Journal of Organizational Behavior, J. Organiz. Behav. 32, 1062–1083 (2011) Published online 7 September 2010 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/job.724

Diversity and organizational innovation: The role of employee involvement
YANG YANG1* AND ALISON M. KONRAD2
Management Department, the Wharton School, University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A. Department of Organizational Behavior, Richard Ivey School of Business, U. of Western Ontario, London, Ontario, Canada
2 1

Summary

This study examined the interactive effects of workplace diversity and employee involvement on organizational innovation. Using a sample of 182 large Canadian organizations, we found a three-way interaction between level of employee involvement, variation in involvement, and racioethnic diversity on innovation. In organizations with high levels of employee involvement, high variation in involvement was associated with higher involvement levels among racioethnic minorities, resulting in a stronger association between diversity and innovation. Furthermore, the association between White employee involvement and innovation was significantly more positive under the condition of high involvement among racioethnic minority group members. Thus, ensuring high levels of involvement among members of historically marginalized racioethnic groups enhances the innovation effects of employee empowerment systems. Copyright # 2010 John Wiley & Sons, Ltd. Keywords: gender and racioethnic diversity, diversity management, employee involvement level, employee involvement variation, organizational innovation

Introduction
The two research streams of employee involvement and diversity management have a number of heretofore unacknowledged commonalities. Both areas seek to improve the impact of human capital on business outcomes by developing and utilizing employee skills and abilities more fully. Employee involvement practices are intended to inject the information and knowledge of non-management employees into higher-level organizational decisionmaking processes (Benson, Young, & Lawler, 2006; Guthrie, 2001; Lawler, 1986). Diversity management seeks to leverage the different information and knowledge bases brought into the organization by members of historically marginalized demographic and identity groups (Cox, 2001; Cox & Blake, 1991; Robinson & Dechant, 1997). The goal of both research streams can be viewed as twofold: Enhancing workplace outcomes for historically disempowered employees, while at the same time creating value for the organization. Employee involvement and diversity management differ in important respects. Employee involvement research has failed to recognize the impact of demographic and identity group memberships on employee outcomes, while diversity management research emphasizes that certain categories of individuals are less likely to be developed and utilized than others. The area of diversity management has failed to develop a coherent set of recommended management practices (Kalev, Dobbin, & Kelly, 2006), and indeed, some of the practices, such as affirmative action, have been controversial (Heilman, Block, & Stathatos, 1997). By comparison, a well-developed set of employee involvement practices has been proposed and studied (Benson et al., 2006; Delery & Shaw, 2001; Pfeffer, 1998).
* Correspondence to: Yang Yang, The Management Department of the Wharton School, University of Pennsylvania, 2000 Steinberg-Dietrich Hall, Philadelphia, Pennsylvania, 19104, USA. E-mail: yang14@wharton.upenn.edu

Copyright # 2010 John Wiley & Sons, Ltd.

Received 18 December 2008 Revised 20 July 2010, Accepted 21 July 2010

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1063

Specifically, employee involvement practices for enhancing employee knowledge, skills, and abilities (KSAs), for empowering employees to make decisions and for motivating employees to perform have been linked to positive organizational outcomes (Combs, Liu, Hall, & Ketchen, 2006; Yalabik, Chen, Lawler, & Kim, 2008). Synthesizing these two streams of thought has the potential to contribute to extant knowledge in both areas. On the one hand, the diversity management perspective helps to predict possible implementation difficulties when employee involvement programs do not take into account cultural and power differences between identity groups. Social closure theory argues that traditionally predominant groups are likely to exclude historically marginalized groups from accessing important resources in order to maintain their power (Tomaskovic-Devey, 1993). Employee involvement programs, which aim at pushing power, information, and knowledge down to the lower job levels (Lawler, 1986), may be viewed as constituting important resources. Thus powerful demographic groups may exclude less powerful groups from participating in involvement processes. Instead, in an environment high in autonomy and without strong diversity management, demographic groups might end up in stereotypical roles (Harrison & Humphrey, 2010), limiting participative opportunities for marginalized groups. As a result, involvement practices may not lead to desired outcomes in diverse workplaces. On the other hand, the employee involvement perspective has the potential to provide the diversity management field with a coherent set of organizational practices for fostering intergroup contact and improving intergroup relations. Specifically, Allport’s contact hypothesis (1954) argues that four features of a contact situation can effectively reduce intergroup prejudice, namely, equal status between the groups in the contact situation; common goals; intergroup cooperation; and institutional support (e.g., authorities, laws, or custom). A recent meta-analysis of research in this field suggests that these conditions are not essential and mere contact can reduce intergroup prejudice (Pettigrew & Tropp, 2006). Employee involvement programs, which emphasize communication and teamwork, may provide opportunities for intergroup contact and lead to favorable outcomes in a diverse workplace. In sum, combining the two perspectives identifies involvement programs as potential tools for developing and leveraging historically underutilized human capital in order to realize the potential of a diverse workforce. Care must be taken in the implementation of involvement programs in order to ensure that all demographic groups have equal participation. Otherwise, involvement programs will be unable to address intergroup power differences as is needed for effective diversity management. This study examines the combined impact of racioethnic diversity, proportion of women, and employee involvement on organizational innovation. Innovation is an often-invoked benefit of employing more members of historically marginalized groups (Cox & Blake, 1991; Robinson & Dechant, 1997) which has not yet received much research attention. Both conceptual and empirical support exists to justify the link between organizational demographic composition and innovation. Conceptual perspectives on creativity emphasize the value of a diversity of perspectives and knowledge bases for generating a wider variety of ideas (Cummings, 2004; Rodan, 2002). Empirically, meta-analysis at the group level has found that gender composition and diversity in attitude, ability, and personality are positively associated with group performance on complex tasks but negatively associated with performance on simple tasks (Bowers, Pharmer, & Salas, 2000). This finding implies that the variety of perspectives present in a diverse group is beneficial when complex problem-solving requiring creativity is needed. At the establishment level, research has also shown that that greater racioethnic diversity (Richard, 2000) and greater participation of women (Francoeur, Labelle, & Sinclair-Desgagne, 2007) lead to better outcomes when innovation and complex problem-solving are required. Previous research has assumed that the favorable outcomes result from the increased presence of historically marginalized groups and are driven by the participation of employees with new perspectives in organizational decision-making, which enhances the variety of viewpoints considered (Harrison & Klein, 2007). Increasing the employment of historically marginalized groups does not automatically increase their opportunities for voice, however. Historically predominant groups often maintain or even gain more power when the proportion of marginalized group members increases (Maume, 1999). Hence, a key research question is whether members of historically predominant and marginalized groups have an equal voice in the workplace. We address this question by assessing whether members of historically predominant and marginalized groups participate equally in employee
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1064

Y. YANG AND A. M. KONRAD

empowerment programs. Hence, beyond examining the interactive effect of employee involvement with racioethnic diversity and proportion of women on organizational innovation, the current study also examines the level of involvement reported by members of historically predominant and marginalized groups as well as the outcome when all groups report a high level of participation. As an outcome variable, innovation is a relatively proximal outcome of effectively-managed workplaces compared to firm financial performance or other measures of overall organizational effectiveness. Innovation does not necessarily result in improved performance, for instance, if an industry downturn occurs or management sets the wrong strategic direction. Employee involvement programs are intended to generate new ideas to improve organizational products and processes (Arthur & Aiman-Smith, 2001; Benson et al., 2006; Combs et al., 2006; Guthrie, 2001), which may eventually have positive impacts on firm financial performance depending upon strategy, industry environment, and other exogenous factors. Hence, at the establishment level, innovation is likely to be a more proximal outcome than performance for studying the joint effects of demographic composition and employee involvement.

Theoretical Overview
Theoretically, we posit that beyond any potential main effects, employee involvement practices are a potential moderator of the relationship between the employment of historically marginalized groups and organizational innovation. As such, our arguments are consistent with previous authors’ conclusions that the impact of workplace diversity is moderated by contextual factors (Joshi & Roh, 2009; Kochan et al., 2003; van Knippenberg & Schippers, 2007). In the following sections, we link demographic composition and employee empowerment to two key creative processes in organizations, specifically, the generative process and the elaboration process. Then, we use the categorization–elaboration model of diversity dynamics (van Knippenberg, De Dreu, & Homan, 2004) to provide a theoretical foundation for examining employee involvement programs as a potential moderator of the link between demographic composition and organizational innovation. While the theoretical basis of our arguments was originally conceptualized at the group level of analysis, we apply them to innovation at the organizational level. Cross-level models of organization learning and innovation are widely accepted which identify group-level effects as crucial factors for creating organizational-level outcomes (Crossan, Lane, & White, 1999). Furthermore, prior research on diversity management has shown that the same mechanism (in that case, diversity climate) had positive effects at both the individual and organizational levels (Gonzalez & DeNisi, 2009).

