The Effects of Test Disclosure on Linear Equating Relationships under the Common Item Nonequivalent Groups Design

Abstract

The proponents of test disclosure argue that disclosure is a matter of fairness; the opponents argue that fairness is enhanced by score equating which is dependent on test security. This research simulated disclosure on a professional licensing examination by placing response keys to selected items in some examinees' records, and comparing their scores on a later form of the same test to scores of examinees who did not receive disclosure. Ten groups varied in number of items disclosed, number of examinees receiving disclosure, ability level of the subgroup receiving disclosure, and whether the items disclosed were anchor test or nonanchor test items. Results depended on the degree of exposure to test items; the greatest score differences were found in the group in which the greatest number of people received disclosure on the greatest number of test items. The effect of item disclosure on passing scores and passing rates was also examined. Results depended on whether the disclosed items were anchor or nonanchor items and whether the benefit of disclosure was direct or indirect. It was a concluded that there is currently no perfect equating method for test disclosure and that test fairness is more complicated than originally thought. Efforts to have fair disclosure in occupational licensing tests must be weighed against the need to protect the public. (JGL)