ASSESSING UNIDIMENSIONALITY OF POLYTOMOUS DATA

This study investigated the performance of Poly-DIMTEST (PD) to assess unidimensionality of test data produced by polytomous items. Two type s of polytomous data were considered: (1) tests in which all items had the same number of response categories, and (2) tests in which items had a mixed number of response categories. Test length, sample size, a nd the type of correlation matrix (used in factor analysis for selecti ng AT1 subset items) were varied in Type I error analyses. For the pow er study, the correlation between Bs and the item-theta loadings were also varied. The results showed that PD was able to confirm unidimensi onality for unidimensional simulated test data, with the average obser ved level of significance slightly below the nominal level. PD was als o highly effective in detecting lack of unidimensionality in various t wo-dimensional tests. As expected, power increased as the sample size and test length increased, and the correlation between the theta s dec reased. The results also demonstrated that use of Pearson correlations to select ATI items led to equally good or better performance than us ing polychoric correlations; therefore Pearson correlations are recomm ended for future use.