The Effects of Repeaters on Test Equating

Abstract

Test fairness implies that successive forms of a test are equivalent in all key areas; since most forms cannot be precisely equivalent it is necessary to equate for scores on the forms to be directly equivalent. The effect of repeaters was explored empirically in the context of test equating for this paper. Results indicate that the group taking the test at different time periods was not a random sample drawn from an underlying population. This effect is due to the presence of repeaters in the second test administration. Determination of equating parameters resulted in different cut-off points for passing on the test. It is now clear that the performance of repeaters tends to move the passing score downward, allowing greater advantage to those who are repeating a test to do so when their numbers are greater.