Improving STEM Success in Elementary School Students Through Gender Integration

• In Event: Gender Gaps in STEM: Teacher Practices and Biases, Other-gender Peer Relationships, and Children’s Stereotypes

Abstract

STEM promotes critical thinking and problem-solving abilities; however, little is known about how to encourage students to become interested in STEM courses. Furthermore, girls lose interest in STEM earlier than boys resulting in a large gender gap in the U.S. In this study, we tested the hypothesis that when boys and girls work cooperatively and supportively in a classroom, they will have more positive STEM-related outcomes. We expected that students who felt gender-integrated (e.g., higher levels of other-gender interaction) would perform better in STEM tests of achievement than students who were relatively lower in gender integration (GI; Goal 1). Further, to explore how GI might relate to achievement, we examined both cognitive and affective processes as mediators of students’ STEM achievement (Goal 2). Given gender stereotypes about STEM performance, girls may be more likely to show positive STEM outcomes when GI is high. Thus, we also tested gender as a moderator (Goal 3).

Participants were 515 3rd, 4th, and 5th graders (51.1% boys; Mage = 9.08, SD = 1.00; 47.2% White, 14% African American) from three public elementary schools in the Southwestern U.S. In the fall (T1), student-reported GI was indicated by five composite scores: other-gender interaction frequency, other-gender interaction quality, other-gender expectancies of inclusion, other-gender expectancies of teasing, other-gender relationship efficacy. Other-gender measures were used following prior research on GI (Author citation). Teachers rated students’ school liking in winter (T2) using selected items from the School Liking and Avoidance Questionnaire (Ladd & Price, 1987). In the spring (T3), students reported their academic beliefs and values (Eccles et al., 1993). T3 academic achievement was formed by teacher-reported academic success and students’ standardized test scores in math.
Table 1 presents descriptive statistics among observed study variables. To examine research goals, we used Mplus with TYPE = COMPLEX to account for class-level variance. For all models, we first estimated the measurement model. We found that T1 GI was not related to T3 achievement (β = .10, S.E. = .06, p = .11). Bootstrapping analyses indicated that T2 school liking mediated the link between T1 GI and T3 achievement, 95% CI [.001, .123], and T3 academic beliefs also mediated the link between T1 GI and T3 achievement, 95% CI [.025, .214]. Multiple group analyses with gender as a moderator (See Figure 1) suggested the mediating effect for school liking was mainly driven by boys, 95% CI [.012, .205]. However, T3 academic beliefs mediated the link between T1 GI and T3 achievement for both girls, 95% CI [.031, .232] and boys, 95% CI [.019, .266].
This study highlights the importance of promoting GI early in students’ education as a way to benefit both girls’ and boys’ STEM beliefs, motivations, and achievement. Interestingly, GI was effective in promoting STEM interests for all students; further research is needed to continue explorations of how to best narrow gender gaps in STEM. We hope the findings will help researchers and educators to gain awareness of the utility of GI in the classroom in promoting students’ academic success.

Authors

• Sonya Xinyue Xiao, Northern Arizona Univeristy
  Presenting Author

• Carol Lynn Martin, Arizona State University

• Laura D Hanish, Arizona State University

• Dawn DeLay, Arizona State University

• Richard A Fabes, Arizona State University

• Krista Oswalt