Gender biases and the role of expectations: An experimental study in a Mexican police institution

Mexico ranks 81 out of 144 countries in the World Economic Forum’s Gender Gap Index 2017 and it ranks 20 out of 24 in the Latin American and the Caribbean Region (WEF, 2017). According to public information, in the main... [ view full abstract ]

Mexico ranks 81 out of 144 countries in the World Economic Forum’s Gender Gap Index 2017 and it ranks 20 out of 24 in the Latin American and the Caribbean Region (WEF, 2017). According to public information, in the main public agencies of the Mexican Federal Government, the percentage of women participation is lower than that of men (39% and 65%, respectively) (LNPP, 2017). Likewise, as the hierarchical level increases the participation of women goes down. In addition, the difference between the wages of women and men is negative and statistically significant, implying that, on average, men earn more than women (the difference is of $319 USD). One possible explanation for these differences is the existence of gender biases in the selection processes and promotion in public agencies. We propose that biases related to suitability and expectation of the performance of candidates negatively affect women. The purpose of this study is to explore the existence of gender biases in the evaluation of hypothetical profiles of candidates to a police corporation in Mexico and to see the interaction of these biases with the socioeconomic level of candidates.

The research question of this study is: how does information about gender and socioeconomic level of candidates to a police corporation influences the evaluation and expectations of them given by current police officers? To answer this question, we designed an experiment conducted in February 2018. In the experiment, Mexican police officers read hypothetical profiles of people interested in entering to the corporation. They were asked about their assessment of the candidate and their expectations about them if they enter the organization. These hypothetical profiles differed only in the gender and socioeconomic level of the candidate. Thus, the experiment design measures if gender and socioeconomic level affect the evaluation and expectations of candidates. We did not found significant differences on expectations in terms of: income, permanence, and performance.