Science Faculty’s Subtle Gender Biases Favor Male Students

Science professors of both genders exhibited an unconscious bias against female students, perceiving them to be less competent than male students.

Introduction

In the last three decades, women have made significant strides in matching, and in some areas, surpassing men in earning educational degrees. More than half of all bachelor’s and master’s degrees earned in the United States now belong to women, with 60% of all non-science PhD’s being attained by females.  Despite these encouraging statistics, women remain largely underrepresented in science, technology, engineering, and math (STEM). Although the proportion of science degrees granted to women has increased, less than half of all doctoral candidates in STEM fields are female, and there is a persistent disparity between the number of women receiving PhDs and those hired as faculty. Explanations for this gender gap often focus on women’s preference and lifestyle factors that steer them away from pursuing a STEM degree. An alternative explanation, however, is that there may be a subconscious bias on the part of existing faculty that in turn leads to a lack of mentoring, professional development, and opportunity for women considering STEM careers. In this study, researchers investigate whether an innate and unintentional gender bias on the part of employers and mentors impedes women from achieving professional placements in the science departments of universities.

Findings

The gender of the applicant had a statistically significant effect on the reviewer’s evaluation of their competence, employability, salary grant, and the mentorship offered.

  • Science professors evaluated female student applicants as less competent and less hireable than male students with identical application materials. On a scale of 1 to 7, applicants with male names were given a score of 4.01 and 4.1 by male and female professors respectively with regard to competence, while female students were given a score of 3.33 and 3.32 by male and female professors respectively. With regard to hireability, male students were given an average score of 3.74 and 3.92 by male and female professors respectively, while female students were given a mean score of 2.96 and 2.84 by male and female professors respectively.
  • Professors offered female students a starting yearly salary of approximately $26,500, whereas male applicants were offered a salary of about $30,200. On average, female faculty offered female students the lowest salary ($25,000) while male faculty offered male students the highest salary ($30,520.83).
  • Female faculty did not rate female students as more competent or hireable than male faculty, nor did they offer higher salary or more mentoring to female students than their male counterparts.
  • Though professors of both genders reported liking the female applicant better, her likability had no bearing on the perception of her competence or hireability.

In short, science faculty demonstrated subtle biases against female science students that may contribute to the stark underrepresentation of women in science, particularly in the higher levels of academia.

Methodology

In this randomized study, 127 biology, chemistry and physics professors evaluated undergraduate applicants for a laboratory manager position. The professors were recruited from six anonymous American universities (3 public and 3 private), all of which were ranked by the Carnegie Foundation as “large, Research University (very high research productivity).” The schools were matched for size and prestige, and were selected from three different geographic regions within the United States. Though application components were exactly the same, half of the professors (63) were randomly assigned a male name, while the other half (64) were randomly assigned a female name. Competence, hireability, salary, and mentoring were evaluated using a two by two ANOVA. To test for factors that may play a role in the professors’ gender bias, the Modern Sexism Scale was applied and multiple regression analyses were conducted. Additionally, mediation analysis was used to determine the processes through which the gender bias occurred.

Related GAP Studies