| Abstract: | In psychological measurement it is important to determine when a particular examinee's test or scale score provides an invalid measure of the trait or attitude being assessed. In this article we present several quantitative indices that have been found to effectively identify some types of inappropriate scores. These measures, termed appropriateness indices, are all derived from item response theory. They are computed directly from the item responses that are combined to form the test or scale of interest; information from other scales or tests is not needed. A decision-theoretic approach to the use of appropriateness indices in selection decisions and theoretical research is introduced. An example is then presented to illustrate how researchers can use appropriateness indices. Finally, we discuss policy options that are available for dealing with individuals who are identified as having inappropriate scores. (27 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
