One hundred and seven psychology students are given questionnaires at the end of a statistics course. These are used to generate a score between 0 and 100 indicating how valuable they judged the course to be. These scores correlate (r = .59) with the students' marks for the course, namely the percentage of correct answers in an exam. The lecturer in charge of the course was curious to find out whether this correlation could be explained entirely on the basis of the students' statistical ability before taking the course. This was assessed by a different test where the students were timed solving a number of mathematical puzzles yielding a score in minutes. This correlates with both the final exam (r = -.71) and with the participants' rating of the value of the course (r = -.65).