Effects of mathematics preparation and prior language exposure on perceived performance in introductory computer science courses

Reviewer: Michael James Folk

This is a short summary of a survey of 467 students in introductory computer language courses, which is designed to determine the effects described in the title. The survey found a strong positive correlation between the number of mathematics courses students had in high school or college and the grades those students expected to receive in Pascal, COBOL, and BASIC courses. It also found positive correlations between preexposure to certain languages and expected grades in other languages. The highest correlations were between preexposure in Pascal and expected grade in COBOL, and between preexposure to BASIC and expected grade in COBOL. Preexposure to BASIC showed little correlation with expected grades in Pascal. A common criticism of studies that use correlation to imply cause and effect, as this one seems to do, was addressed by testing whether certain attitude factors toward mathematics and computer science were also highly correlated. The results were largely insignificant. The results of the survey are not compelling, but they do support a belief that many computer science educators harbor: that learning one language makes it easier to learn another language, with the possible exception of BASIC; and that the more mathematics students have, the more success they are likely to have in computing.