Evaluating teaching performance of teaching-only and teaching-and-research professors in higher education through data analysis

Abstract

We present a study that compares the teaching performance of Teaching-only versus Teaching-and-research professors at higher education institutions. It is a common belief that, generally, Teaching-only professors outperform Teaching-and-Research professors in teaching and research universities according to student perception reflected in student surveys. We present a case study which demonstrates that, in the vast majority of the cases, it is not necessarily true. Our work analyzes these two type of professors at their ability to function as an intellectual challenger, learning guide and their tendency to be recommended to other students. The case study takes place at Tecnológico de Monterrey (Tec), a teaching and research private university in Mexico that has developed a research profile during the last two decades with a mix of teaching-only and teaching and research faculty members and shows a growing accomplishment on world university rankings. We use five datasets from a student survey called ECOA which accounts observations from 2016 to 2019. We present the results of statistical and machine learning methods applied when the taught courses of more than nine thousand professors are taken into account. Methods include Analysis of Variance, Logistic Regression, Recursive Feature Elimination, Coarsened Exact Matching and Panel Data. Contrary to common belief we show that, for the case presented, teaching and research professors perform better or at least the same as teaching-only professors. We also document the differences found on teaching with respect to attributes related to courses and professors.