Tech companies, banks, consulting firms, you name it — all are scrambling to create diverse and inclusive environments. But despite pouring millions of dollars annually into diversity efforts, organizations sometimes fail to capture the benefits that diverse groups reportedly offer.
One possibility for this failure is that the purported benefits of diversity are more hype than reality, but that’s unlikely given the ample research that speaks against this claim. Racially diverse groups of jurors exchange a wider range of information during deliberations than racially homogeneous groups, for example. Diverse groups of traders are less likely to make inaccurate judgments when trading stocks. Gender diversity in top management teams improves firm performance, especially when innovation is a strategic focus. And our own past research helped establish the fact that the mere presence of diversity can lead groups to work harder, share unique perspectives, be more open to new ideas, and perform better, especially when groups need to share information and resolve differences of opinion.
So why the disconnect between the potential of diverse groups and reality? Part of the problem may be the fact that people’s biases about diverse groups, both conscious and unconscious, can undermine the very benefits of diversity.
To examine this hypothesis, we conducted a series of experiments in which participants made judgments about the level of conflict in a group’s interactions. They also rated their willingness to provide that group with the resources it requested. With this approach we were able to hold constant the content of the interaction — every participant either read the same exact transcript, watched the same video interaction, or listened to the same discussion among a group of four members. The only thing that changed was the racial composition of the individuals in the group. Homogeneous groups were either all-white or all-black, and diverse groups had two whites and two blacks in the group.
The findings were striking. When reading a transcript with pictures revealing the group’s composition, racially diverse teams were perceived as having more relationship conflict than homogeneous ones. And they were less likely to receive additional resources because of these biased perceptions of conflict — even though the objective content of the group interaction was exactly the same.
We tried it again — maybe it was something about the written transcript that left some things to the imagination of the participant. This time we hired actors to create videotaped discussions for our participants to watch. We were very careful to ensure that content, tone, and behavior of the actors was the same across the videos. Again, the only thing we changed was the composition of the group — it was either diverse or homogeneous — and we saw the same pattern emerge. Diverse groups were perceived as having more relationship conflict, and because of this, financial resources were less likely to be given to them than to homogeneous groups. The diverse groups were handicapped, potentially derailing future success.
In our final study we used audiotaped discussions and photos of the group members to appear racially diverse or homogenous and found the same pattern of results. Importantly we learned from this last experiment that it is only when groups are experiencing moderate and somewhat ambiguous levels of conflict that there is a clear bias against diverse teams. When the conflict was very clear and high, homogeneous and diverse teams were equally less likely to receive funding.
There are two important observations that make these findings compelling — it wasn’t just that the presence of more black members made people think the groups had more relationship conflict. Groups of all blacks were rated the same as groups of all whites. It was only when there was a diverse racial composition that this biased assessment emerged. Second, participants in our studies also assessed how much task conflict existed in the groups and here no differences were found. It was only the assessment of relationship conflict that varied and undermined the willingness to support the diverse teams.
So what can organizations do to combat this bias against diverse groups? At a basic level, an important first step is to cultivate an awareness of this bias in those responsible for evaluating diverse teams. Relationship conflict, in fact, is not necessarily a sign that things are going completely wrong or can’t be resolved. It may be the result of differing information, perspectives, and worldviews being worked through to allow innovation, better problem solving, and accurate decisions to emerge. Remember that your assessment of the severity of the conflict might be lower if it was happening in an equivalent homogeneous group.
Second, managers should rely upon clear standards of performance set before — not during — group observation instead of making performance and resource determinations in the middle of the process. Another way to measure performance could be having people who were not able to observe the group process be involved in the evaluation of the outcomes. These evaluators should also be blind to the composition of the team if possible. This should help reduce bias against diverse teams the same way blind auditions in orchestras helped eliminate bias against female musicians. By disconnecting the process and composition from the actual performance of the groups, the benefits of diversity might be seen more clearly.
Finally, a little advice for the diverse teams themselves: You have to play offense and ensure that managers see and value when things are going smoothly on the team. Celebrate your outcomes, even when they come from what may look like a messy process to others. And make sure those evaluating you know as much about when things are going well as they do about when things seemingly aren’t. Without counter information, managers’ biases may stunt the progress of diverse teams, and unwittingly undermine the opportunity for the benefits of diversity to emerge in the organization.
Katherine W. Phillips is the Paul Calello Professor of Leadership and Ethics and the senior vice dean at the Columbia Business School. Her research focuses on the areas of diversity, stereotyping, status, identity management, information sharing, minority influence, decision making, and performance in work groups.
Robert B. Lount, Jr. is an associate professor of management at the Fisher College of Business at the Ohio State University. He received his Ph.D. from the Kellogg School of Management at Northwestern University. His research focuses on how group composition and social status shape decision making, motivation, trust, and performance.
Oliver Sheldon is an assistant professor of Management and Global Business at Rutgers Business School. His research investigates triggers of interpersonal competition and conflict within small groups and teams, with the aim of shedding light on how organizations might improve coordination and collaboration among employees.
Floor Rink is a professor in Organizational Behavior at the Faculty of Economics and Business from the University of Groningen in the Netherlands. She examines how people respond to diversity and change within organizations and how people’s decisions are influenced by organizational norms and regulations.
IMAGE CREDITS: http://s3.amazonaws.com/