Abstract:
This paper presents new evidence on the e®ects of attendance on academic performance. We exploit a large panel data set for Intro- ductory Microeconomics students to explicitly take into account the e®ect of unobservable factors correlated with attendance, such as abil- ity, e®ort and motivation. We ¯nd that neither proxy variables nor instrumental variables provide a viable solution to the omitted vari- able bias. Panel estimators indicate that attendance has a positive and signi¯cant impact on performance. Lecture and classes have a similar e®ect on performance individually, although their impact can- not be identi¯ed separately. Overall, the results indicate that, after controlling for unobservable student characteristics, teaching has an important independent e®ect on learning.