UC Irvine

Teach For America (TFA) is a high-profile alternative teacher certification program that seeks to provide high-quality teachers to students in low-income schools. While a growing body of research examines the average impact of TFA, few researchers examine how its impacts may vary across students and whether it positively impacts all children.

This dissertation broadens our understanding of the impacts of TFA by examining variation in its effectiveness across a number of dimensions. The first study uses experimental data from Mathematica Policy Research to examine the effect of TFA teachers on the distribution of student achievement in elementary school. Results suggest that, relative to non-TFA teachers, TFA teachers have a positive average effect on math achievement that is shared across much of the distribution, while in reading TFA has a negative effect on the bottom of the distribution and a positive effect on the top, particularly relative to veteran non-TFA teachers.

The second and third studies use administrative data from North Carolina. The second study tests for variation in the effect of TFA across grade levels and subject areas and finds that TFA has positive effects in most subjects, which are largest in high school science and math. In addition, examinations of the relative fit between TFA and student baseline proficiency suggest that TFA teachers have larger effects for initially higher-performing students in elementary and middle school, but that in high school their effects are larger among initially lower-achievers, underscoring the importance of considering the person-environment fit. The third study examines whether TFA's effects on student achievement have changed over time and demonstrates that TFA's efficacy has increased as the program has matured, and that this change is not accounted for by most observable measures of teacher quality.

Together, the three studies provide new evidence about the benefits and shortcomings of TFA, improving our understanding of the efficacy and potential of TFA and similar programs. These results provide policy-relevant information that speaks to larger debates about alternative certification and teacher quality in high-poverty schools, and provide evidence regarding the effective allocation and training of teachers for high-poverty schools through both alternative and traditional teacher training programs.