Abstract
A simulation investigated use of the difficulty pa rameter (Bejar, 1980) to evaluate item unidimensional ity. Artificial tests were designed to be nonequivalent in both length and dimensionality. Simulated item re sponses to the tests were analyzed with the LOGIST computer program. Two indices were calculated: the slope of the principal axis between the content-area- based item difficulty estimates and corresponding to tal-test-based estimates, and the correlation between the two sets of estimates. Results show that the mag nitude of the correlation coefficient provides no infor mation about dimensionality of a set of test items. The slope of the principal axis, on the other hand, is sensi tive to multidimensionality in the data as well as test length. The size of the slope adequately detects the di mensionality of items for relatively long tests. Index terms: equating, item difficulty parameters, item re sponse theory, unidimensionality.
Get full access to this article
View all access options for this article.