My Fair Physicist? Feminine Math and Science Role Models Demotivate Young Girls

In Study 1, middle-school girls were discouraged by role models who were both successful in STEM subjects and stereotypically feminine in their appearance and hobbies.

• Among STEM-disidentified girls (who did not report math or science among their three favorite subjects), feminine STEM role models decreased intent to take high school and college math classes compared to gender-neutral STEM role models (4.04 vs. 5.57 on a 7-point scale), with medium effect size.
• Among STEM-identified girls (who reported math and/or science among their three favorite subjects) there was no significant difference in future plans to take math classes when presented with feminine vs. gender-neutral STEM role models.
• Among all girls, there was no significant difference in future plans to take English classes (a field with no negative feminine stereotypes) when girls were presented with feminine or gender-neutral role models outside of STEM.
• Among all girls, feminine STEM role models decreased confidence and interest in math.
  • Feminine STEM role models decreased all girls’ ratings of their own interest, ability, and expectations of success in math, compared to gender-neutral STEM role models (4.50 vs 5.53 on a 7-point scale), with large effect size.
  • Effects were more pronounced among STEM-disidentified girls, whose math self-ratings decreased by over 1 point when presented with a feminine vs. gender-neutral STEM role model.
  • There was no significant difference in girls’ self-ratings when presented with feminine or gender-neutral role models outside of STEM.
• Both STEM-disidentified and identified girls were discouraged in math despite viewing all of the role models as equally positive and similar to themselves.

In Study 2, girls whose interest in STEM subjects was already low viewed the success of feminine STEM role models as particularly unattainable.

• STEM-disidentified girls perceived themselves as significantly less likely to match role models’ STEM success and femininity when presented with feminine STEM role models, compared to gender-neutral STEM role models (4.08 vs 5.46 on a 7-point scale), with moderate-to-large effect size.
• STEM-identified girls felt no significant difference in their likelihood of matching role models’ success when presented with feminine or gender-neutral STEM role models.
• STEM-disidentified girls rated STEM success and femininity (or “being girly”) as less compatible, compared to STEM-identified girls (3.5 vs 4.36 on a 7-point scale, marginally significant).

These findings show that attempts to feminize STEM role models can have unintended negative consequences, discouraging middle-school girls with images of STEM success they view as unattainable.

Study 1 included 144 girls in sixth and seventh grades. Participants were considered STEM-identified (54.2%) if they included math or science when reporting their three favorite school subjects. Participants read interviews with three role models, all women attending a local college. The role models were either feminine (pink clothing, make-up, and interest in fashion magazines) or gender-neutral (dark clothing, glasses, and interest in reading), and successful in either STEM fields or unspecified academic fields. Using 7-point scales, participants rated the role models’ positive image (4 items) and similarity to themselves (1 item); reported their own future plans including intent to take high school and college math classes; and rated their current interest, ability, and expectations of short-term success in math (9 items).

Study 2 included 42 girls in sixth and seventh grades (45.2% STEM-identified). Participants read the Study 1 interviews with STEM role models who were either feminine or gender-neutral. Participants rated the likelihood that they could achieve the same level of STEM success and femininity as the role models they read about (1 item, 7-point scale). They also answered the question, “Do being good at math and being girly go together?”