EFFECTS OF A PROBLEM-BASED LEARNING APPROACH ON STUDENTS’ SPREADSHEET SKILLS
This paper reports the preliminary findings from a quasi-experimental project to study the effects on bachelor students’ spreadsheet skills after a section on spreadsheet modelling. The spreadsheet modelling section was part of a compulsory 7.5 ECTS introductory course on business data processing. In the pretest of our quasi-experiment, conducted at the beginning of the course, we asked the students to hand in their answer to a spreadsheet task. The task was voluntary, and 31% of the students (N=117) submitted their model or an empty spreadsheet. The results showed that around 90% of the students lacked basic technical spreadsheet skills, and that they did not know how to structure a spreadsheet so that their model was useful for decision support. The posttest, conducted at the end of the spreadsheet modelling section, showed that more than 90% of the same students (N=117) had acquired basic spreadsheet skills and managed to model a similar task in the spreadsheet according to principles of spreadsheet design. There was, however, one particularly disappointing exception: 26% of the students (N=30) still had constants in some of their formulas and, thus, hidden assumptions in their model. It is not surprising that most students had acquired spreadsheet skills after the course. Therefore, we emphasise the weaknesses of the students’ answers in order to enhance the understanding of how to improve spreadsheet teaching. Limitations of our study are discussed together with suggestions for further research.