Citation
Stout, J., Dasgupta, N., Hunsinger, M., & McManus, M. (2011). STEMing the tide: Using in-group experts to inoculate women’s self-concept in science, technology, engineering, and mathematics (STEM). Journal of Personality and Social Psychology, 100, 255-270. [one of three studies described in a single report]
Highlights
- The study’s objective was to examine the impact of having a professor and teaching assistant (TA) of the same gender on introductory calculus students’ attitudes toward math, compared with English.
- The study used computerized tasks to measure attitudes toward math and collected the expected course grade and actual course performance of female students with a female professor and TA pair (a female teaching pair), compared with those of female students with a male professor and TA pair (a male teaching pair) at both the start and end of the semester.
- The study found that female students with a female teaching pair identified more with math and expected a higher course grade than those with a male teaching pair. Female students with a male teaching pair had more negative attitudes toward math, compared with English, than female students with a female teaching pair.
- The quality of causal evidence presented in this report is low because the authors did not demonstrate comparability of treatment and control groups before the intervention. This means we are not confident that the estimated effects are attributable to having a female teaching pair; other factors are likely to have contributed.
Features of the Study
Without knowing the identity of the professor, 42 women and 49 men registered for one of 15 sections of an introductory calculus course based on the course timetable. To assign instructors to course sections, the authors first drew a sample of male and female calculus professors who were similar on teaching skills, career level, and English fluency to make the study conditions as comparable as possible on attributes other than the instructors’ gender. Next, professors were matched with teaching assistants of the same gender as the professors to create two treatment conditions: a female professor paired with a female TA, and a male professor paired with a male TA. Seven course sections were assigned to a female teaching pair and 8 sections to a male teaching pair. The professors were told they were participating in a study to measure students’ interest in math, though the specific treatment being tested was not revealed to them.
Outcomes were compared across four groups: (1) male students instructed by females, (2) male students instructed by males, (3) female students instructed by females, and (4) female students instructed by males. However, the focus of this review is on the comparison between female students with a female teaching pair and female students with a male teaching pair.
At the beginning of the semester, students completed computerized tasks to measure their implicit and explicit attitudes toward math and English and reported their expected course grades (self-efficacy in math). Students completed the same computerized tasks at the end of the semester, and reported their actual final grades for the course. The authors used statistical analysis to estimate differences in the outcomes across the various study conditions.
Findings
- Female students had similar implicit attitudes toward math and English when they had a female teaching pair, but more negative implicit attitudes toward math when they had a male teaching pair. The difference between the two groups was statistically significant.
- Female students with a female teaching pair identified more with math than those with a male teaching pair, and the difference was statistically significant.
Considerations for Interpreting the Findings
The authors estimated multiple related impacts on outcomes related to attitudes toward STEM. Performing multiple statistical tests on related outcomes makes it more likely that some impacts will be found statistically significant purely by chance and not because they reflect program effectiveness. The authors did not perform statistical adjustments to account for the multiple tests, so the number of statistically significant findings in these domains is likely to be overstated. However, CLEAR was able to adjust for these multiple comparisons using information presented in the report; the findings presented here are those that remained statistically significant after the adjustments were made.
Causal Evidence Rating
The quality of causal evidence presented in this report is low because the authors did not demonstrate comparability of treatment and control groups before the intervention. This means we are not confident that the estimated effects are attributable to having a female teaching pair; other factors are likely to have contributed.
Additional Sources
Stout, J., Dasgupta, N., Hunsinger, M., & McManus, M. (2010). STEMing the tide: Using in-group experts to inoculate women’s self-concept in science, technology, engineering, and mathematics (STEM). Journal of Personality and Social Psychology. Advance online publication.