economics Paradigm Challenge

The professors that students like the most are often the ones who contribute the least to their future earnings.

April 25, 2026

Original Paper

Instructor Value-Added in Post-Secondary Education

Merrill Warnick, Jacob Light, Anthony LokTing Yim

SSRN · 6616698

AI-generated illustration

The Takeaway

Student evaluations are the primary way universities measure the quality of their teachers. Data shows that a student's future salary can increase by 5% if they have a high-quality instructor, but those instructors often get mediocre reviews. There is a very weak correlation between who students enjoy learning from and who actually prepares them for the labor market. Popular teachers might make the classroom experience pleasant without providing the difficult skills that pay off later. Relying on student feedback to promote faculty might be actively harming the long-term financial success of graduates.

From the abstract

Value-added methods are widely used in economics to measure how heterogeneous agents contribute to outcomes, but their credibility hinges on quasi-random exposure to those agents. In higher education, this condition fails: when students select courses and instructors based on unobserved characteristics, conventional value-added estimates will be biased. We develop a method that groups students with similar course histories and identifies instructor effects within these groups by comparing outcom