Demographic composition and the generative process
Organizational innovation requires the generation of novel ideas that address the current organizational situation. Novel ideas result from ‘‘the process of merging thought categories, or mental images, either across or within domains, in ways that have not been done before, in order to develop an original and appropriate solution to a situation or problem’’ (Kilgour, 2006, p. 82). Hence, one foundation of creativity is the ‘‘generative process’’ (Leung, Maddux, Galinsky, & Chiu, 2008) that results in a wider set of options for consideration. As such, access to new sources of information, knowledge, and perspectives enhances the potential for organizational innovation by increasing the number of thought categories and mental images available for modification and recombination. At the organizational level, research has shown that organizations with access to a wider variety of knowledge resources are more effective innovators (Rodan & Galunic, 2004) and are better able to develop and assimilate new processes into their ongoing operations (Fichman & Kemerer, 1997).Although some researchers have discounted the possibility
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1065

that demographic composition can affect task outcomes in organizations (Joshi & Roh, 2009), others have emphasized that differences in values, attitudes, experiences, perspectives, and cultures between gender and racial or racioethnic groups have the potential to increase the number of thought categories and mental images available to the organization (Earley, 1989; Feingold, 1994; Frink et al., 2003; Kashima, Hamaguchi, Kim, Choi, Gelfand, & Yuki, 1995; Konrad, Ritchie, Lieb, & Corrigall, 2000; Richard, 2000; Stedham & Yamamura, 2004). At the individual level, a body of research shows that exposure to information from multiple cultures leads to more creative performance (Leung et al., 2008). At the group level, a study of 50 teams in a Fortune 500 company found that racial diversity was positively associated with the number of innovative ideas generated (Cady & Valentine, 1999).

Employee involvement and the elaboration process
Creative thinking results when people generate a large variety of ideas with potential for use and then ‘‘scrutinize these candidate ideas to determine which ones should receive further processing such as modification, elaboration, and transformation’’ (Leung et al., 2008, p. 171). Hence, a second key foundation of creativity in organizations is the ‘‘elaboration process,’’ defined as, ‘‘the exchange of information and perspectives, individual-level processing of the information and perspectives, the process of feeding back the results of this individual-level processing into the group, and discussion and integration of its implications’’ (van Knippenberg et al., 2004, p. 1011). The elaboration process is critical to organizational innovation and learning. Organizational learning results from links across the individual, group, and organizational levels of analysis through the processes of ‘‘intuiting, interpreting, integrating, and institutionalizing’’ (Crossan et al., 1999). ‘‘Feed-forward’’ learning, i.e., a flow from individuals to organizations, is the process whereby individual creativity is transformed into organizational innovation through the elaboration process (Crossan et al., 1999). ‘‘Feed-back’’ learning, i.e., a flow from organizations to individuals, is concerned with how contextual factors either inhibit or facilitate individual creativity (Crossan et al., 1999). Cross-level research shows that the context within which the individual works can enhance or hinder employee creativity (Hirst, van Knippenberg, & Zhou, 2009). Barriers to individual involvement in the elaboration process results in under-utilization of employee knowledge and ideas for organizational innovation, and employee empowerment programs are intended to remove barriers to employee participation in the organizational learning process. In particular, the knowledge and information held by employees on the lower rungs of the organizational hierarchy is an under-utilized source of novel ideas that have the potential to add value to the organization (Arthur & Aiman-Smith, 2001; Batt, 2000). A considerable body of research has shown that employee involvement programs can indeed result in improvements to organizational products and processes (Appelbaum, Bailey, Berg, & Kalleberg, 2000; Arthur & Aiman-Smith, 2001; Hunter, MacDuffie, & Doucet, 2002; Kochan & Rubinstein, 2000). Employee involvement practices have many positive outcomes for employees and organizations alike because these practices empower workers. More specifically, formalized practices that build workers’ skills, provide them with autonomy and authority to make decisions, and reward them for contributing value to the organization lead to a state of psychological empowerment (Laschinger, Finegan, Shamian, & Wilk, 2001; Spreitzer, 1996). Psychological empowerment, or the experience of meaning, competence, self-determination, and impact at work, has been linked to positive employee attitudes (Riordan, Vandenberg, & Richardson, 2005), retention (Spreitzer & Mishra, 2002; Vandenberg, Richardson, & Eastman, 1999), innovative behaviors (Spreitzer, 1995), and even more peaceful societies (Spreitzer, 2007). The empowerment experienced by workers in organizations utilizing high involvement practices results in innovation because it engages employees in the idea generation and elaboration process to contribute to organizational learning. With these practices, employees participate in more decisions, share more ideas, exchange and process more information, and integrate more of their thinking into their work processes. Employee participation in these involvement behaviors increases the likelihood that a workforce composed of a larger percentage of
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1066

Y. YANG AND A. M. KONRAD

historically marginalized groups will generate a variety of ideas and engage in the elaboration process to select, develop, and implement those ideas.

The categorization–elaboration model of diversity
Diverse groups engaged in organizational decision-making and problem-solving have the potential for greater creativity than homogeneous groups, due to the richness of the resources at their disposal during the generative phase of the creativity process. As a result, demographic composition is linked to the potential for organizational innovation. Diversity has also been linked to destructive conflict and miscommunication (Jehn, Northcraft, & Neale, 1999; Pelled, Eisenhardt, & Xin, 1999), however, which hamper collaboration during the elaboration phase of the creativity process (De Dreu & Weingart, 2003; Edmondson, 1999). The potentially negative outcomes of diversity are attributable to what van Knippenberg et al. (2004) call social categorization effects, defined as, ‘‘the differentiation between in-group others, who are subjectively similar to self, and outgroup others, who are subjectively dissimilar to self’’ (van Knippenberg et al., 2004, p. 1010). Social categorization results in negative outcomes for diverse groups due to the operation of intergroup biases (Brewer & Brown, 1998). Intergroup biases disrupt working relationships by reducing listening, information exchange, and trust (Brewer, 1979; Jehn et al., 1999; van Knippenberg, 1999; van Knippenberg & Wilke, 1992). van Knippenberg et al. (2004) explain the relationship between diversity, social categorization, and the development and generation of new ideas with their categorization–elaboration model (CEM). The CEM proposes that the impact of diversity on group outcomes such as creativity, innovation, and decision quality depends upon the extent to which categorization processes result in intergroup biases. While diverse groups contain a larger variety of information, knowledge, and perspectives, the ability of the group to draw upon these greater resources is inhibited if social categorization processes become triggered by cognitive accessibility of categorization, normative fit of categorization, and/or identity threat (van Knippenberg et al., 2004, Figure 1). In organizations, the CEM implies that increasing the participation of historically marginalized groups will only lead to positive innovation outcomes when organizations minimize the impact of intergroup biases through effective management. Employee involvement practices may help organizations avoid or overcome intergroup biases as employee demographic composition changes. Practices that build employee KSAs strengthen the confidence of all groups that they can contribute positively to the organization, which reduces problems of low collective self-esteem and identity threat (Crocker & Luhtanen, 1990; Crocker, Luhtanen, Broadnax, & Blaine, 1999). Practices that involve employees in teamwork and decision-making create more opportunities for intergroup contact across the organization. Contact is an essential condition that reduces intergroup prejudice and enhances the likelihood of building positive working relationships across demographic and identity group differences (Pettigrew, 1998). Practices that motivate employees help to engage members of historically marginalized group by creating a more positive organizational context for these individuals (Cox, 2001). Because employee empowerment programs reduce identity threat, enhance inter-group contact, and motivate members of all identity groups, they have the potential to minimize categorization processes and maximize the generative and elaboration processes in workplaces undergoing demographic changes. Hence, employee empowerment programs may be quite effective for maximizing organizational innovation as the presence of historically marginalized groups increases.

Research Hypotheses
Research suggests that the mere institution of employee involvement practices by management is insufficient for generating the desired outcomes of increased motivation, innovation, and performance (Riordan et al., 2005; Shadur, Kienzle, & Rodwell, 1999). Rather, the effectiveness of involvement practices depends on the breadth of the program
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1067

and the extent to which employees across the entire organization actually engage in participative behaviors (Ichniowski, Kochan, Levine, Olson, & Strauss, 1996; Lepak, Taylor, Tekleab, Marrone, & Cohen, 2007; Nishii, Lepak, & Schneider, 2008; Takeuchi, Chen, & Lepak, 2009; Vandenberg et al., 1999). In particular, the percentage of workers in a firm engaged in participatory behaviors has been linked to positive outcomes in several studies (Benson et al., 2006; Datta, Guthrie, & Wright, 2005; Guest, Conway, & Dewe, 2004; Guest, Michie, Conway, & Sheehan, 2003; Guthrie, 2001). These findings support Lawler’s (1986) conclusion that employee involvement programs that extend across the organization are more effective than more limited programs. A high involvement culture is only indicated when involvement behaviors are manifested by employees across the organization (Den Hartog & Verburg, 2004; Guest et al., 2003). We posit that the level of actual employee involvement and the amount of variation in involvement between traditionally predominant and historically marginalized groups are potentially important moderators of the association between organizational demographic composition and innovation outcomes. A high level of involvement is needed for employees to generate and elaborate a variety of innovative ideas. Variation in involvement, or unequal involvement between traditionally predominant and historically marginalized groups, reduces the positive impact of a highly-involved workforce.

