When What You Know Is Not Enough

Is expertise, like beauty, in the eye of the beholder? Recent work by Professor Katherine Phillips (Kellogg School of Management) and Melissa Thomas-Hunt (Cornell University) explores the role of gender dynamics in the expression of expertise. In particular, Phillips and Thomas-Hunt find that gender expectations interfere with the expression, perception, and use of expertise, harming both the experts and the groups that depend upon them. This research explores the possibility that the possession of expertise lies in the eye of the beholder.

Previous work by Phillips (née Williams) and collaborators (for example, Gruenfeld, Mannix, Williams, and Neale, 1996) suggests that groups fail to take advantage of their members’ expertise because group discussions are social exchanges, not just information exchanges. When groups convene, their members seek social goals, such as acceptance, status, and power, not just informational goals. Under some conditions, these social goals prevent individuals with expertise from sharing it, and other group members from accepting it.

The ambiguity of real organizational tasks merely confounds the problem. For most such tasks, members cannot easily detect the “correct” solution. In the absence of objective solutions, judgments about others’ information often turn on social expectations. Even if group members never meet prior to their interaction, research demonstrates that they bring preconceived, status-based expectations about others’ competence on particular tasks to the group discussion (Eagly, 1987; Ridgeway, 1982). When a group member makes a statement that diverges from the rest of the group, others use their expectations to judge its relevance and veracity. If members have met prior to the interaction, their status-based expectations and social goals combine to influence judgments about others’ knowledge.

One status-relevant characteristic that consistently confounds judgments about others is gender. Research has demonstrated that women receive lower evaluations than men for similar performance (Biernat and Kobrynowicz, 1997). In addition, perceivers hold women to a higher standard when conferring competence (Shackelford, Wood, and Worchel, 1996). Importantly, most of the research behind these findings derives from traditionally male-oriented tasks, and the findings reverse somewhat for traditionally female-oriented tasks. Nevertheless, the traditional association of men with organizational tasks means that many such tasks are stereotypically male. Thus, women who dissent on organizational tasks should take a “double-hit” on status: one for expressing a divergent opinion, and another for diverging from gender-based expectations.

Drawing from this logic, Phillips and Thomas-Hunt predicted that both social goals and gender expectations interfere with the expression, perception, and incorporation of expertise. They hypothesized that groups, regardless of their gender composition, would judge women as less expert than men, and as a result women would exert less influence and report less confidence about their influence on the group. Phillips and Thomas-Hunt also predicted that when women are experts gender expectations would lead others to perceive them as less expert than even nonexpert women, causing them to exert less influence and report less confidence. In all these predictions, they anticipated the reverse for men. Taken together, these predictions imply a grim, group-level consequence: groups with female experts should actually perform worse than those with male experts. Inferior group performance would result from both gender and its implications for the actual influence of an individual on the group strategy.

To test these hypotheses, Phillips and Thomas-Hunt used a team development simulation (created by Human Synergistics, Inc.) in which participants must determine which survival items are most important for escaping from an Australian bushfire. This is a stereotypically male task, although, it is important to note, the actual performance of men and women on this task did not differ.

Prior to convening in groups, participants ranked the importance of various items (e.g., leather boots) individually. “Experts” were defined as those whose ranking most closely resembled the ranking of real survival experts. The researchers then assembled groups with lone female or male experts and asked the participants to develop a joint ranking of the items. After ranking the items, participants completed individual questionnaires about themselves and the other group members. The researchers measured the expert’s influence as the similarity between the expert’s individual strategy (as reflected in the first and last four items on their ranking) and the group’s collective strategy. They measured perceptions of others and self from the post-ranking questionnaire responses.

Results from this task largely confirmed predictions. Groups judged women as less expert than men, and women reported less influence and confidence than men. Groups judged women experts as less expert than their nonexpert female counterparts, and women experts reported less influence and confidence than those nonexpert women. Groups with women experts performed worse than those with male experts, and gender impacted group performance via individual influence on group strategy. There was one unanticipated finding: expert men did not report more confidence about their individual influence on the group’s strategy than nonexpert men. That is, all men were equally confident.

In light of these results, Phillips and Thomas-Hunt concluded that possessing expertise may actually become a liability for women. In general, social goals prevent anyone with expertise from sharing that information, and those social goals discourage other potential dissenters from supporting them. Group members evaluate team members who comply with gender-based roles more positively. However, gender expectations lead to stereotypes that make the situation worse for women. Specifically, group members use these attributions to dismiss women’s dissenting statements and make negative value judgments about them, whereas they accept divergent opinions from men more readily.

The authors argue that the dismissal of women’s expertise damages their confidence and leads them to self-censor. When it persists, self-censorship takes the form of a self-fulfilling prophecy, in which women experts are seen by others and themselves as less expert than they really are. Over time, moreover, chronic dismissal of women’s expertise damages their confidence and reinforces existing stereotypes about the relative expertise of women and men on particular tasks.

In sum, Phillips and Thomas-Hunt provide evidence that the actual expression, perception, and group harnessing of expertise is affected by gender expectations. Despite the actual distribution of expertise in groups, the value of that expertise for some groups may never be realized.

Further reading:

Biernat, M. and D. Kobrynowicz (1997). “Gender and race-based standards of competence: Lower minimum standards but higher ability standards for devalued groups.” Journal of Personality and Social Psychology, 72(3): 544-557.

Eagly, A.H. (1987). Sex differences in social behavior: A social-role analysis. Hillsdale, NJ: Lawrence Erlbaum.

Gruenfeld, D.H., E. Mannix, K.Y. Williams, and M.A. Neale (1996). “Group composition and decision-making: How member familiarity and information distribution affect process and performance.” Organizational Behavior and Human Decision Processes, 67(1): 1-15.

Ridgeway, C. L. (1982). “Status in groups: The importance of motivation.” American Sociological Review, 47: 76-88.

Shackelford, S., W. Wood, and S. Worchel (1996). “Behavioral styles and the influence of women in mixed-sex groups.” Social Psychology Quarterly, 59(3): 284-293.