Three Types of Workplace Diversity and Why They Matter

Stop Saying "Diversity"—Start Saying Which Kind

By Jonathan H. Westover, PhD

Every leadership team I've worked with over the past decade has wanted the same thing: high-performing teams that also reflect the organization's values around inclusion. And nearly everyone has made the same mistake—treating "diversity" as a single dial that you either turn up or turn down. The research is clear: that approach doesn't work. Not because diversity doesn't matter, but because the word hides at least three fundamentally different phenomena, each with different performance implications. Until leaders learn to distinguish between them, diversity initiatives will keep producing muddled results and frustrated stakeholders.

Three Kinds of Diversity (That Most Leaders Conflate)

The conceptual breakthrough came from Harrison and Klein (2007), who argued that "diversity" actually describes three distinct patterns of difference within a team:

• Variety — differences in kind. Team members bring different functional expertise, educational disciplines, or professional backgrounds. Think of a product team with an engineer, a designer, a marketer, and a data scientist. The value proposition is a wider knowledge base.
• Separation — differences in position. Team members disagree on values, attitudes, or perspectives, forming opposing camps. The risk is that separation fuels in-group/out-group dynamics, erodes trust, and stalls decision-making.
• Disparity — differences in concentration of resources. One or two team members hold disproportionate status, power, or tenure. The danger is that disparity silences lower-status voices and suppresses the very contributions diversity was supposed to unlock.

In my consulting work, I've found that most organizations default to measuring variety (how many different demographic categories are represented) while worrying about separation (interpersonal friction) and ignoring disparity entirely. That disconnect is a root cause of diversity fatigue.

What the Evidence Actually Shows

Bell, Villado, Lukasik, Belau, and Briggs (2011) conducted the most granular meta-analysis to date on demographic diversity and team performance, covering 92 studies and thousands of teams. Their central finding should be required reading for every CHRO and team leader: the specific diversity variable matters more than whether diversity is present or absent.

Here's what they found:

• Functional background variety had a consistent, positive relationship with team performance (ρ = .11), rising to ρ = .18 when the outcome was creativity or innovation, and ρ = .16 for cross-functional and design teams. In plain language: staffing a project team with people from different departments works, especially when the goal is to create something new.
• Educational background variety was unrelated to overall performance but positively related to innovation (ρ = .23) and to top management team effectiveness (ρ = .13). Educational diversity pays off, but only in specific contexts.
• Race and sex variety showed small negative effects in field settings (ρ = –.13 and ρ = –.09, respectively). These effects disappeared in laboratory settings, suggesting the issue isn't demographic difference itself—it's the social dynamics that organizations allow to persist around those differences.
• Age diversity was effectively zero across every conceptualization and setting.
• Mean organizational tenure outperformed tenure diversity as a predictor of efficiency-oriented outcomes (ρ = .14). Sometimes what teams need isn't more difference—it's more collective experience.

I've seen these patterns play out in real organizations again and again. A healthcare system I advised assembled a clinical quality improvement team by pulling one person from every department to "check the diversity box." The team had variety on paper, but the members had been selected for representation, not for the relevance of their expertise to the problem at hand. Performance was mediocre. When the team was reconstituted around functional variety aligned to the specific improvement challenge—a nurse, a pharmacist, a process engineer, and a data analyst—outcomes improved dramatically. The lesson: not all variety is created equal.

Four Things Practitioners Should Do Differently

Based on both the research and my professional experience, here are four shifts that make a meaningful difference.

1. Match diversity type to team purpose.

Before composing a team, define the performance criterion. If the goal is innovation—developing a new product, reimagining a service model, solving an ambiguous strategic challenge—prioritize functional and educational variety. If the goal is execution and efficiency—manufacturing, compliance, process optimization—prioritize collective experience (i.e., high mean tenure) and coordination. This distinction, supported by Bell et al.'s (2011) moderator analyses, is one of the highest-leverage staffing decisions a leader can make.

2. Invest in managing the diversity you already have.

The negative effects of race and sex diversity on team performance are not inevitable. They emerge when social categorization processes go unmanaged—when bias, status hierarchies, and in-group favoritism operate unchecked (van Knippenberg, De Dreu, & Homan, 2004). Psychological safety—the shared belief that the team is a safe place to take interpersonal risks—is the critical intervention. Edmondson (2019) has shown that leaders who model vulnerability, invite input, and respond constructively to dissent create conditions where diverse teams outperform homogeneous ones.

In practice, this means that the single highest-ROI diversity investment is not a new hiring initiative; it's equipping team leaders with the skills to facilitate inclusion after diverse people are in the room. I've seen organizations spend millions recruiting diverse talent only to watch that talent leave within 18 months because the team environment never changed.

3. Stop aggregating diversity into a single metric.

If your diversity dashboard reports one number—a composite index or a single percentage—you're measuring the wrong thing. Track functional diversity and demographic diversity separately. Distinguish between teams designed for innovation and teams designed for execution. Match your metric to your theory of change. As Harrison and Klein (2007) warned, mismatching your measurement to your theoretical mechanism produces misleading null results—and misleading null results produce abandoned initiatives.

4. Talk openly about what diversity is for on each team.

The research suggests that teams perform better when members understand why diversity matters for their specific task (Homan, van Knippenberg, Van Kleef, & De Dreu, 2007). In my experience, this means leaders need to have explicit conversations at team launch: "We've brought together people from different functions because this problem requires perspectives we can't get from a single discipline." That framing reframes diversity from a compliance exercise into a performance strategy—and it gives team members a reason to seek out and value each other's contributions.

The Bottom Line

Diversity is not a monolith, and treating it as one is the single biggest reason organizations are disappointed with their diversity-performance results. Functional variety drives innovation. Surface-level differences create social challenges that demand active management. Collective experience fuels efficiency. These are different phenomena requiring different strategies.

The organizations that will win on diversity are the ones that stop asking "Are we diverse enough?" and start asking "Do we have the right kinds of difference, configured the right way, for the outcomes we're trying to achieve—and are we leading those teams in ways that let every member contribute?"

That's a harder question. But it's the right one.

 

References

Bell, S. T., Villado, A. J., Lukasik, M. A., Belau, L., & Briggs, A. L. (2011). Getting specific about demographic diversity variable and team performance relationships: A meta-analysis. Journal of Management, 37(3), 709–743.

Edmondson, A. C. (2019). The fearless organization: Creating psychological safety in the workplace for learning, innovation, and growth. Wiley.

Harrison, D. A., & Klein, K. J. (2007). What's the difference? Diversity constructs as separation, variety, or disparity in organizations. Academy of Management Review, 32(4), 1199–1228.

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(5), 1189–1199.

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(6), 1008–1022.