Two-way interaction: Demographic composition and employee involvement
Employee involvement practices, which engage employees in relatively high-level organizational decision-making processes (Benson et al., 2006; Delery & Shaw, 2001; Pfeffer, 1998), have the potential to leverage the innovation potential of a workforce composed of a higher percentage of women and greater racioethnic diversity. Employee involvement practices that enhance employee KSAs, empower employees to make decisions, and motivate employees to perform are likely to enhance innovation outcomes. Involvement practices that enhance employee KSAs, such as training and job rotation, increase the quality of the organization’s human capital (Pfeffer, 1998). Members of historically marginalized groups are particularly likely to benefit from practices that build their KSAs because they are most likely to have experienced barriers to their personal development. For instance, women have had relatively little access to training in a number of blue¨ collar occupations (Fransson & Thornqvist, 2006; Harlan & Steinberg, 1989) and people of color tend to have less access to high quality education (Darling-Hammond, 2006). Access to training, therefore, is likely to enhance the abilities of a diverse workforce to contribute value-added ideas to the organizational decision-making process. Building employee KSAs is necessary because openness to employee input without building employee KSAs to the level where they can contribute high quality ideas is unlikely to result in valuable innovations (Lawler, 1986). Involvement practices that increase employee participation in important decision-making processes include suggestion programs, information-sharing, job design, and self-managed teams (Delery & Shaw, 2001; Pfeffer, 1998). Members of historically marginalized groups may particularly benefit from such involvement practices because they are most likely to be segregated into low-level routine jobs that provide little discretion or autonomy. Women and racial minorities are significantly more likely to be found at the bottom of wage distributions (Pendakur, Pendakur, & Woodcock, 2008) and significantly less likely than white men to be promoted from the lowest-level organizational positions, a phenomenon called, ‘‘the sticky floor’’ (Yap & Konrad, 2009). Hence, accessing the ideas of historically marginalized groups requires the organization to push decision-making opportunities downward to empower employees at the bottom of the wage distribution. Moreover, according to intergroup contact theory, involvement practices may increase opportunities for intergroup contact and in turn reduce intergroup prejudice (Pettigrew, 1998; Pettigrew & Tropp, 2006). Under this condition, the generative and elaboration processes proposed by CEM are more likely to be facilitated. Involvement practices must also motivate employees to contribute ideas in a way that adds value to the organization. Engaging in participative behaviors relatively frequently enhances employee motivation through the fulfillment of individual needs (Deci & Ryan, 2000). Teamwork meets the need for affiliation; involvement in work
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1068

Y. YANG AND A. M. KONRAD

design meets needs for flexibility and self-development; and involvement in communication meets the need for intrinsic job qualities such as opportunities to have voice and influence (Benson et al., 2006; Vandenberg et al., 1999). Teamwork provides people with opportunities to work together closely, enhances chances of face-to-face interaction, increasing communication and social integration (Evans & Davis, 2005). In these ways, teamwork fulfills social needs and increases information-sharing across the organization. In sum, employee involvement activities are likely to be highly motivating in ways that enhance the generation and elaboration of a large variety of new ideas. In summary, employee involvement practices have the potential to improve the KSAs, participation in decisionmaking, motivation, and quality of working life for workers at the bottom of the organizational hierarchy. Because members of historically marginalized groups are over-represented in these low-level positions, employee involvement practices have the potential to help organizations better utilize the variety of information, knowledge, and perspectives of these groups. Organizations that create a relatively high level of actual employee participation in involvement behaviors access the variety of ideas present across the organization more effectively, resulting in a higher level of innovation. Hypothesis 1: Employee participation in involvement behaviors positively moderates the association between the representation of women and racioethnic minorities and organizational innovation.

Three-way interaction: Variation in employee involvement
Employee involvement practices are likely to reduce barriers to the novel contributions of members of historically marginalized groups; however, as previous research has shown, management implementation of practices is often inconsistent (Lawler, 1986; Riordan et al., 2005; Shadur et al., 1999). Furthermore, a high level of involvement does not necessarily imply that traditionally predominant and historically marginalized groups participate equally in involvement processes. Tomaskovic-Devey’s (1993) social closure theory suggests that powerful social groups hoard resources within the in-group’s social network to maintain their higher status and material outcomes. His research has documented social closure effects among white employees and among men, who fail to share resources such as job opportunities, authority, and training with people of color and women, respectively (Tomaskovic-Devey & Skaggs, 2002; Tomaskovic-Devey, Thomas, & Johnson, 2005). In the terminology of social closure theory, employee involvement opportunities constitute a valuable organizational resource that provide visibility, impact, and the psychological benefits of empowerment (Spreitzer, 1995, 1996). As such, the identity groups who have the ability to claim involvement opportunities for themselves will do so, and will tend to deny them to out-group members to the extent they can. The involvement of historically marginalized groups has important implications for organizational innovation. Successful innovation requires the development and implementation of ideas that are novel for the organization (Kilgour, 2006). Novel ideas are most likely to come from individuals whose beliefs, assumptions, and perspectives differ from the organization’s historical norm. The quest for novelty, therefore, implies that organizational innovation can be increased by heeding the voices of members of historically marginalized groups (Cox, 2001). Therefore, given the foundation of a relatively high level of employee involvement, the low variation in involvement derived from the active participation of members of historically marginalized groups will have the potential to enhance organizational innovation even further. Achieving the goal of a relatively high level of voice for members of historically marginalized groups is not easy. In addition to social closure theory, the CEM (van Knippenberg et al., 2004) and status characteristics theory (Ridgeway, 1991) emphasize that there are many barriers to overcome. The CEM (van Knippenberg et al., 2004) supports the value of focusing on new ideas through the elaboration process and cautions that many organizations composed of multiple demographic groups will fail to garner these potential benefits due to the operation of intergroup biases. Status characteristics theory (Ridgeway, 1991) emphasizes that members of historically marginalized groups are the individuals least likely to be heard and supported in task groups due to negative
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1069

expectations for their ability to add value. Together, these perspectives highlight that members of historically marginalized groups experience barriers when they attempt to contribute new ideas to support organizational innovation. Hence, it is likely that many organizations with employee involvement programs do not always fully leverage the potential benefits to innovation. We anticipate that those organizations that successfully promote the voices of historically marginalized groups will show the highest levels of innovation. This set of contingencies means that the most organizational innovation will be associated with a combination of a relatively high level and low variation in employee involvement. A high level of involvement indicates that the organization is attempting to leverage employee ideas, and the low variation in involvement indicates that members of historically marginalized groups and traditionally predominant groups exercise an equal level of voice. Hypothesis 2: The positive moderating effect of employee involvement level on the association between the representation of women and racioethnic minorities and organizational innovation will be stronger when variation in involvement behaviors is low.

Dominant and marginalized group involvement
Employee involvement for both traditionally dominant groups and historically marginalized groups will be important to organizational innovation. As argued above, historically marginalized groups need to be involved because they are most likely to bring perspectives, knowledge, and ideas that are new to the organization. Hence, high involvement of traditionally dominant groups without involving members of historically marginalized groups will lead to less innovation due to reduced creativity for idea generation. High involvement of historically marginalized groups without involving members of traditionally dominant groups is also likely to limit organizational innovation. First, members of traditionally dominant groups are likely to have the longest tenures and deepest levels of organization-specific knowledge. Hence, their inclusion in the process of elaborating new ideas (van Knippenberg et al., 2004) is likely to maximize fit to the organizational situation and benefits to performance outcomes. Second, due to status distinctions (Ridgeway, 1991) and social closure dynamics (Tomaskovic-Devey, 1993), innovations developed either wholly or primarily by members of historically marginalized groups are likely to become derailed at the implementation phase. Traditionally dominant groups generally hold critical positions in organizations and are capable of disrupting the implementation of changes they do not support (Fronda & Moriceau, 2008; Harrison & Humphrey, 2010). Participation in the development of the innovation reduces resistance to change by enhancing awareness of the need for the change, increasing understanding of the nature of the change, and allowing input into the design of the change so that it will lead to desirable outcomes for the implementers (Lawler, 1986). Hence, the inclusion of members of traditionally dominant groups in involvement programs is critical to innovation due to the importance of gaining their support during the implementation phase of the innovation process. Hypothesis 3: Involvement behaviors by historically marginalized groups moderate the association between involvement behaviors by traditionally dominant groups and innovation such that the slope becomes significantly more positive (i.e. upward-slanting) as involvement by historically marginalized groups increases.

Research Methodology
We used the Workplace and Employee Survey (WES) data collected by Statistics Canada in 2001 and 2002 to test the above hypotheses. Based on industry, region, and size, Statistics Canada uses a stratified sampling method to draw
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1070

