Portal de Periódicos da CAPES

Comparison between computer‐aided testing and traditional multiple choice: an equivalence study

2007; Wiley; Volume: 11; Issue: 1 Linguagem: Inglês

10.1111/j.1600-0579.2007.00432.x

1600-0579

Matthias Karl, Friedrich Graef, Stephan Eitner, Nikolas Beck, Manfred Wichmann, Stefan Holst,

Digital Communication and Language

Aims: Computer‐aided learning (CAL) is a viable alternative to traditional teaching methods. While CAL is used for teaching only, computer‐aided testing (CAT) can be applied as a computer‐based examination method. The aim of the present study was to compare the performance of undergraduate students in computer‐interactive tests and traditional multiple choice tests and to evaluate the influence of question length and language skills on test results. Materials and methods: Fifty‐two native German‐speaking students and five non‐native speaking students were randomly assigned to two groups. They were required to answer 20 conventional multiple choice questions and 20 questions using the interactive MobiTed® system. An analysis of variance and t ‐tests ( α = 0.05) were used to compare scores. Results: Non‐native speakers scored significantly lower on the interactive test compared with the native speaking students ( P = 0.048 for long questions and P = 0.003 for short questions). There were no significant group differences on the conventional multiple choice test ( P = 0.052 for long questions and P = 0.089 for short questions). Based on the results of the native speaking students neither test method nor question length had an influence on overall performance. Conclusion: Interactive multiple choice tests may provide an alternative to written tests, especially when considering time expenditure. Language skills and question length should be taken into account when setting time limits for examination.