A series of consecutive progress test scores indicates how a studentâ€™s knowledge is developing. However, scores are affected by measurement errors due to variations in test difficulty and student-item interactions (for example: when the anatomy questions in the next progress test are more difficult, a student who generally does quite well on anatomy may do less well because the questions are more difficult to answer). The cumulated score is used to reduce the effects of measurement errors in a series of scores. The cumulated score is obtained by calculating the mean score on the progress test in question and all the preceding tests. In other words, the cumulated score is the mean score so far. The effect of measurement error is expected to be much smaller in cumulated scores, because cumulated scores are calculated over a much larger number of questions than the score on one progress test. Indeed, the cumulated score can be regarded as a more reliable indicator of the development of knowledge than the original series of raw scores. This effect is even more marked with series of scores per category or per discipline, where the number of items per progress test can be very small and the measurement error very large as a result.
Go back to the page ‘Basic concepts’