Sections 3 and 4 describe the data set and the econometric methodology,
respectively. Section 5 presents the results of the empirical analysis. Section
6 concludes with a discussion of the main findings and the implications of
the analysis.
2 Literature review
In a widely cited study, Romer (1993) reported evidence on absenteeism in
undergraduate economics courses at three major US universities, finding an
average attendance rate of about 67 per cent. The paper also presented
regression results, based on a sample of 195 Intermediate Macroeconomics
students, indicating a positive and significant relationship between student
attendance and exam performance. This result was found to be qualita-
tively robust to the inclusion among the explanatory variables of students’
grade point average and the fraction of problem sets completed.3 On the
basis of these findings, Romer suggested that measures aimed at increasing
attendance, including making attendance mandatory, could be considered.4
Prior to Romer, Schmidt (1983) had investigated student time allocation
in a sample of 216 macroeconomic principles students, finding that time
spent in lectures and discussion sections has a positive and significant effect
on exam performance, even after controlling for hours of study. Park and
Kerr (1990) had found an inverse relationship between students’ attendance
and their course grades in a money and banking course over a four-year
period, even after controlling for the effect of unobservable motivation by
means of students’ self-reported hours of study and their perceived value of
the course.5
Following the controversial conclusions of Romer (1993), in the past
decade a number of empirical studies in the economic education literature
3 In order to control for the effects of motivation, Romer also examined the results
obtained by restricting his sample to students who had completed all the problem sets
assigned during the semester.
4 “I believe that the results here both about the extent of absenteeism and its relation to
performance are suggestive enough to warrant experimenting with making class attendance
mandatory in some ungergraduate lecture courses.” (Romer, 1993, p. 173).
5See McConnell and Lamphear (1969), Paden and Moyer (1969), Buckles and McMahon
(1971), Browne, et al. (1991) for early studies finding no significant impact of attendance
on academic performance. See also Siegfried and Fels (1979) for a comprehensive survey
on research on teaching college economics.