Y. YANG AND A. M. KONRAD

the workplace sample from the Business Registration List containing all businesses operated in Canada. All surveyed companies were given a sampling weight to facilitate estimation of population characteristics. The sampling weights reflect the probability that any particular data point was included in the WES sample. For instance, 568 retail workplaces and 265 forestry, mining, oil, and gas extraction workplaces were included in the 2001WES workplace sample; however, the 568 retail workplaces represent a total of 229 893 retailers in Canada, while the 265 forestry, mining, oil, and gas extraction workplaces represent a total of only 10 362 Canadian workplaces. Researchers are required by Statistics Canada to weight the WES data in analysis to reflect the differences in the probability that each type of workplace was selected into the sample. The reason for this requirement is to ensure that the disseminated findings accurately reflect the population of Canadian workplaces. The major respondent to the workplace survey was an HR manager, while in small companies, the major respondents were the general managers or the business owners (Zatzick & Iverson, 2006). Statistics Canada used computer assisted telephone interviews to collect workplace data. For large organizations, Statistics Canada required more than one contact person to respond to the workplace survey. For example, accounting personnel were asked to provide financial information (Zatzick & Iverson, 2006). The employee sample was drawn based on the lists provided by each responding employer. Twenty-four employees were randomly selected from each company, and all employees were surveyed if the companies had less than four employees. Both the employers and employees who had responded to the 2001 surveys were also surveyed in 2002, allowing for the creation of a longitudinal database. In this study, we measured organizational innovation using the 2002 workplace survey and assessed diversity and involvement using aggregations from the 2001 employee survey. Drawing information from different surveys for the dependent and independent variables reduces common method bias. Also, creating a one-year lag between the independent and the dependent variables facilitates testing causal relationships. For the workplace survey, response rates were 85.9 and 84 per cent in 2001 and 2002, respectively. For the employee survey, response rates were 86.9 and 90.9 per cent in 2001 and 2002, respectively. The datasets consisted of 6223 employers and 20 377 employees in 2001, and 5818 employers and 16 813 employees in 2002. We only retained organizations that responded to both 2001 and 2002 workplace surveys so that we could test the one-year lagged effects of the hypothesized predictors on innovation. We focused on organizations with more than 500 employees because size can influence organizational innovation (Damanpour, 1996). Among these large organizations, we selected those organizations with at least five respondents to the employee surveys because aggregation biases diminish when five or more data points are used to assess the characteristics of collectivities (Bliese & Halverson, 1998; Liao, Joshi, & Chuang, 2004). We also left supervisors out of this calculation because involvement opportunities are likely to be different for supervisors and non-supervisors. The final sample used to test the hypotheses consisted of 183 organizations and 1289 employees. The average number of employees in the sampled establishments was 1108. Means, standard deviations, and correlations for major study variables are shown in Table 1.
Table 1. Means, standard deviations, and correlations Mean 1. Innovation strategy 2. People strategy 3. Unionization 4. % Professionals 5. Racioethnic diversity (Blau index) 6. % Women 7. Squared % women 8. Involvement level 9. Involvement variation 10. Organization innovation 3.02 3.29 0.88 0.30 0.18 0.58 0.40 0.34 0.80 0.48 SD 1 2 3 4 5 6 7 8 9 10

0.95 1.00 0.74 0.38 1.00 0.32 0.09 À0.02 1.00 0.22 À0.01 À0.18 0.23 1.00 0.20 À0.28 0.06 À0.33 À0.16 1.00 0.26 0.02 0.02 0.16 0.56 À0.06 1.00 0.26 À0.02 0.04 0.17 0.53 À0.04 0.97 1.00 0.40 0.17 0.24 À0.16 0.04 0.18 0.02 0.01 1.00 0.22 0.00 0.08 À0.03 0.09 0.14 À0.05 À0.08 0.32 1.00 0.35 0.29 0.21 0.03 À0.14 À0.02 0.09 0.07 0.14 0.00 1.00

Note. N ¼ 182. Correlations > 0.16 (0.21) in magnitude are significant at the 0.05 (0.01) level (2-tailed).

Copyright # 2010 John Wiley & Sons, Ltd.

J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1071

Measures
Demographic composition Because the workplace survey did not provide information concerning overall gender and racioethnic composition, aggregation of individual demographic characteristics was used to measure organizational-level gender and racioethnic diversity. The 2001 employee survey data provided information on gender and race, and organizationallevel demographic composition was then assessed by aggregating individual employee responses. Harrison and Klein (2007) recommend the use of Blau’s (1977) index to measure demographic composition when conceptual arguments focus on the variety of knowledge, information, and ideas present in an organization. Following their recommendation, we calculated the Blau index for the racial composition of each workplace based on the categories of White, Black, or African origin, Bi-Racial or Multiple Minority backgrounds, and Asian/Filipino/Middle Eastern origin or other. Blau’s (1977) index is problematic for dimensions such as gender which have only two categories because its range is truncated such that the maximum value is only 0.5 rather than the theoretical 1.0 (Harrison & Sin, 2006). For ˆ this reason, a proportional measure generally has more statistical power (Williams & Mean, 2004). We followed prior authors (e.g., Frink et al., 2003) and used the percentage of women as the measure of gender composition, adding a squared term to capture any nonlinear effects resulting from the impact of variety. To check the validity of aggregating the demographic data reported by employees to the establishment level, we compared the WES establishment-level data to 2001 Census statistics provided by Statistics Canada (2009). Statistics Canada does not provide industry comparisons on the basis of race, ethnicity, or immigration status; however, it does provide industry comparisons by gender. We compared the average percentage of women calculated for establishments based on the WES employee data to the percentage of women employed in each of the 14 industry categories based on the Census data. The correlation between the two percentages was 0.99 ( p < 0.001), the Spearman r was 0.98 ( p < 0.001), and the Kendall t was 0.93 ( p < 0.001). These comparisons showed that our measure of the percentage of women in the establishment was very consistent with the Census data percentage of women in the same industry and supported the validity of using the employee data to measure establishment-level demographic diversity. Table 1 shows that the average percentage of women in the sampled establishments was 59 per cent, while the average percentage of people of color was 14 per cent. The average amount of racial diversity as assessed by Blau’s index was 0.18. Employee involvement The 2001 employee survey included several questions about participation in employee involvement activities. The items were not designed to cover the three areas of building skills, motivating, and empowering employees, yet together, they tap into each of these dimensions. We indicate links to these three conceptual dimensions in parentheses next to each item. The specific questions asked how frequently the individual participated in each of the following activities: Completing employee surveys (motivating, empowering), offering suggestions (empowering, skill building), receiving job rotation or cross-training (skill-building, motivating), being informed about important issues such as organizational performance (motivating, empowering), participating in task teams or labormanagement committees (empowering, motivating, skill-building), problem-solving teams (empowering, motivating, skill-building), and self-directed teams (empowering, motivating, skill-building). The items assessing the use of teams used four-point scales (1 ¼ Never, 2 ¼ Occasionally, 3 ¼ Frequently, 4 ¼ Always), while three-point scales were applied to other items (1 ¼ Never, 2 ¼ Occasionally, 3 ¼ Frequently). The Cronbach’s a of the index combining all of these items was 0.94. To measure organizational-level employee involvement, we first conducted a factor analysis of these seven items at the individual level and calculated each employee’s factor score in order to deal with the scaling differences on the original survey items. Then, we averaged the factor scores across employees in the same organization to assess the level of employee involvement. A higher mean indicated a higher level of employee involvement. Following
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1072

Y. YANG AND A. M. KONRAD

previous studies of organizational climate (e.g., Schneider, Salvaggio, & Subirats, 2002), variation in employee involvement was measured by taking the standard deviation of the employee factor scores. A higher standard deviation indicated a higher level of variation. The employee involvement measures utilized survey responses of non-supervisory employees only. We calculated intraclass correlation coefficients (ICCs), ICC(1) and ICC(2), to assess the appropriateness of aggregating data to assess employee involvement. The ICC(1) of individual involvement was 0.29, which exceeded the conventionally accepted value of 0.05 (Bliese, 2000); and the ICC(2) was 0.67, which exceeded the conventionally accepted value of 0.60 (Ostroff & Schmit, 1993). We also calculated the rwg for this measure. The denominator for the rwg statistic is dependent upon the scaling of the item used (Cohen, Doveh, & Eick, 2001); therefore, we calculated a separate rwg for the survey items with three response options (0.83) and the survey items with four response options (0.62) (average rwg across all items ¼ 0.73). These rwg statistics were adequate in size to support the aggregation of the individual employee involvement scores to create an organizational-level measure. Involvement of historically predominant and marginalized groups Similar to the measure of employee involvement, women’s and men’s involvement was assessed by taking the average involvement factor scores of women and men in the same organizations. For involvement of racioethnic groups, we first recoded the racioethnic categories into white or historically predominant group and people of color or historically marginalized groups. We then calculated the average involvement factor scores for each of these two groups to assess the involvement of historically predominant and marginalized racial groups, respectively. Organizational innovation The organizational innovation measure was taken from managers’ responses to the workplace survey. The 2002 workplace survey asked whether the organization had (1) introduced new products/services, (2) improved products/ services, (3) introduced new processes, and (4) improved processes in the previous year (Yes ¼ 1, No ¼ 0). The average of these four items was used to assess organizational innovation (a ¼ 0.74). Table 1 shows that the average innovation score reported for these establishments was 0.48, indicating that the average establishment agreed with almost two of the four innovation survey items. Controls We controlled for industry, unionization, percentage of professionals, innovation strategy, and people strategy, all measures taken from managers’ responses to the 2001 workplace survey. We coded the 14 industries in the sample into 13 dummy variables (i.e., Labor Intensive Tertiary Manufacturing, Primary Product Manufacturing, Secondary Product Manufacturing, Capital Intensive Tertiary Manufacturing, Construction, Transportation/Warehousing/ Wholesale Trade, Communication and Other Utilities, Retail Trade & Consumer Services, Finance and Insurance, Real Estate, Rental, Leasing Operations, Business Services, Education, Information and Cultural Services) with the Forestry/Mining/Oil and Gas extraction industry serving as the comparison category. This is a common procedure for studies conducted at the establishment level (e.g., Derfus, Maggitti, Grimm, & Smith, 2008; Kim, Hoskisson, & Wan, 2004; Shane, 2001). To assess unionization, we coded those establishments with at least one non-managerial employee being covered by a collective agreement as ‘‘1’’, and others as ‘‘0’’. Table 1 shows that 88 per cent of the establishments in the sample were unionized. We also controlled for the percentage of professionals in an organization as it may influence innovation (average for the sample was 31 per cent). Because gender and race composition are associated with occupational distributions reflecting task differences and opportunities for innovation (Konrad & Pfeffer, 1991; Tomaskovic-Devey, Zimmer, Stainback, Robinson, Taylor, & McTague, 2006), we controlled for the gender (race) composition of the firm when examining the impact of racial (gender) composition. Becker and Huselid (2006) recommend controlling for other management factors when studying the impact of human resource management, and previous studies have shown that innovation strategy influences the effects of diversity on organizational outcomes (Richard, McMillan, Chadwick, & Dwyer, 2003). This measure consisted of
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1073

the average of three items assessing the importance of: Undertaking research and development, developing new products or services, and developing new production or operating techniques (1 ¼ ‘‘not important’’, 5 ¼ ‘‘crucial’’; a ¼ 0.79, average for the sample ¼ 3.00). Similarly, people strategy, which reflects how important employees are to a company’s business strategy, influences the effectiveness of employee involvement (Zatzick & Iverson, 2006). It was measured by the average of two items assessing the importance of increasing employee skills and increasing employee involvement (1 ¼ ‘‘not important’’, 5 ¼ ‘‘crucial’’; a ¼ 0.86, average for the sample ¼ 3.28).

