Essays on Instructor Quality in Higher Education

How do teachers affect students’ academic and labor market outcomes? Research into teacher quality has been thoroughly scrutinized for the K-12 sector, while there is a requirement for examining these questions at post-secondary education level. In the past few decades, several important trends of faculty employment among higher education institutions have emerged. First, faculty employment in higher education in the United States has gradually transformed from a bifurcated system based on tenure status into a trifurcated system, constituting three types of faculty: those which are tenure eligible, fulltime but not tenure eligible, and part-time faculty. Second, due to aging of the American professoriate, particularly those faculty members hired in the 1960s and 1970s, colleges and universities have been recruiting more diverse candidates, such as female faculty, to fill positions. My dissertation examines the implications behind these two important trends. In the first chapter, I provide a detailed portrait of non-tenure-track faculty in terms of their demographic information, personal attributes, and employment trajectory across institutional sectors and academic subjects. Based on unique datasets linking college administrative information on student transcripts to Unemployment Insurance (UI) data on faculty employment history, I find that there is significant variation in individual characteristics and employment patterns across non-tenure-track faculty who were hired through different types of contracts with the colleges. In the second chapter (co-authored with Di Xu), we examine the impact of non-tenure track faculty by types of employment on students’ academic outcomes in two- and four-year colleges using a two-way fixed effects model and an instrumental variable approach. We also analyze how the estimated effects on student outcomes can be explained by observable instructor characteristics and employment features. We find that non-tenure track faculty have positive impacts on current course grades but negative impacts on subsequent course outcomes. These negative impacts are stronger for non-tenure track faculty hired through temporary appointments than those hired with long-term contracts, which can be explained partly by observable instructor characteristics. In the third chapter, I document the existence of long-term effects of faculty gender on female students’ occupational choices, likelihood of employment, and earnings six years after the initial term of college enrollment, based on a novel dataset that links college administrative data with Unemployment Insurance (UI) records from a state college system for both public two- and four-year colleges. To minimize bias from student systematic sorting by the gender of instructors, I use an instrumental variable (IV) approach which exploits term-by-term variations in total course enrollments with female faculty in each college-department, after controlling for fixed effects of the course set students took during the first term. I find that female students in four-year colleges who take more course credits with female faculty in their initial semester are more likely to be employed overall, be employed in industries with more Science, Technology, Engineering, and Math occupations (STEM), and have higher annual earnings six years after; no effect is detected in two-year colleges.