Wallflowers: Experimental Evidence of an Aversion to Standing Out

While public recognition for altruism inspires some people to donate more, women tend to choose contributions closer to the average, suggesting that they do not want positive or negative attention.


Instinctively, most people want to have a positive reputation. One way to achieve this is through pro-social behavior, like donating to a charity or contributing to the public good. Previous literature documents a positive relationship between visibility and pro-social behavior, as people like being recognized for their good deeds and charitable donations. However, public recognition for altruistic acts might have a more nuanced effect. The authors of this study argue that some people, especially women, experience a “wallflower” effect, in which individuals respond to visibility by acting in a way that minimizes potential scrutiny.  This includes giving donations that is likely to be seen as “average” when  charitable behavior is publicly acknowledged.

In this study, researchers asked participants to choose how much they would like to donate to a charity, and provided participants with scenarios about other people’s contributions. The authors test whether visibility affected contributions by placing study participants in groups and randomly assigning  groups to treatments where contributions are made privately or publicly (revealed  to the group). . To find if women were more likely to exhibit the “wallflower” effect, the authors evaluated the behavior of men and women in these two treatments.


Contributing visibly significantly increased the likelihood that a participant chose within-range group contributions, i.e. contributions that fell between his or her group members’ contributions.

  • When a person’s donations were visible to the group, their average contribution follows group members’ donations closely compared to when donations are not visible to the group. This “wallflower effect” increased participants’ contributions in scenarios where other group member’s contributions are high, and reduced contributions when the other group member’s contributions were low.
  • Women are more likely to be exhibit the “wallflower” effect”:   
    • Women in the visibility treatment were almost twice as likely to choose charitable contributions that would fall within the group range, relative to women in the private contribution group (38% compared to 21%).
    • Men’s in-range contributions did not significantly change in the visibility treatment (28% in private vs. 25% in visible). and in fact shifted slightly in the direction of matching the max.
  •  Reanalyzing data from existing experimental work where visibility is manipulated and gender data is collected (Gächter and Fehr 1999, Dufwenberg and Muren 2006, Andreoni and Bernheim 2009) and conducting a small real effort field experiments, the authors find similar patterns of gender differences in reputational concerns.

Overall, women were especially likely to choose contributions closer to others when their behavior was public. The authors posit that women may be more concerned than men about standing out or being judged as image motivated.


Participants were recruited through the Pittsburgh Experimental Economics Laboratory database. Each session included 15 participants seated at computer terminals and randomly assigned to groups of three and identified only by anonymous participant IDs. Five sessions of each of the two treatments were conducted, with 15 participants per session. The proportions of men and women in each session were roughly equal, with 75 men and 75 women participating overall.

Individuals were split between a baseline and a visibility treatment, where those in the visibility treatment were told that their identity and donations would be publicized to the rest of their group members. There were three phases to the experiment: (1) unconditional contributions, (2) conditional contributions and (3) belief elicitation.

In the unconditional contribution task, participants were asked to indicate how much of their $10 endowment they wanted to donate to a specific charity. The authors restricted contributions to multiples of $2. This restriction and the small group sizes were chosen to limit the number of choices that participants would face in the conditional contribution task.

In the conditional contribution task, participants were given the opportunity to change their contribution based on a series of scenarios about what the other two members of their group chose. Participants chose contributions conditional on every possible combination of their group members’ contributions from the unconditional phase—a total of 21 decisions.

In the belief-elicitation task, participants were asked to guess the number of people who chose each of the possible unconditional contributions to gauge whether they believed their donations were in line with their group members.

Related GAP Studies