Results
Table 2 shows the results of the regressions examining the relationship between gender and racial diversity and organizational innovation. In all regressions, the predictors were mean centered to control for multicollinearity when testing interactions. Moreover, because field data usually lacks power for detecting interaction effects, we set significance levels to p < 0.10 for the interaction terms (Harrison, Price, & Bell, 1998). In both sets of regressions, an innovation strategy and the percentage of women employed were significant positive predictors of the amount of organizational innovation reported by the establishment. None of the other main effects were significant.

Table 2. Diversity and organizational innovation DV ¼ Organizational innovation Step 1 Predictor Controls Industry: 13 categories controlled Innovation strategy People strategy Unionization (1 ¼ yes, 0 ¼ no) % of professionals (1) Racioethnic diversity (1) % of women (2) Squared % of women (3) Involvement level (4) Involvement variation Two-way interactions (1) Â (3) (2) Â (3) (1) Â (4) (2) Â (4) (3) Â (4) Three-way interaction (1) Â (3) Â (4) (2) Â (3) Â (4) Change in R2 at step 1 Change in R2 at step 2 Change in R2 at step 3 F Overall adjusted R2 N
ÃÃ

% Women final b

Racioethnic diversity final b

0.21Ã À0.02 0.09 À0.03 À0.04 0.40ÃÃ À0.02 0.06 0.03 À0.02 À0.02 À0.09 À0.09 À0.03 À0.15y 0.17Ã 0.01 0.01 2.14ÃÃ 0.15 182

0.20Ã 0.00 0.10 À0.03 À0.04 0.68Ã À0.21 0.04 À0.04 À0.10 0.02 À0.05 0.16y 0.26ÃÃ 0.02 0.02y 2.47ÃÃÃ 0.17 182

2

3

y

p < 0.10; Ãp < 0.05;

p < 0.01;

ÃÃÃ

p < 0.001.

Copyright # 2010 John Wiley & Sons, Ltd.

J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1074

Y. YANG AND A. M. KONRAD

Hypothesis 1 predicted that employee participation in involvement behaviors would positively moderate the association between the representation of women and racioethnic minorities and organizational innovation. For the dimension of gender, this hypothesis was tested by examining the significance of the two-way multiplicative interaction terms, percentage of women  involvement level and percentage of women squared  involvement level (Cohen, Cohen, West, & Aiken, 2003, p. 292–295). For the dimension of racioethnicity, it was tested with the two-way interaction of Blau’s index of racioethnic diversity  involvement level (Table 2). All of these interaction terms were non-significant, indicating no support for the hypothesis. Hypothesis 2 predicted that low variation combined with a high level of employee participation in involvement behaviors would positively moderate the association between representation of women and racioethnic minorities and organizational innovation. For the dimension of gender, this hypothesis was tested by examining the significance of the three-way multiplicative interaction terms: percentage of women  involvement level  involvement variation and percentage of women squared  involvement level  involvement variation (Cohen et al., 2003). These terms did not add significantly to the proportion of variance explained in the prediction equation. For the dimension of racioethnic diversity, Hypothesis 2 was tested by examining the three-way interaction between racioethnic diversity  involvement level  involvement variation (Table 2). This three-way interaction term added significantly to the proportion of variance explained in the dependent variable ( p ¼ 0.052), supporting further examination of the form of the interaction to determine whether it was consistent with the direction of the effect predicted in the hypothesis. The form of the interaction is shown in Figure 1a and b. At high levels of employee involvement, the simple slope of the regression line indicating the association between racioethnic diversity and innovation was substantially larger and more positive under the condition of high variation in involvement (1.66, t ¼ 1.84, p ¼ 0.068) than under the condition of low variation in involvement (0.78, t ¼ 1.53, ns). This difference in slopes indicates that when level and variation in employee involvement were both high, the level of racioethnic diversity was positively related to organizational innovation. When level of involvement was high and variation was low, racioethnic diversity was unrelated to innovation. At low levels of employee involvement, the simple slope of the regression line indicating the association between racioethnic diversity and innovation was non-significant for either high or low variation in involvement. This finding was consistent with our reasoning that when the employer is not committed to a high level of employee involvement, racioethnic diversity and variation in involvement are unrelated to innovation. The form of this interaction did not fit our reasoning that a high level of involvement with high variation would not lead to a positive association between diversity and innovation. We made this prediction because we had assumed that when involvement variation was high, members of historically marginalized groups would be less active in involvement behaviors than members of historically predominant groups. This assumption proved to be incorrect. Examining establishments above the sample median on level of participative behaviors, we found that when racioethnic diversity was present (51 per cent of the sampled establishments were 100 per cent White), members of racioethnic minority groups reported a higher level of participation in involvement behaviors than their White counterparts did. When variation in participation was low, the average participation score for White employees was 0.57 compared to 0.74 for racioethnic minorities (no. of establishments ¼ 23, t ¼ À1.39, ns). When variation in participation was high, the average participation score for White employees was 0.74 compared to 1.02 for racioethnic minorities (no. of establishments ¼ 35, t ¼ 1.38, ns). In other words, in racioethnically diverse high involvement workplaces, racioethnic minority groups consistently showed greater involvement than their White counterparts did, although the differences were not statistically significant. Furthermore, when variation in involvement was high, racioethnic minority employees showed a tendency toward higher average levels of involvement (t ¼ À1.70, p < 0.10) compared to when variation was low. Linking these findings to the plot of the triple interaction effect in Figure 1a, the simple slope of the association between diversity and innovation was significantly more positive when there was high variation in involvement, which reflected higher average involvement of racioethnic minority employees.
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1075

Figure 1. (a) Interaction of racioethnic diversity and involvement variation in establishments with a high level of involvement. (b) Interaction of racioethnic diversity and involvement variation in establishments with a low level of involvement

Hypothesis 3 predicted that involvement behaviors by historically marginalized groups would moderate the association between involvement behaviors by traditionally dominant groups and innovation such that the slope becomes significantly more positive (i.e. upward-slanting) as involvement by historically marginalized groups increases. For the gender dimension, this hypothesis was tested with the data from those 155 workplaces employing both women and men and examining the significance of the two-way multiplicative interaction term of women’s average involvement  men’s average involvement (Table 3). Findings indicated that the interaction between women’s and men’s level of involvement was not a significant predictor of organizational innovation. For the racioethniciy dimension, Hypothesis 3 was tested with the data from those 89 workplaces employing both White and racioethnic minority employees and examining the significance of the two-way multiplicative interaction term of White employees’ involvement  ethnic minority’s involvement (Table 3). This interaction term was a
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1076

Y. YANG AND A. M. KONRAD

Table 3. Involvement and Organizational Innovation in Diverse Workplaces DV ¼ Organizational innovation Gender and involvement final b Racioethnic diversity and involvement final b

Step 1

Predictor Controls Industry: 13 categories controlled Innovation strategy People strategy Unionization (1 ¼ yes, 0 ¼ no) % of professionals % of women Squared % of women Racioethnic diversity (centered Blau index) Main effects (1) Male (white) involvement (2) Female (minority) involvement Two-way interaction (1) Â (2) Change in R2 at step 1 Change in R2 at step 2 Change in R2 at step 3 F Overall adjusted R2 N
ÃÃ

0.11 À0.09 À0.01 0.03 0.82y À0.38 À0.05 0.25Ã À0.07 À0.11 0.18y 0.11ÃÃ 0.03 2.39ÃÃ 0.19 155

0.08 0.02 0.15 À0.16 0.96Ã À0.41 À0.05 0.23Ã 0.25Ã 0.37ÃÃ 0.03 0.04Ã 2.56ÃÃ 0.27 89

2 3

y

p < 0.10; Ãp < 0.05;

p < 0.01;

ÃÃÃ

p < 0.001.

significant and positive predictor of innovation. The form of the interaction is plotted in Figure 2. The simple slope of the regression line indicating the association between the White involvement and innovation was significant and positive under the condition of high racioethnic minority involvement (0.34, t ¼ 2.73, p < 0.01), but non-significant under the condition of low minority involvement (0.06, t ¼ 1.14, ns). This finding was consistent with the prediction Hypothesis 3.

