higher status unit has the authority to adopt an innovation that another part of the organization must implement (Fidler & Johnson, 1984), resistance may send a message that management is insensitive to organizational members' needs. On a deeper level, resistance may symbolize dissatisfaction with the organization's innovation-related decision-making structure. Ultimately, however, when potential idea champions raise concerns about an innovation, orchestrators need to listen. Stakeholders' valuable input can be used by managers to modify innovations mid-stream or reconsider their communication strategies for promoting innovation.
Implications for Further Research
The significant functional role differences in perceptions of the pros and cons of innovation and weak ties suggest that distinct groups play key roles in the innovation process; further, that these groups have varying perceptions of the utility of innovations. In his work on rethinking technology transfer, Dearing (1993) stresses the importance of measuring effectiveness by comparing sources' and receivers' perceptions of utility. Future research needs to explore more systematically the factors that contribute to the discrepancy in perceptions of utility among key stakeholders in the process of innovation, paying special attention to the features of the structural, cultural, and political environment in which innovation occurs.
Interestingly, unexpected findings suggest that those who have the power to champion innovation are also capable of challenging it. In light of these findings, researchers might consider the utility of adding a new character to Galbraith's (1984) existing cast of innovation roles: The role of the "devil's advocate" might offer a standpoint for those who voice objections to innovation adoption or implementation for very legitimate reasons. Expanding the innovation role typology to include the devil's advocate addresses Kanter's (1988) criticism that past innovation role research has a managerial or pro-innovation bias.
Moreover, an examination of the role of the devil's advocate has a great deal of heuristic value for future communication research about organizational innovation. Despite the fact that the role of the devil's advocate has an extensive theoretical and empirical base in small group research (e.g., Janis, 1983; Pavitt, 1994), there is a paucity of scholarly research about the role of the devil's advocate in organizational innovation processes. Although essays in non-scholarly journals speculate that the devil's advocate can play an important part in generating thoughtful criticism about new proposals and new technology (e.g., Foegen, 1988; Woodruff, 1991), up to this point there has not been a systematic program of scientific inquiry designed to test the extent to which theoretically-derived propositions are supported by empirical observations.
There are a variety of possible directions for future scholarly research about the role of the devil's advocate in the process of innovation. First, researchers might ask how organizational members define the role of the devil's advocate. Pavitt and Curtis' conceptualization of the devil's advocate as "a person who may not disagree with the group consensus but who does not think that the agreed-upon proposal has undergone enough examination" (1994, p. 194) assumes that certain conditions exist: Notably, a group consensus about a proposal and a perceived need for additional examination of the facts and assumptions that have led to consensus. One might ask whether these conditions, which are salient to the context of group decision making, are necessary and sufficient terms with which to define the role of the devil's advocate in the innovation process.
Second, scholars might examine the origin of the devil's advocate role. For example, is the role of the devil's advocate assigned by a group leader, as Janis (1983) suggests? Or rather, does the role of the devil's advocate arise from conditions in which an organizational member feels compelled to voice resistance to an innovation for legitimate reasons? Is this role one that rotates among different members in the group (Janis, 1983) over time, or is it monopolized by a certain organizational member who is most adept at playing the role? What constitutes an "effective" devil's advocate? Do the same characteristics that reduce groupthink also lead to innovation? Is a devil's advocate more desirable in some phases of innovation (e.g., piloting) than in others (e.g., idea generation)?
Third, researchers might investigate various communication choices used to perform the role of the devil's advocate in innovation processes. Although the concept of weak ties is a useful one for the study of innovation, most operational definitions are rather limited in that they reduce the rich, complex, emotional phenomenon of communication to a numerical indicator of communication contacts (Weenig, 1999). Alternatively, scholars could examine the efficacy of various communication styles (e.g., argumentativeness vs. verbal aggressiveness) or compliance-gaining strategies (e.g., reason vs. assertiveness) employed by devil's advocates during the innovation process. How are these communication choices perceived by other organizational members? Are devil's advocates perceived as team players in the innovation process or adversaries against whom other (powerful) organizational members form coalitions to achieve discursive closure (Deetz, 1992)?
