LINKING STATEWIDE TESTS TO THE NATIONAL ASSESSMENT OF EDUCATIONAL-PROGRESS - STABILITY OF RESULTS

The adequacy of linking statewide standardized test results to the Nat ional Assessment of Educational Progress (NAEP) by using equipercentil e equating procedures was investigated. Statewide mathematics test dat a for eighth-grade students in 1990 and 1992 were obtained from four s tates. NAEP data for samples from these four states were obtained from the results of the Trial State Assessment administrations in the same years. Equating functions for male and female students in two states providing gender identification were similar at the low end of the sca le but diverged at the high end of the scale. Applications of the equa ting functions obtained for 1990 data to the statewide test results ob tained in 1992 provided estimates that were generally similar to actua l NAEP results near the median, but not in the tails of the distributi on. These results suggest that such linking, although reasonable for e stimating average performance for the state, are not sufficiently trus tworthy to use for making comparisons based on the tails of the distri bution.