What's in a Match? Disentangling the Impact of Teacher Race/Ethnicity

• In Event: Promising Scholarship in Education: Dissertation Fellows and Their Research

Abstract

The 2014/2015 school year marked a turning point in the demography of American education. For the first time, nonwhite students constituted the majority of elementary and secondary public school enrollment (National Center for Education Statistics (NCES) 2014). Even as America’s students have grown increasingly diverse, however, its teaching force has remained predominantly white. Despite three decades of initiatives to recruit teachers of color, white teachers still constitute over 82% of the teaching force (Villegas, Strom and Lucas 2012; NCES 2013).

This growing gap between students and teachers has generated considerable interest among both policymakers and researchers. As of 2012, at least 31 states as well as the U.S. Department of Education had instituted programs to recruit teachers of color (Bireda and Chait 2011; Villegas, Strom and Lucas 2012). These efforts find support in research demonstrating that black and Hispanic students benefit from increased exposure to teachers who share their racial/ethnic background (e.g., Meier 1984; Dee 2004; Hanushek et al. 2005; Pitts 2007; Grissom, Nicholson-Crotty and Nicholson-Crotty 2009; Egalite, Kisida and Winters 2015; Grissom and Redding 2016).

Although researchers have documented the positive effects associated with student-teacher racial/ethnic matching (hereafter, “matching”), they have been unable to provide any direct evidence on the mechanisms that underlie these effects. From the students’ perspective, why does matching matter? How do matched and non-matched teachers differ in terms of their pedagogy, their standards, their ability to connect with and motivate their students? Answering these kinds of questions does not exhaust the ways in which matching may promote student success. However, understanding how students’ experiences vary by teacher background provides the essential baseline for future work.

The current study provides that baseline. Using over 56,000 new student surveys from six districts, I provide the first large-scale evidence on the specific ways that matching affects student experience. The richness of my data, collected as part of the Measures of Effective Teaching (MET) study, allows me to analyze multiple dimensions of teacher practice and student engagement. Comparing students within the same school, I find that matched black and Hispanic students report significantly better experiences than their non-matched peers. Moreover, the distinct profiles of benefits that these students report are consistent with prior qualitative research on the styles and successes of black and Hispanic teachers (e.g., Valenzuela 1999; Delpit 2006). These results emerge primarily among middle school students reporting on English Language Arts (ELA) classes. Interestingly, white students do not report significant variations across matched and non-matched classrooms. Finally, the positive matching effects reported by black and Hispanic students do not seem to driven by problems experienced by their non-matched peers. Instead, matching effects occur on top of an already high baseline reported in non-matched classrooms.

The full paper (available upon request) is structured as follows. First, I review the existing evidence on matching effects. Second, I highlight the central gap in this literature: the inability to provide evidence on the mechanisms that underlie its results. Third, I introduce my data and discuss my analytical approach. Fourth, I review my results and demonstrate that they are robust to two main threats to inference. Finally, I discuss the findings, exploring the ways in which they both reinforce and refine existing work. I conclude with an acknowledgement of limitations and present implications for future research and policy.

Author

• Joel Mittleman, Princeton University