The STEM disciplines have been shown to recruit and, perhaps more importantly, retain less female students to their higher education programs. A great deal of work has been devoted to an explanation of this observation but efforts aimed to study the potential positive effects of an enlarged female participation are less prominent. Sullivan et al. (2018) exploit the existence of more balanced gender ratios in the field of biology to examine these effects.
The authors performed an interesting experiment over three sections of an introductory biology course for non-biology majors. Their design involved a manipulation of the gender ratios of the small groups (tables) to which the students attending the lectures were assigned over the entire course. The final grades of the students, their peer- and self-evaluations together with their gender were recorded and linked to the gender ratios of their corresponding table. Two major effects were observed. First, the performance scores were found to be higher in those groups with a larger female presence and no significant gender difference was observed in the extent of this improvement. Second, self- and peer-evaluations presented an intricate but rather interesting pattern. Although all students delivered better evaluations for themselves than their evaluated partners this seemed to be less so for female students. Also, the more critical evaluation of their peers exerted by women was attenuated when the assigned group displayed a larger female share. Finally, an examination of the potential impact in social belonging delivered no significant results.
Comment: The targeted students of this study are of a very special kind which may raise concerns when it comes to its reproducibility. Indeed, a group of non-biology majors where only 5% of them were STEM majors may not give a picture with enough external validity for the research question. In all fairness, this is a point conceded by the very same authors.
A more general view over the purpose of this paper and the associated related research is in order. Studies like these are usually motivated by using slightly different versions of two major arguments: imperatives about the ethical necessity to introduce diversity in the classroom and potential positive impacts of diversity in student performance. The former should not be invoked when the latter is examined. If so, an immediate question may appear in our mind. What if we find negative effects of diversity? Does that qualify any ethical consideration? If the answer is no, the motivation of the research question may be compromised. Why study this in first place? Statistical discrimination may be magnified among those universities aspiring to achieve academic excellence. If the answer is yes, the intellectual honesty of the authors may be compromised instead.
Finally, the authors also bring up the present increasing demand for trained workers with STEM degrees in our societies as a reason to attract more women into this field. But in such case, a more effective tool would be to study the psychological traits and mechanisms that scare away any individual from such fields. Reducing everything to one single dimension (gender) with which more complex personal characteristics may correlate is not the most efficient way to proceed.
Text: Iñaki Rodríguez Longarela
