The Impact of Visualizing the Group on Children’s Persistence in and Perceptions of STEM

• In Event: Girls in STEM: Exploring the Developmental Origins of the Gender Gap in Science

Abstract

Women are underrepresented in STEM fields in the United States and across the world (NSF, 2019), though they achieve better grades in science than men throughout school (Voyer & Voyer, 2014). This presentation investigates the developmental origins of this gender disparity, with a focus on the perceptual mechanisms that help shape children’s stereotypes about what it takes to succeed as a scientist. Much of the research surrounding gender stereotypes in STEM has involved adults (e.g. Bian, Leslie, Murphy, & Cimpian, 2018; Cheryan et al., 2017), but young girls are also impacted by negative stereotypes about women in STEM (e.g. Cvencek et al., 2011). These negative stereotypes about girls in STEM negatively impact girls’ performance in and engagement with math and computer science (Master, Cheryan, Moscatelli, & Meltzoff, 2017; Huguet & Regner, 2007).

Previous research has studied how linguistic cues impact children’s preferences and performances in STEM (e.g. Rhodes et al., 2019; Bian, Leslie, & Cimpian, 2017). In two studies, we extended this work by exploring how visual information about the gender composition of a group of scientists impacts children’s persistence on a STEM task and their evaluations of group members. In Study 1, 129 4- and 5-year-old children viewed one of three groups of scientists: all-male, all-female, or a lone female among all-males. After viewing the groups, children were asked to play a science game in which they guessed whether objects would sink or float in water. The game was rigged such that all children failed on the first trial; we measured how many trials children persisted after this failure. We then asked children a forced choice question about the individual scientist in the group: did they think that the scientist was smart or hardworking?

Children’s persistence on the STEM task did not change across conditions (hazard ratios = 0.8, 1.02), but children judged the lone female scientist as “smart,” they judged scientists in the same gender conditions as “hardworking” (β = 1.976, SE = 0.489, p < 0.000). These findings begged the question: did children judge the lone female scientist as “smart” rather than “hardworking” because of her gender or because of an assessment of probability (i.e. because she was the exceptional character in the group)? To determine what drove children’s judgment, in Study 2, 37 children viewed a lone male scientist among all-females. The study followed the same procedure as in Study 1. Children’s persistence on the science task did not differ from their persistence in Study 1 (hazard ratios = 0.76, 1.04, 1.03, see Figure 1). Unlike in Study 1, in Study 2, children were equally as likely to judge the lone male scientist smart as hardworking (β = -1.074, SE = 0.478, p = 0.03, see Figure 2).

These findings suggest that children use both gender stereotypes and understanding of probability when making trait judgments about individual scientists in STEM. Taken together, Study 1 and Study 2 contribute to research on how stereotypes about who belongs in STEM impact children’s perceptions of others.

Authors

• Sona Kumar, Boston University
  Presenting Author

• Amanda S Haber, Fairfield University
  Non-Presenting Author

• Samantha Barbero
  Non-Presenting Author

• Kathleen H. Corriveau, Boston University
  Non-Presenting Author