Association for Public Policy Analysis & Management

There are more than 700 honors programs at four-year U.S. universities and colleges, enrolling upwards of 250,000 students.  Honors programs have become so popular that U.S. New and World Report has a separate ranking for the top programs.  Despite the growth in honors colleges and programs, there is astonishingly little academic research on their impact on outcomes, either during college or after.  Our paper is, to the best of our knowledge, to undertake to estimate the causal impact of an honors program.

Our study examines the impact of an honors program at a large multi-campus public urban university.  We have administrative data on the population of students (both honors and non-honors) enrolling in the university from fall 1999 to spring 2012.  These data allow us to follow the students through their academic career and include transcript information, financial aid information, and admissions data like SAT scores and high school GPA.  These data have also been linked to quarterly earnings data from the unemployment insurance (UI) system in the state of the university, allowing us not only to examine the impact of the honors program on within-college outcomes like GPA and completion rates, but also labor market outcomes like employment and earnings. We also have access to the application roster of the honors program (including students who did not enroll), allowing us potentially to track non-enrollees schooling outcomes through the National Student Clearinghouse.

We use two separate research designs to examine the outcomes of honors student enrollees relative to different control groups.  We will examine incident of completion, time to completion, graduate school attendance, post-graduation employment, and post-graduation earnings.

In the first design, we match the admissions rosters to the population of enrolled students in our large public urban university, including the non-honors students.  To the extent that there is a discontinuity in certain admission criteria like SAT scores and high school GPA, we compare the outcomes for enrolled honors students who were just above the admission cutoff to those who were not admitted to the honors program but still enrolled in the university under study.

It is possible comparisons based on freshmen enrollments will be somewhat noisy.  To evaluate the impact of the honors program, we will perform another comparison at the halfway point of students’ 4-year college careers.   Given the very large sample sizes at our disposal, we can match the honors students to non-honors students with similar GPAs, majors, etc., after their sophomore year and then estimate the differences in completion rates, graduate school attendance, etc.

Lastly, in the most complex analysis, we can use the admissions office roster to examine the difference between honors college enrollees and those who were admitted but went elsewhere by finding them in the National Student Clearinghouse data.  We will use a “value added” framework to examine the impact of attending an honors college relative to an alternative college ranked by selectivity.