What do Scientists Look like? Drawing Scientists in Kindergarten and Middle School

• In Event: Poster Session 03
  In Poster Session: PS 03 Topic - Sex, Gender

Abstract

What do children think scientists look like? The Draw-A-Scientist (DAST) task (Chambers, 1983) asks children to draw a scientist at work. A meta-analysis of five decades of DAST studies (Miller et al., 2018) reported that children aged 5-6 drew roughly the same numbers of males and females, but that began to change as children grew older. By middle school, children were drawing four times as many males as females. Chambers (1983) found that out of nearly five thousand drawings from children ages 5-11, only 28 were of women, and all of those were drawn by girls. Miller et al. (2018) found that across 29 samples, boys drew male scientists 96% of the time, and boys drew only male scientists in more than half of those samples.

We used the DAST to investigate the gender of scientist depictions using a cross-sectional sample. The two cohorts were comprised of 92 kindergarteners (F=54, M=38) 4.9 to 6.9 years of age (M=5.5, SD=4.28mo) and 51 middle schoolers (F=22, M=29) 11.3 to 14.3 years of age (M=12.65, SD=0.65). We also asked children to choose the “smart one” (Bian et al., 2017) or the “scientist” (Diaz et al., 2020) from gender-balanced photo arrays. Testing took place in small groups at the childrens’ school.

Our results were as follows: kindergarten boys drew males 74% of the time, females 5.5%, and “other” 21.1%. “Other” were non-human characters. Kindergarten girls drew females 62% of the time, males 19%, and “other” 19%. In middle school, boys drew males 93% of the time and “other” 7% (no females), while girls drew equal numbers of males and females. The proportion of “other” drawings decreased developmentally. Correlations revealed that kindergarten boys’ scientist drawings were related to their likelihood of choosing a male as the smart person (r(38)=.45, p=.005) or as the scientist (r(38)=.32, p=.044) from a set of pictures. For middle school boys, DAST was related to their likelihood of picking a male as the smart person (r(25)=.47, p=.018). Notably, DAST was not related to girls’ choices in kindergarten or in middle school.

Although cross-sectional in nature, our results demonstrate a developmental transition in girls suggesting a more equitable and representative view of scientists. In contrast, boys display a seemingly immutable perspective of representation of males as scientists. Differences in who they chose as the smart person and as the scientist suggest that for boys, these stereotypes are stable and also present in different domains. We propose that while interventions towards increasing female representation in STEM often focus on girls’ attitudes (Liben & Coyle, 2014), future work should focus on designing interventions for boys and tracking their attitudes throughout adolescence.

Authors

• Kelly Elizabeth Runyon, Virginia Polytechnic Institute and State University
  Presenting Author

• Vanessa Diaz, Virginia Polytechnic Institute and State University
  Non-Presenting Author

• Carolyn J. Kroehler, Virginia Tech
  Non-Presenting Author