This is the 13th in our series of posts by job market candidates.
Globally, men commit nearly all homicides and are considerably more likely than women to die from suicide and violence. Agreeing to traditional masculinity norms – the beliefs about what men should or not do (e.g. “men should not cry”, “it is acceptable for men to use violence”) – may, at least partially, explain these gender gaps as it predicts worse mental health outcomes and aggressive behaviors. But what if these norms are based on incorrect views of what others believe? Misperceiving others’ views may make people act against their own beliefs just to conform to the perceived norm. In the context of masculinity norms, overestimating others’ beliefs in traditional masculinity can perpetuate harmful behaviors. For example, men may avoid help-seeking behaviors or use violence if they incorrectly think that is what others expect they should do.
Hence, shedding light on the mechanisms that sustain undesirable norms in equilibrium might be important. What factors contribute to the persistence of misperceived social norms? A hypothesis is that people do not talk to their same-generation peers. Alternatively, people could talk only with a selected group, such as more vocal or closer friends.
To test these hypotheses, I conducted two field experiments (N = 2,608) that examine whether a lack of communication or biased communication contributes to the persistence of misperceived masculinity norms among 13 to 15-year-old boys and girls across 25 public schools in Rio de Janeiro. Organized crime dominates 18% of Rio’s territory, making masculinity norms particularly relevant in this context. Men may think their peers will see them as stronger and more admirable if they engage in criminal activities. In addition, as boys transition to manhood, they may feel compelled to "man up", suppressing their emotions and, in some cases, resorting to violence to earn respect.
Boys Overestimate Peers’ Support For Traditional Masculinity Norms
I measure masculinity norms along two dimensions: emotional restriction–which dictates that men should suppress vulnerable emotions–, and aggression–which dictates that it is acceptable, and sometimes encouraged, for a man to use violence. To measure these actual norms, I ask whether adolescents agree or disagree with the beliefs about masculinity “men who cry are weak” and “men should use violence to get respect if necessary”. To measure perceptions of norms, I ask them to guess the percentage of their male or female classmates, separately, that they think agree with each of these statements. I define misperceptions as the difference, in percentage points, between their perceptions of what their classmates think and the fraction of classmates who agree with each statement. In this post, I focus on the results for boys.
Boys largely overestimate their classmates’ views about masculinity. Only 10% of boys agree with “men who cry are weak”, but they guess that 31% of other boys in their class agree with the statement, resulting in an average misperception of 21 percentage points (pp) (Figure 1, Panel A). Boys overestimate girls’ beliefs about crying even more, by 28pp (Figure 1, Panel B). They also overestimate other boys’ and girls’ beliefs about violence are 12pp and 10pp, respectively. These misperceptions might reflect outdated views about masculinity transmitted from, for example, older family members. It might also reflect the views of a selected group of friends, such as those who are more vocal about masculinity. So encouraging communication with a broad set of peers might be important to reduce them.
Broad Communication With Peers Reduces Misperceived Views About Masculinity
In the first experiment (N=2,249), I test whether a lack of communication drives the persistence of misperceived norms. I randomly assigned boys and girls from the same classroom into a 15-minute discussion about masculinity or a control discussion about recycling. In the masculinity discussions, a mediator asked adolescents’ opinions on the statements about crying and violence. To further test for biased communication (i.e., communication with a selected set of peers), I randomized whether the masculinity discussions were of two types: Voluntary, in which participants raise their hand if they want to speak, or Randomized, in which I randomly selected the participants asked to speak. Observers took notes during these discussions, allowing me to compare their public views with those in the pre-discussion survey (i.e., private). The control discussions elicited adolescents’ views about recycling practices. I surveyed participants immediately after as well as three weeks later.
Masculinity discussions reduce boys’ misperceptions by about 50% immediately after the discussion, similarly across the Randomized and Voluntary groups (Figure 2). The Voluntary and Randomized arms yield similar treatment effects. Vocal students – those who speak in the Voluntary arm – are privately less likely to agree with the beliefs about masculinity than the representative speakers in the Randomized group. This suggests that even vocal types do not talk about masculinity at school; otherwise, participants in the Voluntary group would not have learned any new information. Nevertheless, the public views shared by Voluntary and Randomized speakers are similar. This happens because even though Randomized speakers are representative by construction, they are less likely to agree with the masculinity statements in public than in private.
Three weeks later, the effects of the discussion persist, suggesting that those in the masculinity discussions retained the information learned, and that the information did not spill over to their classmates in the recycling discussions. This is striking, considering students interact daily, so strong barriers to natural conversations about masculinity might exist.
Communication With Closer Peers And Lower Mediation Also Reduces Misperceptions
In the second experiment implemented a year later in different schools, but similar in terms of observable characteristics, (N=359), I test for the effects of biased communication with chosen peers by letting adolescents choose the classmates they will discuss masculinity with. This design allows for variation in the sex composition of the groups. 25% were boys-only, and 25% were girls-only, with an average of 5 peers. The discussion features were similar to those of the first experiment, except the mediator was not present all the time—they would just make sure that students were discussing masculinity. I surveyed students immediately before and immediately after the discussions. In addition, to understand whether barriers to discussing masculinity exist, I asked participants to indicate, before and after the discussions, their perceptions of comfort and interest. A hypothesis is that people tend to overestimate the awkwardness of deep conversations.
When in conversations with chosen peers, boys’ misperceptions about other boys’ views reduce, suggesting they do not discuss masculinity even with closer peers. Even when in boys-only group, boys’ misperceptions about other boys’ views do not increase, which is evidence that they do not engage in locker-room talk by expressing very masculine views in more private conversations. Nevertheless, boys’ misperceptions about girls’ views do not change, because when in boys-only group, their misperceptions about girls’ views increase.
Why, then, don’t adolescents engage in conversations about masculinity? They underestimate their interest in and comfort with discussing masculinity.
Importantly, all the results are robust to including a social desirability index in the specifications. In addition, the social desirability index does not significantly predict the levels of misperceptions among the control group.
What do we learn?
My findings contribute to recent calls in the public arena to listen to boys' narratives about masculinity to achieve gender equality. Interventions that create safe spaces for boys and girls to engage in open conversations about masculinity may shift collective social norms. Trained facilitators may not even be necessary, so policymakers should consider incorporating discussions about masculinity into school activities. Longer-term programs of a similar type are encouraged, as meaningful behavioral changes might take time. Policies that break the barriers to natural conversations about masculinity should also be considered.
Ieda Matavelli is a postdoctoral fellow at the University of New South Wales.
Join the Conversation