Teacher Perceptions and Responses to Student Ideas in Mathematics: An Experimental Study of Racial Biases

• In Event: Examining the Role of Racial-Ethnic Biases in Teacher Perceptions and Decision Making

Abstract

Racial-ethnic disparities in student learning outcomes are persistent features of mathematics education. Students who are Black, Indigenous, and People of Color (BIPOC) have historically performed lower than their white counterparts in mathematics, contributing to magnified disparities in later educational outcomes, engagement in STEM fields, lifetime earnings, and overall access to opportunity in American society (Gutiérrez, 2012). While differences in math achievement across student racial-ethnic groups are well established, the specific instructional mechanisms by which these patterns occur are not well understood (Warikoo et al., 2016). Further, while a focus on differences in achievement illuminates important educational disparities, it inadvertently locates the source of such differences on the child rather than on the mechanisms that contribute to them (Flores, 2007; Ladson-Billings, 2006; Martin, 2009; Milner, 2010). In the present study, we thus shift the focus to teachers’ fine-grain instructional practices that may be contributing to disparities in instructional rigor and subsequent opportunities to learn in math.

Taking the lens of implicit social cognition theory (Greenwald & Banaji, 1995), we build on prior studies that describe racial-ethnic biases in teachers’ academic expectations (van den Bergh et al., 2010) and perceptions of student mathematical ability from partially correct responses (Author et al., 2019). We use an experimental design to investigate the extent to which teacher responses to novel student ideas in mathematics are subject to racial-ethnic biases. Using a three-factorial, between-subjects design, we randomly assigned teachers (N = 271) to a simulated classroom environment in which they responded to a hypothetical Black, Latinx/e, or white student. In each condition, teachers responded to six different instructional scenarios containing 4th grade mathematics tasks ranging from fraction to addition operations. For example, in one vignette, a student presented their approach for 12x5 using a non-standard approach taking the sum of the partial products. Following the student’s explanation, teachers recorded what they would say and do next. Their responses were coded along four dimensions: elicitation and use of student thinking, entertainment of student sensemaking, and teacher affirmation. We also asked teachers to rate their perceptions of the student response (e.g. difficulty for the student, sophistication) using seven Likert-type items.

We found that teachers reported a higher perceived difficulty of the math tasks for Black and Latinx/e students relative to white students. We did not detect differences by condition for the other dimensions of teacher perceptions and responses to the student. Our results demonstrate significant differences in teacher expectations for BIPOC learners in math. However, we also see that when prompted to elicit and respond to student ideas and unique approaches to solving problems, teachers respond equitably across student groups, despite differences in perceived difficulty. Our findings suggest that surfacing and putting student thinking on the table and prompting teachers to respond may play a leveling role in math instruction. We discuss ideas for future study design and measurement of instructional constructs in this area as well as implications for opportunities to learn and teacher education in service of more just math pedagogy.

Authors

• Jeannette Garcia Coppersmith, Harvard University

• Hannah Kleen, DIPF | Leibniz Institute for Research and Information in Education

• Cynthia Pollard, Harvard University

• Heather C. Hill, Harvard University