ABSTRACT
In many testing applications, the framework for the test is specified initially, and this specification subsequently guides the development of items and scoring procedures. Empirical relationships may then be used to inform decisions about retaining, rejecting, or modifying items. Interpretations of scores from tests developed by this process have the advantage of a logical/theoretical and an empirical foundation for the underlying dimensions represented by the test. (American Educational Research Association, American Psychological Association, & National Council on Measurement in Education [AERA, APA, & NCME], 1999, p. 41)
