On the Construct Validity of Multiple- Choice Items for Reading Comprehension

Abstract

In this study 590 third-grade students took one of four reading comprehension tests with either multiple- choice items or open-ended items. Each also took 32 tests indicating 16 semantic Structure-of-Intellect (si) abilities. Four conditions or groups were distinguished on the basis of the reading comprehension tests. The four 33 x 33 correlation matrices were analyzed si multaneously with a four-group LISREL model. The 16 intellectual abilities explained approximately 62% of the variance in true reading comprehension scores. None of the SI abilities proved to be differentially re lated to item type. Therefore, it was concluded that item type for reading comprehension is congeneric with respect to the SI abilities measured. Index terms: construct validity, item format, free response, reading comprehension, Structure-of-Intellect model.