Pascal errors—empirical evidence
J Pugh, Deborah Simpson · ITNOW · 1979
During the past ten years or so, a number of papers have been published which have offered critical accounts of the language. Most of these papers look at the more advanced aspects of the language and so seem particularly aimed at the use of Pascal for serious systems programming, Wirth [2] and Welsh et al [3] are sensible examples. The other group of papers talk in generalities about the advantages of using Pascal for teaching programming; Nutt [4] is a good example. In this experiment we have followed the second approach but report empirical results of the errors made by students when following a first programming course using Pascal. The results obtained may give teachers guidelines on which parts of the language need careful explanation and may also be of use to designers of future languages. On the first year of the B.Sc. Computer Studies course at Sheffield City Polytechnic all students follow a course in programming, in which course students become proficient at designing algorithms for (fairly simple) problems and programming these in two languages. For the past three years these languages have been Fortran/Cobol, Algol W/Cobol and currently Pascal/Cobol. In parallel with this course students also study the usual first year undergraduate curriculum.