"The Emergence of Gender Gaps in Math Learning during a Single High-Quality Instructional Opportunity"
American Educational Research Association (AERA) Annual Meeting
Theme: Leveraging Educational Research in a “Post-Truth” Era: Multimodal Narratives to Democratize Evidence
April 5-9, 2019
Title: The Emergence of Gender Gaps in Math Learning during a Single High-Quality Instructional Opportunity (Paper)
Session: The Role of Gender in Shaping Mathematics Trajectories
Authors: Emily McLaughlin, Almaz Mesghina, Lindsey Richland
Abstract: Multiple previous studies unexpectedly found large and consistent gender gaps showing girls learning more from a mathematics lesson. The two studies presented here aim to clarify how individual differences across students and learning contexts may contribute to this gender gap. Study 1 tested the gender gap with a controlled videotaped, interactive math lesson and examined contributions of student’s prior knowledge, cognitive resources and math anxiety. Study 2 experimentally manipulated increased pressure to test impacts on the gender gap.
Measures of achievement requiring sustained effort in low-stakes settings (e.g. investing time in homework, Gershenson & Holt, 2015; persisting to graduation, Stetser & Stillwell, 2014) often favor girls, while the gender gap disappears or reverses on high-stakes measures of math achievement such as standardized tests (Reardon et al., 2018). Gender gaps may be shaped by both individual differences, such as math anxiety (Hembree, 1990; Devine et al., 2012), and by contextual factors, such as whether learning is pressured. Research suggests a larger performance boost in response to incentives (Levitt et al., 2016) and high-stakes testing contexts (Attali, Neeman & Schlosser, 2011) among males.
We implemented a pretest, lesson-and-immediate-posttest, delayed-posttest design, assessing both immediate learning and retention. Students interacted with a high-quality video-taped math lesson in their regular math classrooms - maximizing ecological validity, while allowing for controlled stimuli. We also collected measures of math anxiety and cognitive resources.
For Study 1, 346 diverse 5th grade students (180 girls) completed study procedures in a non-pressured classroom context.
For Study 2, 178 diverse 5th grade students (89 girls) completed study procedures. Half the students were randomly assigned (within classroom) to receive a pressure manipulation before instruction. They were told their performance on a test would determine whether their whole class would receive a desired incentive.
In Study 1, girls exhibited greater learning gains, immediately following the lesson and at a delay (both p s < 0.01). Prior knowledge and cognitive resources both predicted gains, but did not explain gender differences. Math anxiety negatively predicted gains, but also did not explain gender differences. Indeed, observed relationships between gender, math anxiety and learning are seemingly paradoxical- with girls exhibiting both higher math anxiety and larger gains- despite math anxiety negatively predicting gains in the sample as a whole. These relationships will be explored more fully in the paper. In study 2, the no-pressure condition replicated Study 1, with girls learning more than boys immediately and at a delay (both p s < 0.05). With pressure, however, the gender gaps disappeared, with pressure predicting smaller gains among girls and larger gains among boys, findings that will be explored more completely in the paper.
Although gendered patterns of achievement are well documented, less research has considered the role of gender in shaping initial learning. Through examining the in-the-moment emergence of gender differences in learning during a single high-quality instructional opportunity in pressured and non-pressured contexts, these studies aim to elucidate factors shaping gendered patterns of math trajectories.