[Excerpt] This paper analyzes how the composition of an institution’s student body affects the performance of that institution’s students. In particular, we investigate how the average student quality and the dispersion in student quality within the student body affects the future earnings of individual students. We begin in section II by examining the two primary reasons why one’s fellow students would affect future labor market success: peer effects and employer screening. Peer effects are important because one’s peers can augment or detract from human capital accumulation through numerous types of interactions. The literature examining these potential effects is growing, and we summarize the work and relate the findings to our specific question.
The second reason, employer screening, captures the role of one’s fellow students in shaping the beliefs of employers about the quality of students at the institution. These beliefs may affect employers’ actions in two manners. First, the level of student quality may affect the intensity by which employers recruit at that institution. Second, the dispersion in student quality may affect the degree to which employers screen by ability when interviewing the institution’s students. We formalize this discussion of employer behavior with a screening model.
The analyses in Section II generate several predictions, and we turn next to tests of their validity. After reviewing past research on the topic in Section III, we use the 1982 cohort of the High School and Beyond survey in Section IV to provide additional evidence. Our findings are consistent with much of the previous literature. We find that the level of student quality at an institution is an important determinant of earnings; a 100 point increase in the median SAT is related to a 3 percent increase in a student’s annual earnings. In addition, students with lower SAT scores appear to receive the highest premium from attending an institution with a high median SAT. We do not find that the range in ability of one’s peers is a strong predictor of earnings. While some evidence suggests that that those at the bottom of the institution’s ability distribution suffer a wage penalty from an increased spread in student ability, the relationship is fairly weak and not statistically significant.