Finally, what is the role of power in determining the effectiveness of devil's advocates? If one conceptualizes power in its broadest sense (i.e., power derived from social roles and interpersonal power as well as the more traditional notion of positional power), one can ask how organizational members respond to devil's advocates of different gender roles, sexual orientation, race/ethnicity, class, or able-bodiedness. How does an organization's cultural context, including prevailing assumptions about the nature of gender roles, sexual orientation, etc., affect organizational members' perceptions about the credibility of devil's advocates who are labeled as "others"? Answering these types of questions through empirical research can offer a great deal of insight for communication scholars who are interested in exploring the practical implications of theorizing about organizational innovation.
Implications for Practice
Organizational stakeholders' perceptions about the advantages and disadvantages associated with innovations may provide managers of innovations with an assessment procedure (Dearing & Meyer, 1994) that can determine the extent to which individuals are supportive of new ways of doing things. Organizational members who "buy in" to the innovation implementation process become more active, enthusiastic participants because they perceive that there are certain advantages to doing so. In contrast, stakeholders who perceive high levels of disadvantage associated with an innovation may resist change (Albrecht & Bach, 1997). Although resistance to change is not necessarily a bad thing, it can result in dysfunctional outcomes (e.g., half-hearted implementation, delays, and so on) if organizational members do not perceive that their voices are heard in the dialogue about innovation.
Finally, the findings of this study provide direction for researchers and organizational members who are collaborating in the process of piloting innovations. The CISRO exemplifies an innovative strategic alliance between researchers and practitioners within a geographically-dispersed network. In this context, researchers play an important role as idea generators responsible for developing new products or processes. Key organizational members enact the roles of idea champions, orchestrators, and devil's advocates. As idea champions, highly influential individuals promote the innovation by lending support and resources to the innovation. As orchestrators, prominent individuals build support for the innovation by creating an innovation-related communication structure across the network. As devil's advocates, knowledgeable organizational members question the legitimacy of an innovation.
In their roles as orchestrators of innovation, prominent organizational members need to develop competent conflict management skills in order to mediate conflicts that may arise as a result of competing interests between groups of stakeholders. Given that researchers and practitioners represent distinct groups of stakeholders in the innovation process, they are likely to have differing interests in the innovation process (Weiss, 1983). For example, practitioners may be primarily concerned with the practical consequences of innovation for their organization, whereas researchers, having invested a great deal of ownership in their ideas, may be chiefly interested in testing theory and its generalizability to other contexts. Although involving multiple stakeholders in the process of piloting innovations can lead to unproductive conflict that may bog stakeholders down in time-consuming discussion and debates, it can also democratize access to innovation-related information (Weiss, 1983). Clearly, productive confl ict among organizational members, from "souls of fire" to "devil's advocates," can enhance an organization's ability to take ownership of the innovation process.
This article is based in part on the author's doctoral dissertation, which was completed under the direction of J. David Johnson in Michigan State University's Department of Communication and earned the 1996 W. Charles Redding Dissertation Award from the International Communication Association. An earlier version of this paper was presented to the Organizational Communication Division of the 1998 National Communication Association Annual Convention in New York City. The research was supported by Subcontract No. 737-4241 from the AMC Cancer Research Center for the P01 CA57586-01A1 grant from the National Cancer Institute. The conclusions reached in this paper are solely those of the author and do not necessarily reflect the views of the National Cancer Institute (NCI), the Cancer Information Service Research Consortium (CISRC), or the AMC Research Cancer Center. I would like to thank the Office of Cancer Information, Communication, and Education/National Cancer Institute project staff and the members of the 19 regional CIS offices who participated in the data collection. I would also like to thank the members of the Network Analysis Advisory Board for their help throughout the many phases of the research process: Donna Cox, Jo Beth Speyer, William Stengle, Marsha Woodworth, Maureen McClatchey, and Diane Ruesch. I am particularly thankful for data management support from Team for Evaluation & Audit Methods members Betty LaFrance, Hui-Jung Chang, and Caroline Ethington. Thanks also to Deb Tigner for her help in preparing and mailing the questionnaires. Finally, special thanks to Dave Johnson for his feedback on earlier drafts of this paper.
Marcy Meyer is an associate professor in the Department of Communication Studies at Ball State University.
(1.) Organizational members who play central roles in the innovation process are likely to have more favorable attitudes toward innovations than do other stakeholders: Key players tend to "buy in" to the innovation process because they have a great deal at stake. This tendency may be particularly pronounced within the larger political context of this research: At the time of this study, CIS program priorities were being evaluated within an environment of increased fiscal constraints. At the same time, the role of the CIS as a research laboratory in support of NCI's research mission was being explored. One implicit understanding related to the research was that the results could be used to demonstrate that the CISRC had the potential to become a research arm of the CIS (Marcus, 1998).
