Computerized Adaptive Testing with Item Cloning (CT-01-03)
by Cees A. W. Glas and Wim J. van der Linden, University of Twente, Enschede, The Netherlands
A major impediment to cost-effective implementation of computerized adaptive testing (CAT) is the amount of resources needed for item pool development. One of the solutions to the problem currently pursued is generating pools of items by using item-cloning techniques. The approach is based on a formal description of a set of “parent items” along with algorithms to derive a larger set of operational items from them. These parent items have been known as “item forms,” “item templates,” or “item shells,” whereas the items generated from them are now widely known as “item clones.” An important consequence is that these techniques may cause variability on the item parameters.
To tackle this problem, a multilevel IRT model is presented where it is assumed that the item parameters of a 3PL model describing response behavior are sampled from a multivariate normal distribution associated with the parent item. A marginal maximum likelihood estimation procedure for calibration of a pool of item clones is described. A two-stage item selection process for computer adaptive testing is outlined. First, a set of items cloned from the same parent item is selected to be optimal at the ability estimate. Second, a random item from this set is administered. A simulation study illustrates the accuracy of the item pool calibration and ability estimation procedures relative to the accuracy of regular adaptive testing from a pool of individual items, and relative to procedures based on the 3PL that ignore the multilevel structure.