Mentor 1

Rob Deaner, Ph.D.


The purpose of this study is to explore the relationship between childhood play behavior and adult sports interest. More specifically, we examine whether gender-typed childhood behavior can be used to predict the type of sports adults choose to participate in. This study is part of a larger study that seeks to establish a biological component of gender differences in sports participation and interest that can add to the current theories of socialization. This work could shape how we frame certain U.S. educational policies such as Title IX, and have an impact on how we approach certain societal goals such as reducing the prevalence of obesity through increased physical activity. Despite the focus on socialization as the sole cause of observed gender differences in behavior, there is reason to believe that biology contributes to these gender differences as well. For example, previous research by Campbell and Eaton (1999) and Alexander and Hines (2002) has established a link between androgens and childhood behavior by showing very early gender differences in behavior in humans and distinct gender differences in play behavior among non-human primates. Furthermore, Berenbaum and Beltz (2011) found that females with congenital adrenal hyperplasia (CAH) – who, for various reasons were exposed to abnormally high levels of androgens in the womb – show more male-typical play behavior and preferences than females without the condition.

In order to explore the link between childhood behavior and adult sports interest, we used a correlational design. Using SurveyMonkey, we developed a survey to send to collegiate athletes from various schools across the country. The survey included Zucker’s 23 item Recalled Childhood Gender Identity/ Gender Role Questionnaire and questions regarding the athlete’s current primary sport, which sport they would play if given the opportunity and resources, and in which sport they would like to be considered the best in the world. In total, we had 307 participants from Division I and Division III schools (200 females, 107 males). We classified the different sports as feminine, neutral, or masculine based on previous cross-cultural research of sports stereotyping (Lauriola et. al., 2004; Riemer et. al., 2003). Deriving from the premise that biology matters, we suspected that more male-typical childhood play behavior would correlate with a preference for more masculine sports interests and more female-typical childhood play behavior would correlate with a preference for more feminine sports interests.

We found that women who recalled feminine childhood play behavior were significantly more likely to participate and indicate interest in feminine sports. We also found that men who recalled masculine childhood play behavior were significantly more likely to report interest in masculine sports. It is likely that we may have found stronger statistical significance for the male results if the response rate had been higher. Overall, however, our hypothesis was supported. Therefore, we have evidence consistent with the hypothesis that biology is at least partly responsible for the observed gender differences. This is not to rule out the role of socialization, but merely to provide a more accurate representation of the issue.