A STUDY OF THE COMPARABILITY OF SCORES OBTAINED ON THE ATP MATHEMATICS LEVEL I AND LEVEL II TESTS1

This report describes the results of two studies which were conducted to obtain information regarding the comparability of scores on the Admissions Testing Program Mathematics Level I and Mathematics Level II tests. It has been assumed that if the scaling and equating process used with the tests is functioning properly, those students with adequate preparation to take both tests should receive similar scaled scores regardless of the test they choose to take. Results of previous studies had raised some concern that the scaling and equating methods used to date have not resulted in the proper alignment of the scales of the two tests. In addition, concern has been raised that the inclusion of Plane Geometry items in the Mathematics Level I exam disadvantages examinees prepared to take the Level II exam in that they have, in general, taken their course in plane geometry some time prior to the test. In the first study, an experimental single-group equating of scores from a form of the Mathematics Level II test to scores from a form of the Mathematics Level I test was conducted using data from a sample of examinees who elected to take both tests at the December 1983 administration. The resulting raw-to-scale conversion was compared to the operational conversion which was derived from a common-item equating of the Mathematics Level II form to a previous form of the same test. In the second study, the relative performance of groups of students with comparable amounts of course preparation but different amounts of time since they last took a plane geometry course was examined. The results of the equating study, though not definitive, suggest that the scale of the Mathematics Level II test may be somewhat higher than the scale of the level I test. The results of the second study indicate that students prepared to take both tests are probably not disadvantaged by the inclusion of plane geometry items in the Mathematics Level I test.