Figure 2. Interaction of racioethnic majority and minority involvement Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1077

Discussion
We brought together the two previously separate streams of research on workplace diversity and employee involvement to examine their joint effects on organizational innovation. The business case for diversity (Cox, 2001; Cox & Blake, 1991; Robinson & Dechant, 1997) as well as conceptualizations of the creativity process (Kilgour, 2006; Leung et al., 2008) argue that diverse organizations will be more innovative, due to the wider variety of information and knowledge resources available for generating new ideas. Also, employee involvement perspectives argue that involving employees more fully in learning and decision-making processes will result in the development of more and better ideas (Appelbaum et al., 2000; Arthur & Aiman-Smith, 2001; Hunter et al., 2002; Kochan & Rubinstein, 2000). Integrating these lines of argument, we predicted that the association between diversity and innovation would be maximized under the condition of greater employee involvement, and particularly with low variation in involvement where historically marginalized and predominant groups participate equally in involvement behaviors. Our findings do not completely support this prediction; however, the pattern of findings does justify the importance of garnering employee involvement among members of historically excluded groups for organizational innovation. The three-way interaction effect, we observed, shows that when level of employee involvement is high, racioethnic diversity is positively related to innovation under the condition of high variation in involvement where minority employees are equally or more active in involvement behaviors than their White counterparts. Racioethnic diversity is unrelated to innovation when level of employee involvement is low or under the condition of high level with low variation in involvement. One reason why a high level combined with high variation in involvement behaviors strengthened the association between racioethnic diversity and innovation could be because under conditions of higher variation in involvement, a subset of employees becomes very highly involved in the learning and decision-making processes of the organization. Under the condition of both high level and high variation, both White and racioethnic minority employees show a relatively high average in level of involvement, which indicates the development of a core group of highly involved employees who pull up the establishment’s average. The significant positive link between diversity and innovation in this set of establishments may well reflect the value of including members of a variety of racioethnic groups among this subset of very highly involved employees. Consistent with this logic, subsequent analysis supported our prediction that higher levels of involvement by the historically marginalized group would strengthen the association between the level of involvement among the traditionally predominant group and organizational innovation. High involvement by historically marginalized groups is most likely to bring new ideas to the organization, while high involvement by traditionally predominant groups is important to the elaboration process (van Knippenberg et al., 2004) in order to maximize fit and benefits to the organization. Traditionally predominant groups also often control organizational systems and positions crucial to the implementation of new ideas (Harrison & Humphrey, 2010), such that their involvement in innovative processes is essential for overcoming resistance to organizational change.

Theoretical implications
Previous authors have pointed at inconsistency in the implementation of employee involvement across the organization as a major reason why the results of involvement programs are sometimes less than stellar (Riordan et al., 2005; Shadur et al., 1999). The three-way interaction effect, we observed, adds inclusion of historically marginalized racioethnic groups as another factor that can enhance the value of employee involvement. This finding suggests that at minimum, empowerment practices that successfully generate a high overall average of involvement across employees are needed before diversity becomes associated with innovation. Then, supporting a very high level of involvement among members of racioethnic minority groups creates innovation by enhancing the impact of the White majority’s involvement behaviors. High involvement by racioethnic minorities creates variation across the
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1078

Y. YANG AND A. M. KONRAD

establishment, however, that variation can be beneficial to involvement so long as both the traditionally predominant and the traditionally marginalized groups have access to empowerment processes. Consistent with the CEM model (van Knippenberg et al., 2004), the jointly positive effect of high involvement of both traditionally predominant and marginalized groups on organizational innovation suggests that creativity can be boosted when the elaboration process is facilitated. The CEM model argues that elaboration will be hindered when categorization between demographic groups operates. This implies that the mere presence of diversity is not sufficient for generating and elaborating upon a variety of ideas. Our finding indicates that practices that effectively empower different demographic groups and integrate their various views can be beneficial to organizational innovation. The relationship between gender composition and innovation was a much more straightforward positive main effect of the percentage of women. The differences in our findings for gender and racioethnicity are consistent with the conclusions of other researchers that gender dynamics are different from the inter-group dynamics between racioethnic groups (Goldberg, 2005). As such, our research also supports a recent suggestion that diversity is meaningful only when it is narrowly dimensionalized (Harrison & Sin, 2006). Combining several types of diversity, such as gender, racioethnicity, tenure, functional specialty, etc., into a single measure of diversity ignores the possibility that different mechanisms drive inter-group dynamics across different dimensions of diversity.

Managerial implications
Our findings suggest that beyond ensuring consistent implementation of involvement programs across the organization (Riordan et al., 2005; Shadur et al., 1999), managers of diverse organizations need to obtain relatively high involvement levels among members of historically marginalized racioethnic groups. Diversity management practices might be useful for attaining this goal. Although the field of diversity has not generated a clear set of effective management practices (Kalev et al., 2006), evidence has emerged supporting the effectiveness of certain diversity management practices. For instance, the development of affinity groups or employee networks helping to make connections among racioethnic minorities or women have been linked to positive attitudinal and career outcomes for historically marginalized groups (Friedman & Holtom, 2002; Friedman, Kane, & Cornfield, 1998). Strengthening the networks of marginalized groups in the organization could enhance information-sharing about opportunities to participate effectively in involvement activities. Evidence also supports the effectiveness of explicit communication regarding the value of diversity to the organization. Top management support for the value of diversity has been linked to the effectiveness of diversity ´ ´ management in several studies (Camelo-Ordaz, Fernandez-Alles, & Martınez-Fierro, 2006; Combs, 2002; Konrad & Linnehan, 1995; Leonard & Grobler, 2006). Recent evidence suggests that when groups are persuaded of the value of diversity to team performance, diverse teams outperform homogeneous ones (Homan, van Knippenberg, Van Kleef, & De Dreu, 2007). Together these findings suggest that clear communication from top management articulating the reasons why diversity enhances the organization’s ability to fulfill its mission helps employees to overcome intergroup biases and cooperate in involvement activities.

Limitations and directions for future research
One limitation of this study is that we were unable to measure the theoretical mechanisms which, we argued, explain the joint effects of diversity and employee involvement on organizational innovation. The CEM posits that categorization and elaboration processes affect innovation outcomes (van Knippenberg et al., 2004). Employee involvement research suggests that employee empowerment (Spreitzer, 1995, 1996) is a critical process resulting in innovation. Social closure theory points to network building (Tomaskovic-Devey, 1993) as a key factor driving the decision-making processes influencing organizational innovation. None of these processes were measured in the WES data we used to test our hypotheses. The benefits of being able to examine the joint effects of diversity and employee involvement in a large representative sample of establishments made the use of this dataset highly attractive, however. Future research could make a contribution to knowledge in this area by gathering direct
Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1079

measures of categorization and elaboration processes, employee empowerment, and social networks in workplaces and linking these processes to the level of diversity and employee involvement. Another limitation of this study is the small number of employees responding to the survey measures for each of the participating organizations. The number of employee respondents ranged from 5 to 21, and the average is six employees per organization. The aggregated measures based on such a small number of employees may not accurately reflect the organizational-level variables of demographic composition and employee involvement. However, inter-class correlations and rwg statistics supported the aggregation of the employee involvement measures, and comparisons of demographic composition estimates with industry-level data validated the measure of gender composition. It is difficult to get large numbers of survey responses from a representative number of establishments (Becker & Huselid, 2006), so future research might overcome this problem by gathering a larger number of employee responses from a more limited set of organizations to complement the strengths and weaknesses of this study. Finally, the WES data included no information about diversity management practices. Such practices might include the development of employee affinity groups, training leaders to manage diverse teams effectively, examining employment practices to remove barriers to career development for members of historically marginalized groups, and clear communication about how diversity helps the organization to perform better. Diversity management practices might be a necessary addition to employee involvement programs in order to ensure that members of historically marginalized groups are fully empowered to participate in involvement activities. Future research should examine the possible joint effects of employee involvement and diversity management practices for generating innovation in diverse organizations.

Acknowledgements
Dr Alison M. Konrad thanks the Corus Entertainment Chair in Women in Management at the Richard Ivey School of Business as well as by Social Sciences and Humanities Research Council of Canada (SSHRCC) Standard Research Grant no. 410-2007-1673 for providing support to this project. Yang Yang gratefully acknowledges the support of the postdoctoral fellowship award from the Social Sciences and Humanities Council of Canada.

Author biographies
Yang Yang, Ph.D., is a Postdoctoral Fellow at the Management Department of the Wharton School, University of Pennsylvania. Her research interests include workplace inclusivity, diversity initiatives, human resource management, and organizational justice. Alison M. Konrad, Ph.D., joined the Richard Ivey School of Business, U. of Western Ontario in 2003 as a Professor of Organizational Behavior and holder of the Corus Entertainment Chair in Women in Management. Her research focuses on organizational diversity initiatives and making workplaces more inclusive.