(2.) Despite an extensive pretest, the communication log was modified twice during the course of the research project, in response to issues raised by respondents (see LaFrance, Johnson, Ethington, & Meyer, 1997, for the actual communication logs). Nine months into the research project, the communication log was modified to focus on communication that occurred at the national level between individuals in different regional offices. Three months later, the communication log was changed to include communication conducted by facsimile and electronic mail. The first change may have reduced the level of communication reported across the network, especially intraoffice and other work-related communication. The second change may have led respondents to substitute one mode of communication for a variety of others. As this study is concerned with interpersonal innovation-related communication contacts only, the impacts of the change in measurement instruments on communication structure is minimal.
(3.) Confirmatory factor analysis is a superior technique when the a priori specification of items expected to cluster together is possible (Fink & Monge, 1985; Hunter & Gerbing, 1982). This approach also represents an alternative approach for dealing with multicollinearity in network data (Kilduff & Krackhardt, 1994). Three criteria proposed by Hunter (Hunter, 1980; Hunter & Gerbing, 1982) were used to determine unidimensionality: homogeneity of item content (i.e., face validity); internal consistency (i.e., the correlation between indicators of the same underlying construct is equal to the correlation of their factor loadings); and parallelism (i.e., the correlation between two separate constructs is the product of their factor loadings and the correlation between constructs). Tests of unidimensionality are essential to scale development because it has been demonstrated that Cronbach's coefficient alpha provides an unbiased estimate of reliability only if scale items are unidimensional (Hunter, 1980; Hunte r & Gerbing, 1982).
(4.) STRUCTURE is based on the work of the sociologist Ron Burt (1991), and has explicit linkages to his theoretical work (e.g., Burt, 1982, 1992). STRUCTURE calculates a large number of indices related to an individual's structural positioning within the network. Structural Autonomy indices, revealed in part in range measures, relate to the extent to which an individual's relationships may constrain his/her opportunities for individual action within a network. Prominence indices measure the extent to which an Individual is powerful in terms of the degree to which his or her communication is in demand from other network members.
(5.) Paired t-tests are used to determine whether there is a significant difference in subjects' scores when they are measured at two points in time (Frey, Botan, & Kreps, 2000). This procedure is simply an alternative to conducting a repeated measures ANOVA with two levels, appropriate for use when N is less than or equal to 100 (Stacks & Hocking, 1999).
(6.) The significant decline in weak ties may be in part a measurement artifact caused by the change in the communication log discussed in note two. As the central purpose of this study is not to test for significant differences in weak ties over time, this measurement artifact should have no effect on the confidence with which the reader interprets findings related to the research hypotheses of this study. See Meyer (1996) for a complete discussion of the changes in weak ties over four points in time, as they relate to the test of a longitudinal model of the effects of weak ties on perceived organizational innovativeness and innovation characteristics.
Adelman, M. B., Parks, M. R., & Albrecht, T. L. (1987). Beyond close relationships: Support in weak ties. In T. L. Albrecht & M. B. Adelman (Eds.), Communicating social support (Vp. 126-147). Newbury Park, CA: Sage.
Albrecht, T., & Bach, B. (1997). Communication in complex organizations: A relational approach. Orlando, FL: Harcourt Brace.
Albrecht, T. L., & Ropp, V. A. (1984). Communicating about innovation in networks of three U.S. organizations. Journal of Communication, 34, 78-91.
American Psychological Association. (1992). Ethical principles of psychologists and code of conduct. American Psychologist, 47, 1597-1611.
Armenakis, A. A., Harris, S. G., & Moseholder, K. W. (1993). Creating readiness for organizational change. Human Relations, 46, 681-703.
Ashmos, D. P., McDaniel, R. R., & Duchon, D. (1990). Differences in perception of strategic decision-making processes: The case of physicians and administrators. Journal of Applied Behavioral Science, 26, 201-218.
Brimm, I. M. (1988). Risky business: Why sponsoring innovations may be hazardous to career health. Organizational Dynamics, 16, 28-41.
Burt, R. S. (1982). Toward a structural theory of action: Network models of social structure, perception, and action. New York: Academic Press.
Burt, R. S. (1991). Structure reference manual Version 4.2. New York: Columbia University Center for the Social Sciences.
Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press.
Burt, R. S. (1997). The contingent value of social capital. Administrative Science Quarterly, 42, 339-365.
Damanpour, F. (1991). Organizational innovation: A meta-analysis of effects of determinants and moderators. Academy of Management Journal, 34, 555-590.
Dearing, J. (1993). Rethinking technology transfer. International Journal of Technology Management, 8, 478-485.
Dearing, J. W., & Meyer, G. (1994). An exploratory tool for predicting adoption decisions. Science Communication, 16, 43-57.
Deetz, S. (1992). Democracy in an age of corporate colonization: Developments in communication and the politics of everyday life. Albany, NY: SUNY Press.
Devellis, R. (1991). Scale development: Theory and applications. Newbury Park, CA: Sage.
Dillman, D. A. (1978). Mail and telephone surveys: The total design method. New York: John Wiley.
Dillman, D. A. (1991). The design and administration of mail surveys. Annual Review of Sociology. 17, 225-249.
Farace, R. V., & Johnson, J. D. (1974). Comparative analysis of human communication networks in selected formal organizations. Paper presented at the International Communication Association Conference, New Orleans, LA, May, 1974.
Fidler, L. A., & Johnson, J. D. (1984). Communication and innovation implementation. Academy of Management Review, 9, 704-711.
Fink, E., & Mange, P. (1985). An exploration of confirmatory factor analysis. In B. Dervin & M. Voight (Eds.), Progress in communication science (Vol. 6, pp. 167-197). Norwood, NJ: Ablex.
Foegen, J. H.(1988). A devil's advocate approach to technology. Business Horizons, 36, 43-46.
Frey, L, Botan, C., & Kreps, G. (2000). Investigating communication: An introduction to research methods (2nd ed.). Boston: Allyn and Bacon.
Galbraith, J. R. (1984). Designing the innovating organization. In D. Kolb, I. Rubin, and J. McIntyre, (Eds.), Organizational psychology: Readings on human behavior in organizations. New Jersey: Prentice-Hall.
Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78, 1360-1380.
Granovetter, M. (1982). A network theory revisited. In P. V. Marsden & N. Lin (Eds.), Social structure in network analysis (pp. 105-130). Beverly Hills, CA: Sage.
Howell, J. M., & Higgins, C. A. (1990). Champions of technological innovations. Administrative Science Quarterly, 35, 317-341.
Hunter, J. E. (1980). Factor analysis. In P. R. Monge & J. N. Cappella (Eds.), Multivariate techniques in human communication research (pp. 229-258). New York: Academic Press.
Hunter, J., & Gerbing, D. (1982). Unidimensional measurement, second order factor analysis, and causal models. Research in Organizational Behavior, 4, 267 320.
Hunter, J. E., & Lim, T. S. (1987). LIMSTAT. Unpublished manuscript. Michigan State University, East Lansing, MI.
Jablin, F. M. (1987). Formal organization structure. In F. M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organizational communication: An interdisciplinary perspective (pp. 711-746). Newbury Park, CA: Sage.
Janis, I. L. (1983). Groupthink (2nd ed.). Boston: Houghton Mifflin.
Johnson, J. D. (1993). Organizational communication structure. Norwood, NJ: Ablex.
Johnson, J. D., & Chang, H.-J. (2000). Internal and external communication, boundary spanning, and innovation adoption: An over-time comparison of three explanations of internal and external innovation communication in a new organizational form. The Journal of Business Communication 37, 238-263.
Kanter, R. M. (1983). The change masters: Innovation and entrepreneurship in the American corporation. New York: Simon & Schuster.
Kanter, R.(1988). Three tiers for innovation research. Communication Research, 15, 509-523.
Kanter, R. M., Kao, J., & Wiersema, F. (1997). Innovation: Breakthrough thinking at 3M, DuPont, GE, Pfizer, and Rubbermaid. New York: HarperCollins.
Kilduff, M., & Krackhardt, D. (1994). Bringing the individual back in: A structural analysis of the internal market for reputation in organizations. Academy of Management Journal, 37, 87-108.
King, N. (1990). Innovation at work: The research literature. In M. A. West and J. L. Farr (Eds.), Innovation and creativity at work: Psychological and organizational strategies. West Sussex, England: John Wiley.
Kossek, E. E. (1989). The acceptance of human resource innovation by multiple constituencies. Personnel Psychology, 42, 263-281.
