| Thomas Dee
Swarthmore College
A teacher like me: Does race, ethnicity or gender matter?
FINAL REPORT:
It is widely believed that the racial, ethnic, and gender dynamics within classrooms (e.g., role-model effects, stereotype threat, teacher biases) are an important determinant of student achievement, in general, and of the minority/non-minority achievement gap, in particular. However, the available empirical evidence is relatively limited and contradictory. In large part, this is because it is difficult to establish the causal effects of student-teacher interactions when sorting across and within schools can reflect a variety of inherently unobservable traits. This project presents new evidence on the academic consequences of assignment to a demographically similar teacher based on the National Education Longitudinal Study of 1988 (NELS:88). The research methodology identifies the causal effects of these student-teacher interactions by exploiting a unique, "matched-pairs" feature of NELS:88's survey design. Specifically, NELS:88 contains evaluations of students from multiple teachers as well as test scores for the subjects taught by these teachers. Panel-based econometric models that rely on these multiple observations purge the potential biases related to the unobserved traits of both students and teachers. In other words, this research design can effectively compare how the same student performs with (or is perceived by) different teachers. The results indicate that a demographically dissimilar teacher is more likely to view a given student pejoratively. These results also indicate that assignment to an opposite-gender teacher can significantly harm a given student's intellectual engagement and achievement.
Back to Funded Research Grants Page |