Conditional Covariance-Based Subtest Selection for Polytomous DIMTEST

Dimensionality assessment is an important component of test validation. While the vast majority of statewide standardized tests contain both dichotomous and polytomous items, much of the work in dimensionality assessment has focused on the case of dichotomous item exams. Poly-NEWDIM is a polytomous version of DIMTEST that is a nonparametric hypothesis testing procedure for dimensionality assessments based on comparison of an assessment subtest (AT) with a partitioning subtest (PT). A good AT and PT selection is vital to the performance of Poly-NEWDIM. The proposed two new AT-selection procedures, HCA/CCPROX-PolyDIMTEST (HCP) and HCA/CCPROX-PolyNEWDIM (HCN), perform better than the previous AT-selection method (HCD) in terms of power. Moreover, HCN shows less sensitivity than HCP to sample size, correlation, and structural complexity. The results also indicate that 70% of data for AT selection is appropriate for all three kinds of tests (dichotomous only, polytomous only, and mixed) with large sample sizes and dichotomous tests with small sample sizes, while 50% is good for polytomous tests and mixed tests with small sample sizes.