LaFrance, B. H., Johnson, J. D., Ethington, C. & Meyer, M. (1997, May). Studying the Cancer Information Service overtime. Paper presented at the annual meeting of the International Communication Association, Montreal, Canada.
Lewis, L. K., & Seibold, D. R. (1996). Communication during intraorganizational innovation adoption: Predicting users' behavioral coping responses to innovation in organizations. Communication Monographs, 63, 131-157.
Marcus, A.C. (1998). The Cancer Information Service Research Consortium. Preventive Medicine, 27, S1-S2.
Marcus, A.C., Morra, M. E., Bettinghaus, E., Crane, L.A., Cutter, G., Davis, S., Rimer, B. K., Thomsen, C, & Warnecke, R. B.(1998). The Cancer Information Service Research Consortium: An emerging laboratory for cancer control research. Preventive Medicine, 27, S3-S15.
Marcus, A. C., Woodworth, M. A., Strickland, C. J. (1993). The Cancer Information Service as a laboratory for research: The first 15 years. Journal of the National Cancer Institute Monographs, 14, 119.130.
Meyer, M. (1996). The effects of weak ties on perceived organizational innovativeness and innovation characteristics. Unpublished doctoral dissertation, Michigan State University, East Lansing, MI.
Meyer, M., Johnson, J. D., & Ethington, C. (1997). Contrasting attributes of preventive health innovations. Journal of Communication, 47, 112-131.
Mitchell, R.K., Agle, B.R., & Wood, D. J. (1997). Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts. Academy of Management Review, 22, 853-886.
Monge, P., Cozzens, M., & Contractor, N. (1992). Communication and motivational predictors of the dynamics of organizational innovation. Organizational Science, 3, 250-274.
Morra, M., Van Nevel, J. P., Nealon, B., Mazan, K., & Thomsen, C. (1993). History of the Cancer Information Service. Journal of the National Cancer Institute Monographs, 14, 7-34.
Nohria, N., & Gulati, R. (1996). Is slack good or bad for innovation? Academy of Management Journal, 39, 1245-1264.
Pavitt, C., & Curtis, E. (1994). Small group discussion: A theoretical approach (2nd ed.). Scottsdale, AZ: Gorsuch Scarisbrick.
Rakowski, W., Dube, C. E., Marcus, B. H., Prochaska, J. O., Velicer, W. F., & Abrams, D. B. (1992). Assessing elements of women's decisions about mammography. Health Psychology, 11, 111-118.
Roberts, N., & Bradley, R. (1991). Stakeholder collaboration and innovation: A study of public policy initiation at the state level. Journal of Applied Behavioral Science, 27, 209-227.
Rogers, E. (1995). The diffusion of innovations (4th ed.). New York: Free Press.
Stacks, D. W., & Hocking, J. E. (1999). Communication Research (2nd ed.). New York: Longman.
Stjernberg, T., & Philips, A. (1993). Organizational innovations in a long-term perspective: Legitimacy and souls-of-fire as critical factors of change and viability. Human Relations, 46, 1193-1219.
Tichy, N. M. (1974). Agents of planned social change: Congruence of values, cognitions, and actions. Administrative Science Quarterly, 19, 164-182.
Van de Ven, A. (1986). Central problems in the management of innovation. Management Science, 32, 590-607.
Weenig, M. W. H. (1999). Communication networks in the diffusion of an innovation in an organization. Journal of Applied Social Psychology, 29, 1072-1092.
Weick, K. E. (1995). Sensemaking in organizations. Newbury Park, CA: Sage.
Weimann, G. (1983). The strength of weak conversational ties in the flow of information and influence. Social Networks, 5, 245-267.
Weiss, C. H. (1983). Toward the future of stakeholder approaches in evaluation. In A. S. Bryk (Ed.) Stakeholder-based evaluation (pp. 83-96). San Francisco: Jossey-Bass.
Woodruff, M. J. (1991). Understanding--and combatting--groupthink. Supervisory Management, 36, 8.
|Printer friendly Cite/link Email Feedback|
|Publication:||The Journal of Business Communication|
|Date:||Oct 1, 2000|
|Previous Article:||Innovation, Corporate Strategy, and Cultural Context: What Is the Mission for International Business Communication?|
|Next Article:||Effects of Communication Direction on Job Performance and Satisfaction: A Moderated Regression Analysis.|