Thursday, June 21, 2012

Feminine Role Models

This week's guest blogger is Kate Follette. Kate is a graduate student at Steward Observatory and an adjunct instructor at Pima Community College in Tucson, Arizona. Her scientific research focuses on planet formation in circumstellar disks, and she is also engaged in educational research on mitigating quantitative illiteracy through introductory science courses for non-majors. 

I’m sure I wasn’t the only one who was dismayed when I read the headline “My Fair Physicist? Feminine Math and Science Role Models Demotivate Young Girls”.

The article, which was posted to the WIA blog on April 16th and is linked here, was published in Social, Psychological and Personality Science. Its abstract reads:

Women in science, technology, engineering, and mathematics (STEM) are labeled unfeminine, a costly social label that may discourage female students from pursuing these fields. Challenges to this stereotype include feminine STEM role models, but their counterstereotypic-yet-feminine success may actually be demotivating, particularly to young girls.
• Study 1 showed that feminine STEM role models reduced middle school girls’ current math interest, self-rated ability, and success expectations relative to gender-neutral STEM role models and depressed future plans to study math among STEM-disidentified girls. These results did not extend to feminine role models displaying general (not STEM-specific) school success, indicating that feminine cues were not driving negative outcomes. 
• Study 2 suggested that feminine STEM role models’ combination of femininity and success seemed particularly unattainable to STEM-disidentified girls.
The results call for a better understanding of feminine STEM figures aimed at motivating young girls.

After reading the study myself, the bitter aftertaste of its primary conclusion - that “feminine” STEM role models demotivate girls who are STEM-disinclined - stuck with me for several months. I kept coming back to it and thinking “this CAN’T be true, can it??”

I do a fair amount of outreach with middle-school aged girls, and I’d like to consider myself a “feminine” STEM role model.  I don’t want to believe that my femininity is “demotivating”. Of course, just because I don’t want to believe it doesn’t mean that it’s not true.

The root of my personal discontent is that as a STEM role model, the message I received was that I should choose to be either discipline-independently feminine OR gender-neutral if I want to motivate young girls (or avoid demotivating them). I loathe this idea, because to consciously cultivate a “gender-neutral” appearance/demeanor or avoid specific mention of STEM success means not practicing what I preach. I want to be myself when I work with girls, and encourage them to do the same.

So before accepting their conclusion at face value, I suggest that we approach this paper with the same rigor afforded any other published scientific paper. Let’s  examine the data and experimental method and decide for ourselves whether the conclusion is warranted.  Here’s what I found when I did so:

1) Just 144 and 42 girls’ data were analyzed to draw conclusions for Studies 1 and 2 respectively.

2) Although statistics on the race distribution and grade level of participants are provided, no other demographic information is given. A few simple and potentially revealing questions might include how many/what type of schools were included, geographic (urban vs. rural) information, socioeconomic status, etc.

3) The crux of Study 1 was three interviews with university students, which the girls read and answered questions about. The setup is described as follows: “Participants then read magazine-type interviews with three female university students displaying feminine (e.g., wearing pink clothes and makeup, likes reading fashion magazines) or gender-neutral appearance and characteristics (e.g., wearing dark-colored clothes and glasses, likes reading).” Is this the definition of femininity?  Feminine women don’t read books or wear black?  This strikes me as almost comically narrow.

4) Since the students were only reacting to a small number of role model interviews (n=3) and rating them in general categories such as “positivity” and “perceived similarity”, it seems to me that conducting interviews with participants regarding WHY they chose certain rankings would be advisable. This could serve to reassure the reader that the girls are basing their rankings on the characteristics that the study designers claim – femininity and STEM success. In the educational literature this is called establishing “content validity” and involves answering the question “does your instrument measure what you think it does?” I’m not a social scientist, but I imagine that such a thing is (or should be) standard practice.

5) The second study used a similar set of interviews but asked two more direct questions
a. “How likely do you think it is that you could be both as successful in math/science AND as feminine or girly as these students by the end of high school”
b. “Do being good at math and being girly go together?”
The effect here was the same, but more marginal than in the first study (see Figure 3 of the paper) and had fewer participants (n=42), a less-standard setup (some girls participated in a classroom and some at a county fair) and a procedural error through which an (unspecified) number of girls didn’t receive item 2.

While this study is an interesting and thought-provoking result worthy of further investigation, I would have liked to see more of an effort on the part of the authors to emphasize the small and preliminary nature of the study.  Scientists of all persuasions need to be careful about how their work will be interpreted by non-experts, and this study reaches some particularly dangerous and counterproductive conclusions to be throwing around before they are fully supported by evidence. It is NOT the final word on the advantageousness of feminine STEM role models.