References
Allport, G.W. (1954). The Nature of Prejudice. Cambridge, MA: Addison-Wesley. Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. (2000). Manufacturing advantage: Why high-performance work systems pay off. Ithaca, NY: Cornell University Press. Copyright # 2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1080

Y. YANG AND A. M. KONRAD

Arthur, J. B., & Aiman-Smith, L. (2001). Gainsharing and organizational learning: An analysis of employee suggestions over time. Academy of Management Journal, 44, 737–754. Batt, R. (2000). Strategic segmentation in front-line services: Matching customers, employees and human resource systems. International Journal of Human Resource Management, 11, 540–561. Becker, B. E., & Huselid, M. A. (2006). Strategic human resources management: Where do we go from here? Journal of Management, 32, 898–925. Benson, G. S., Young, S. M., & Lawler, E. E. III, (2006). High-involvement work practices and analysts’ forecasts of corporate earnings. Human Resource Management, 45, 519–537. Blau, P. M. (1977). Inequality and heterogeneity. New York: Free Press. Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K. Klein, & S. W. Kozlowski (Eds.), Multilevel theory, research and methods in organizations (pp. 349–381). San Francisco: Jossey-Bass. Bliese, P. D., & Halverson, R. (1998). Group size and measures of group-level properties: An examination of eta-squared and ICC values. Journal of Management, 24, 157–172. Bowers, C. A., Pharmer, J. A., & Salas, E. (2000). When member homogeneity is needed in work teams: A meta-analysis. Small Group Research, 31, 305–327. Brewer, M. B. (1979). In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 105, 475–482. Brewer, M. B., & Brown, R. J. (1998). Intergroup relations. In D. T. Gilbert, & S. T. Fiske (Eds.), Handbook of Social Psychology (fourth edition, pp. 554–594). Boston: McGraw-Hill. Cady, S., & Valentine, H. J. (1999). Team innovation and perceptions of consideration: What difference does diversity make? Small Group Research, 30, 730–750. ´ ´ Camelo-Ordaz, C., Fernandez-Alles, M., & Martınez-Fierro, S. (2006). Influence of top management team vision and work team characteristics on innovation: The Spanish case. European Journal of Innovation Management, 9, 179–201. Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (third edition). Mahwah, NJ: Erlbaum. Cohen, A., Doveh, E., & Eick, U. (2001). Statistical properties of the rWG(J) index of agreement. Psychological Methods, 6, 297–310. Combs, G. M. (2002). Meeting the leadership challenge of a diverse and pluralistic workplace: Implications of self-efficacy for diversity training. Journal of Leadership Studies, 8, 1–16. Combs, J., Liu, Y., Hall, A., & Ketchen, D. (2006). How much do high-performance work practices matter? A meta-analysis of their effects on organizational performance. Personnel Psychology, 59, 501–528. Cox, T. Jr., (2001). Creating the multicultural organization: A strategy for capturing the power of diversity. San Francisco: Jossey-Bass. Cox, T., Jr., & Blake, S. (1991). Managing cultural diversity: Implications for organizational competitiveness. Academy of Management Executive, 5, 45–56. Crocker, J., & Luhtanen, R. (1990). Collective self-esteem and ingroup bias. Journal of Personality & Social Psychology, 58, 60–67. Crocker, J., Luhtanen, R., Broadnax, S., & Blaine, B. E. (1999). Belief in U.S. government conspiracies against blacks among black and white college students: Powerlessness or system blame? Personality & Social Psychology Bulletin, 25, 941–952. Crossan, M. M., Lane, H. W., & White, R. E. (1999). An organizational learning framework: From intuition to institution. Academy of Management Review, 24, 522–537. Cummings, J. N. (2004). Work groups, structural diversity, and knowledge sharing in a global organization. Management Science, 50, 352–364. Damanpour, F. (1996). Organizational complexity and innovation: Developing and testing multiple contingency models. Management Science, 42, 693–716. Darling-Hammond, L. (2006). No child left behind and high school reform. Harvard Educational Review, 76, 642–667. Datta, D. K., Guthrie, J. P., & Wright, P. M. (2005). Human resource management and labor productivity: Does industry matter? Academy of Management Journal, 48, 135–145. De Dreu, C. K. W., & Weingart, L. R. (2003). Task versus relationship conflict, team performance, and team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88, 741–749. Deci, E. L., & Ryan, R. M. (2000). The ‘‘what’’ and ‘‘why’’ of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11, 227–268. Delery, J. E., & Shaw, J. D. (2001). The strategic management of people in work organizations: Review, synthesis, and extension. In G. R. Ferris (Ed.), Research in personnel and human resources management (Vol. 20, pp. 165–197). Stamford, CT: JAI Press.

Copyright # 2010 John Wiley & Sons, Ltd.

J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1081

Den Hartog, D. N., & Verburg, R. M. (2004). High performance work systems, organisational culture and firm effectiveness. Human Resource Management Journal, 14, 55–78. Derfus, P. J., Maggitti, P. G., Grimm, C. M., & Smith, K. G. (2008). The red queen effect: Competitive actions and firm performance. Academy of Management Journal, 51, 61–80. Earley, P. C. (1989). Social loafing and collectivism: A comparison of the United States and the People’s Republic of China. Administrative Science Quarterly, 44, 565–581. Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44, 350–383. Evans, W. R., & Davis, W. D. (2005). High-performance work systems and organizational performance: The mediating role of internal social structure. Journal of Management, 31, 758–775. Feingold, A. (1994). Gender differences in personality: A meta-analysis. Psychological Bulletin, 116, 429–456. Fichman, R. G., & Kemerer, C. F. (1997). The assimilation of software process innovation: An organizational learning perspective. Management Science, 43, 1345–1363. Francoeur, C., Labelle, R., & Sinclair-Desgagne, B. (2007). Gender diversity in corporate governance and top management. Journal of Business Ethics, 81, 83–95. ¨ Fransson, S., & Thornqvist, C. (2006). Some notes on workplace equality renewal in the Swedish labor market. Gender, Work & Organization, 13, 606–620. Friedman, R. A., & Holtom, B. (2002). The effects of network groups on minority employee turnover intentions. Human Resource Management, 41, 405–421. Friedman, R. A., Kane, M., & Cornfield, D. B. (1998). Social support and career optimism: Examining the effectivenes of network groups among black managers. Human Relations, 51, 1155–1177. Frink, D. D., Robinson, R. K., Reithel, B., Arthur, M. M., Ammeter, A. P., Ferris, G. R., et al. (2003). Gender demography and organization performance: A two-study investigation with convergence. Group & Organization Management, 28, 127–147. Fronda, Y., & Moriceau, J.-L. (2008). I am not your hero: Change management and culture shocks in a public sector corporation. Journal of Organizational Change Management, 21, 589–609. Goldberg, C. (2005). Relational demography and similarity-attraction in interview assessments and subsequent offer decisions: Are we missing something? Group & Organization Management, 30, 597–624. Gonzalez, J. A., & DeNisi, A. S. (2009). Cross-level effects of demography and diversity climate on organizational attachment and firm effectiveness. Journal of Organizational Behavior, 30, 21–40. Guest, D. E., Conway, N., & Dewe, P. (2004). Using sequential tree analysis to search for ‘bundles’ of HR practices. Human Resource Management Journal, 14, 79–96. Guest, D. E., Michie, J., Conway, N., & Sheehan, M. (2003). Human resource management and corporate performance in the UK. British Journal of Industrial Relations, 41, 291–314. Guthrie, J. P. (2001). High-involvement work practices, turnover and productivity: Evidence from New Zealand. Academy of Management Journal, 44, 180–190. Harlan, S., & Steinberg, R. J. (1989). Job training for women: The promise and limits of public policies. Philadelphia, PA: Temple University Press. Harrison, D. A., & Humphrey, S. E. (2010). Designing for diversity or diversity for design? Tasks, interdependence, and withinunit differences at work. Journal of Organizational Behavior, 31, 328–337. Harrison, D. A., & Klein, K. J. (2007). What’s the difference? Diversity construct as separation, variety, or disparity in organizations. Academy of Management Review, 32, 1199–1228. Harrison, D. A., Price, K. H., & Bell, M. P. (1998). Beyond relational demography: Time and the effects of surface- and deep-level diversity on work group cohesion. Academy of Management Journal, 41, 96–107. Harrison, D. A., & Sin, H. (2006). What is diversity and how should it be measured? In A. M. Konrad, P. Prasad, & J. K. Pringle (Eds.), Handbook of Workplace Diversity (pp. 191–216). Thousand Oaks, CA: Sage. Heilman, M. E., Block, C. J., & Stathatos, P. (1997). The affirmative action stigma of incompetence: Effects of performance information ambiguity. Academy of Management Journal, 40, 603–625. Hirst, G., van Knippenberg, D., & Zhou, J. (2009). A cross-level perspective on employee creativity: Goal orientation, team learning behavior, and individual creativity. Academy of Management Journal, 52, 280–293. Homan, A. C., van Knippenberg, D., Van Kleef, G. A., & De Dreu, C. K. W. (2007). Bridging faultlines by valuing diversity: Diversity beliefs, information elaboration, and performance in diverse work groups. Journal of Applied Psychology, 92, 1189– 1199. Hunter, L. W., MacDuffie, J. P., & Doucet, L. (2002). What makes teams take? Employee reactions to work reforms. Industrial & Labor Relations Review, 55, 448–472. Ichniowski, C., Kochan, T. A., Levine, D., Olson, C., & Strauss, G. (1996). What works at work: Overview and assessment. Industrial Relations, 35, 299–333.

Copyright # 2010 John Wiley & Sons, Ltd.

