Diversity in Design Teams: A Grounded Theory Approach
Salas, Dr. Eduardo
Doctor of Philosophy
Recent workforce trends have virtually guaranteed that employees need to be able to work effectively with a diverse group of colleagues. First, US workforce demographics have changed dramatically; the population is now 16.4% Hispanic or Latino, 11.7% Black or African American, and 5.5% Asian. Women now make up 46.8% of the labor pool (BLS, 2015). A second trend is the widespread use of interdisciplinary teams to tackle cognitively demanding tasks, as well as to spur creativity and innovation (Cooke, Salas, Cannon-Bowers, & Stout, 2000; Fay, Borril, Amir, Haward, & West, 2006; Salas, Cooke, & Rosen, 2008). Taken together, it is clear that understanding diversity in teams is an issue that needs to be at the forefront of research and practice. In the science of teams, diversity has been labeled a double-edge sword (Chi, Huang, & Lin, 2009). Diverse members offer a wider range of expertise and ideas from which the team can draw (e.g., Watson, Kumar, & Michaelsen, 1993). According to information processing theories (e.g., van Knippenberg, De Dreu, & Homan, 2004), this leads to team behaviors (e.g., constructive debate) that result in higher quality team processes (e.g., decision making; Horwitz & Horwitz, 2007) and performance (e.g., creativity; McLeod, Lobel, & Cox, 1996). Alternatively, similarity-attraction (Byrne, 1971) and social categorization (Hogg & Turner, 1985) theories posit that diversity leads to subgrouping based on perceived similarity, and ultimately to bias, reduced social integration, and increased conflict that can act as barriers to realizing the team-level benefits of diversity (Stahl, et al., 2010). Indeed, the influence of diversity on team process and outcomes is anything but straightforward. In fact, empirical evidence has failed to find consistent relationships (e.g., Webber & Donahue, 2001), highlighting the importance of potential moderators (see Table 1 for a sample of these factors). Towards this end, a series of meta-analytic investigations have been conducted that emphasize the importance of team type, task difficulty, task type and inter-industry factors such as competition (Bell et al., 2011; Bowers, Pharmer, & Salas, 2000; Joshi & Roh, 2009). Clearly, context matters, and understanding the diversity-performance relationship requires a deeper dive into a specific context of interest. Qualitative inquiry is one tool through which this can be achieved. Toward this end, qualitative research allows for an inductive approach to the relationship between diversity and team performance under specific contexts, and recent efforts (e.g., Shachaf, 2008) have shown that while diversity affects similar team processes (e.g., communication), the way this effect unfolds uniquely differs by context. Therefore, in line with these efforts, the current investigation takes a structured and systematic grounded theory approach (Strauss & Corbin, 1990) to look closely at a context where creativity and innovation are demanded from today’s workforce, namely, design teams. In an attempt to shed light on persistent equivocal findings in both the team diversity and team conflict literatures, I put forth: (1) a temporally-based framework of diverse design team performance using the Input Mediator Output Input model (IMOI; Ilgen, et al., 2005), (2) an integrative theory, and a (3) set of testable findings. I argue that in engineering design teams, diversity on assertiveness, previous experience, and demographics (and the underlying cultural dimension of collectivism/individualism) can either positively or negatively influence communication behaviors (i.e., information exchange and elaboration; Van Knippenberg et al., 2004) through the avenues of frequency, timelines, equality, and comprehension of exchange. Furthermore, and in line with meta-analytic findings (Stahl et al., 2010), it is proposed that the nature of diverse teams and the engineering design process itself suggest the diversity-task conflict relationship will be strong. While previous research has largely looked at task and relationship conflict in silo (Behfar, et al., 2008), I suggest that these strongly correlated states (de Wit, Greer, & Jehn, 2012) are inextricable linked and often develop together. Furthermore, the underlying mechanism that is responsible for task conflict spiraling into relationship conflict is diversity in the directness of conflict expression (Weingart et. al, 2015), a variable associated with both assertiveness and collectivism (Oyserman & Kemmelmeier, 2002). Indeed, members varied greatly in their preference for straightforward, goal-directed expression versus ambiguous, relationship-centered expression. It is argued that heterogeneity in this construct among team members has the potential for expressions to be seen as rude, insincere, or argumentative by those less direct, or as avoidant and passive aggressive by those more direct (Taras et al., 2007), thus triggering the onset of relationship conflict. In a unique contribution to the literature, the theory then develops further to address many of the limitations cited in the conflict literature related to measurement, failure to consider reciprocal effects between conflict state, and temporal issues (Loughry & Amason, 2014). What follows is a unique look at how conflict-communication cycles unfold across a team’s lifecycle, and the team level states (e.g., social integration; O’Reilly, Caldwell and Barnett, 1989) that can act as a buffer against potentially negative effects on team performance.