Building a Better Jury

I've been reading this fascinating study by Samuel R. Sommers of Tufts University that was published last year in the Journal of Personality and Social Psychology. I mentioned the study on the blog when it came out, but I didn't get the opportunity to read it until now. Basically, it measures the performance difference between racially homogeneous and heterogeneous juries (the former being all White, the latter consisting of four Whites and two Blacks), all of whom separately deliberated on the same "case". On a variety of metrics, blind observers found that members of the heterogeneous juries outperformed their homogeneous peers. Specifically, the heterogeneous juries deliberated longer, discussed more facts, were less likely to state factual inaccuracies, were more likely to correct factual inaccuracies when they did occur, discussed more "missing" evidence (e.g., inquire as to why certain testimony wasn't presented or why there was no fingerprint evidence, and what it meant), were more likely to discuss racism-related issues, and were less likely to reflexively object when race issues were brought up. These benefits were shared by the White and Black jurors alike. That is to say, the increased deliberation wasn't because Black jurors wished to press deliberations for a longer period, pushing the average up. Rather, in heterogeneous environments, both White and Black jurors both displayed a stronger commitment to a deeper, more rigorous examination of the facts and issues surrounding the case (the specific methodology and mechanics of the study are laid out in the article).

Another interesting implication is how these findings relate to other studies which have elucidated some of the harms of diversity. These harms generally play out in terms of reduced group cohesion and morale, or increased intra-group conflict. While Sommers did not find any difference between the diverse and homogeneous juries in the perception of intra-group conflict, he argues that even where that is a significant risk, in some circumstances it might be outweighed.

But even when conflict accompanies the potential benefits of diversity, one wonders whether this is often a risk worth taking. Threats to morale can be temporary and overcome as a group acclimates to heterogeneity (Jehn et al., 1999; Watson et al., 1993). Furthermore, a little discomfort may be good if, as the present data suggest, groups’ natural tendency is to stifle discussion of controversial or unpopular topics. Many a group has goals beyond a harmonious existence, whether the decision making of committees or the performance of students in a classroom. The present findings raise the possibility that dwelling on the negative interpersonal effects of racial diversity can be shortsighted and may prevent realization of long-term performance benefits. This leads to the more general conclusion that too little attention is often paid to the threat posed by group homogeneity. Debate regarding diversity usually centers on the costs and benefits of seeking heterogeneity, but what about the alternative status quo? An extreme interpretation of the present data is that compared with racially diverse groups, homogeneous groups were lazy information processors, prone to inaccuracies, unwilling to consider uncomfortable topics, and superficial in their discussions. A kinder conclusion would be that homogeneous groups spent less time on their decisions, made more errors, and considered fewer perspectives. In either case, homogeneity was associated with performance decrements, and this is not the first time such a relationship has been noted (Janis, 1982; Kameda & Sugimori, 1993; Wilkenfeld, 2004). Nonetheless, in both popular discourse and scientific examination, cost-benefit analyses of homogeneity are too often left implied or ignored altogether in efforts to evaluate diversity. (608-609)

Where maximum performance is considered to be more important than maximum harmony, diversity should be pursued even where there are risks of fostering some reduction in intra-group morale. Juries would seem to be an obvious instance where this is the case, and I'd argue that this is also so in many of our social and democratic institutions.

I've long argued that diversity (including racial diversity) is a just end for governmental and social bodies to pursue because diversity makes institutions better at what they do. Sommers' study has laid some of the empirical groundwork that indicates that this is, in fact the case. Where racial diversity can be tied to a bona fide job requirement, even conservatives have conceded that affirmative action to pursue it is constitutional and appropriate (see Wittmer v. Peters, 87 F.3d 916, 920 (7th Cir. 1996) (opinion by Judge Richard Posner)). Sommers' study indicates the possibility that these situations exist in far more circumstances than many of us had imagined.

***

Samuel R. Sommers, "On Racial Diversity and Group Decision Making: Identifying Multiple Effects of Racial Composition on Jury Deliberations," Journal of Personality and Social Psychology, Vol. 90, No. 4, pp. 597–612 (2006)