J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

1082

Y. YANG AND A. M. KONRAD

Jehn, K. A., Northcraft, G. B., & Neale, M. A. (1999). Why differences make a difference: A field study of diversity, conflict, and performance in workgroups. Administrative Science Quarterly, 44, 741–763. Joshi, A., & Roh, H. (2009). The role of context in work team diversity research: A meta-analytic review. Academy of Management Journal, 52, 599–628. Kalev, A., Dobbin, F., & Kelly, E. (2006). Best practices or best guesses? Assessing the efficacy of corporate affirmative action and diversity policies. American Sociological Review, 71 589–617. Kashima, Y., Hamaguchi, S., Kim, U., Choi, S., Gelfand, M. J., & Yuki, M. (1995). Culture, gender, and self: A perspective from individualism-collectivism research. Journal of Personality & Social Psychology, 69, 925–937. Kilgour, M. (2006). Improving the creative process: Analysis of the effects of divergent thinking techniques and domain specific knowledge on creativity. International Journal of Business & Society, 7, 79–107. Kim, H., Hoskisson, R. E., & Wan, W. P. (2004). Power dependence, diversification strategy, and performance in Keiretsu member firms. Strategic Management Journal, 25, 613–636. Kochan, T. A., Bezrukova, K., Jackson, S., Joshi, A., Jehn, K. E., Leonard, D., et al. (2003). The effects of diversity on business performance: Report of the diversity research network. Human Resource Management, 42, 3–21. Kochan, T. A., & Rubinstein, S. A. (2000). Toward a stakeholder theory of the firm: The Saturn partnership. Organization Science, 11, 367–386. Konrad, A. M., & Linnehan, F. (1995). Formalized HRM structures: Coordinating equal employment opportunity or concealing organizational practices? Academy of Management Journal, 38, 787–820. Konrad, A. M., & Pfeffer, J. (1991). Understanding the hiring of women and minorities in educational institutions. Sociology of Education, 64, 141–157. Konrad, A. M., Ritchie, J. E., Jr., Lieb, P., & Corrigall, E. A. (2000). Sex differences and similarities in job attribute preferences: A meta-analysis. Psychological Bulletin, 126, 593–641. Laschinger, H. K., Finegan, J., Shamian, J., & Wilk, P. (2001). Impact of structural and psychological empowerment on job strain in nursing work settings: Expanding Kanter’s model. Journal of Nursing Administration, 31, 260–272. Lawler, E. E. III, (1986). High-involvement management. San Francisco: Jossey-Bass. Leonard, A., & Grobler, A. F. (2006). Exploring challenges to transformational leadership communication about employment equity: Managing organizational change in South Africa. Journal of Communication Management, 10, 390–406. Lepak, D. P., Taylor, M. S., Tekleab, A., Marrone, J. A., & Cohen, D. J. (2007). An examination of the use of high-investment human resource systems for core and support employees. Human Resource Management, 46, 223–246. Leung, A. K., Maddux, W. W., Galinsky, A. D., & Chiu, C. (2008). Multicultural experience enhances creativity. American Psychologist, 63, 169–181. Liao, H., Joshi, A., & Chuang, A. (2004). Sticking out like a sore thumb: Employee dissimilarity and deviance at work. Personnel Psychology, 57, 969–1000. Maume, D. .J Jr., (1999). Glass ceilings and glass escalators: Occupational segregation and race and sex differences in managerial promotions. Work & Occupations, 26, 483–509. Nishii, L. H., Lepak, D. P., & Schneider, B. (2008). Employee attributions of the ‘‘why’’ of HR practices: Their effects on employee attitudes and behaviors, and customer satisfaction. Personnel Psychology, 61, 503–545. Ostroff, C., & Schmit, N. (1993). Configurations of organizational effectiveness and efficiency. Academy of Management Journal, 36, 1345–1361. Pelled, L. H., Eisenhardt, K. M., & Xin, K. R. (1999). Exploring the black box: An analysis of work group diversity, conflict, and performance. Administrative Science Quarterly, 44, 1–28. Pendakur, K., Pendakur, R., & Woodcock, S. D. (2008). A representation index: Measuring the representation of minorities in the income distribution. B. E. Journal of Economic Analysis & Policy, 8: Iss. 1 (Advances), Article 44. DOI: 10.2202/19351682.1942 Available at: www.bepress.com/bejeap/vol8/iss1/art44 Pettigrew, T. F. (1998). Intergroup contact theory. Annual Review of Psychology, 49, 65–85. Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality & Social Psychology, 90, 751–783. Pfeffer, J. (1998). Seven practices of successful organizations. California Management Review, 40, 96–124. Richard, O. (2000). Racial diversity, business strategy, and firm performance: A resource-based view. Academy of Management Journal, 43, 164–177. Richard, O., McMillan, A., Chadwick, K., & Dwyer, S. (2003). Employing an innovation strategy in racially diverse workforces: Effects on firm performance. Group & Organization Management, 28, 107–126. Ridgeway, C. L. (1991). The social construction of status value: Gender and other nominal characteristics. Social Forces, 70, 367– 386. Riordan, C. M., Vandenberg, R. J., & Richardson, H. A. (2005). Employee involvement climate and organizational effectiveness. Human Resource Management, 44, 471–488. Robinson, G., & Dechant, K. (1997). Building a business case for diversity. Academy of Management Executive, 11, 21–31.

Copyright # 2010 John Wiley & Sons, Ltd.

J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job

DIVERSITY, EMPOWERMENT AND ORGANIZATIONAL INNOVATION

1083

Rodan, S. (2002). Innovation and heterogeneous knowledge in managerial contact networks. Journal of Knowledge Management, 6, 152–163. Rodan, S., & Galunic, C. (2004). More than network structure: How knowledge heterogeneity influences managerial performance and innovativeness. Strategic Management Journal, 25, 541–562. Schneider, B., Salvaggio, A. N., & Subirats, M. (2002). Climate strength: A new direction for climate research. Journal of Applied Psychology, 87, 220–229. Shadur, M. A., Kienzle, R., & Rodwell, J. J. (1999). The relationship between organizational climate and employee perceptions of involvement: The importance of support. Group & Organization Management, 24, 479–503. Shane, S. (2001). Organizational incentives and organizational mortality. Organization Science, 12, 136–160. Spreitzer, G. M. (1995). Psychological empowerment in the workplace: Dimensions, measurement and validation. Academy of Management Journal, 38, 1442–1465. Spreitzer, G. M. (1996). Social structural characteristics of psychological empowerment. Academy of Management Journal, 39, 483–504. Spreitzer, G. M. (2007). Giving peace a chance: Organizational leadership, empowerment and peace. Journal of Organizational Behavior, 28, 1077–1095. Spreitzer, G. M., & Mishra, A. K. (2002). To stay or to go: Voluntary survivor turnover following an organizational downsizing. Journal of Organizational Behavior, 23, 707–729. Statistics Canada. (2009). Employment by industry and sex (CANSIM Table 282-0008). Available at: http://www40.statcan.gc.ca/101/cst01/labor10a-eng.htm Stedham, Y. E., & Yamamura, J. H. (2004). Measuring national culture: Does gender matter? Women in Management Review, 19, 233–243. Takeuchi, R., Chen, G., & Lepak, D. P. (2009). Through the looking glass of a social system: Cross-level effects of highperformance work systems on employees’ attitudes. Personnel Psychology, 62, 1–29. Tomaskovic-Devey, D. (1993). Gender and racial inequality at work: The sources and consequences of job segregation. Ithaca, NY: ILR Press. Tomaskovic-Devey, D., & Skaggs, S. (2002). Sex segregation, labor process organization, and gender earnings inequality. American Journal of Sociology, 108, 102–128. Tomaskovic-Devey, D., Thomas, M., & Johnson, K. (2005). Race and the accumulation of human capital across the career: A theoretical model and fixed-effects application. American Journal of Sociology, 111, 58–89. Tomaskovic-Devey, D., Zimmer, C., Stainback, K., Robinson, C., Taylor, T., & McTague, T. (2006). Documenting desegregation: Segregation in American workplaces by race, ethnicity, and sex, 1966–2003. American Sociological Review, 71, 565–588. van Knippenberg, D. (1999). Social identity and persuasion: Reconsidering the role of group membership. In D. Abrams, & M. A. Hogg (Eds.), Social identity and social cognition (pp. 315–331). Oxford, UK: Blackwell. van Knippenberg, D., De Dreu, C. K. W., & Homan, A. C. (2004). Work group diversity and group performance: An integrative model and research agenda. Journal of Applied Psychology, 89, 1008–1022. van Knippenberg, D., & Schippers, M. C. (2007). Work group diversity. Annual Review of Psychology, 58, 515–541. van Knippenberg, D., & Wilke, H. (1992). Prototypicality of arguments and conformity to in-group norms. European Journal of Social Psychology, 22, 141–155. Vandenberg, R. J., Richardson, H. A., & Eastman, L. J. (1999). The impact of high involvement work processes on organizational effectiveness: A second-order latent variable approach. Group & Organization Management, 24, 300–339. ˆ Williams, H. M., & Mean, L. J. (2004). Measuring gender composition in work groups: A comparison of existing methods. Organizational Research Methods, 7, 456–474. Yalabik, Z. Y., Chen, S.-J., Lawler, J., & Kim, K. (2008). High-performance work system and organizational turnover in east and southeast Asian countries. Industrial Relations, 47, 145–152. Yap, M., & Konrad, A. M. (2009). Gender and racial differentials in promotions: Is there a sticky floor, a mid-level bottleneck, or a glass ceiling? Relations Industrielles/Industrial Relations (RI-IR), 64, 593–620. Zatzick, C. D., & Iverson, R. D. (2006). High-involvement management and workforce reduction: Competitive advantage or disadvantage? Academy of Management Journal, 49, 999–1015.

Copyright # 2010 John Wiley & Sons, Ltd.

J. Organiz. Behav. 32, 1062–1083 (2011) DOI: 10.1002/job