In this paper the relationship between attendance and attainment in a standard calculus course is investigated. Calculus could in principle be studied without attending lectures due to the wealth of material available (in hardcopy and online). However, in this study we will show that the pass rate of students attending classes regularly (>75% of the classes) is much higher than the pass rate of students attending fewer classes. We use a logistic model to investigate whether this correlation is significant. We will argue why we believe that this correlation between attendance and attainment is causal, i.e. why it is necessary for most students to attend classes in order to (improve their chances to) pass the exam.