Validating summative assessment
He defined this argument as an evaluation of the proposed uses and interpretations of test scores.Describing the traditional trinity of validity conceptions (criterion, content, and construct) as “strands within a cable of validity argument,” Cronbach emphasized the need to play devil’s advocate in the development of a persuasive validity argument.
The implications of this idea are clear: if test scores are used or interpreted for a purpose other than the one being validated, then a new validity argument must be crafted.
Messick (1989) was one of the first to outline a unified approach.
Using the Construct model as a basis for this unified approach, he defined validity as “an integrated evaluative judgment of the degree to which empirical evidence and theoretical rationales support the based on test scores or other modes of assessment” (Messick, 1989, p. One issue with this conception is that it does not provide much guidance for the validation effort.
The argument should not only seek to confirm, but also to falsify and contribute to revision — especially for a “young” instrument, such as that presented in this study.
A very approachable summary of this unified conception of validation and a guide for structuring validation efforts is presented in latest edition of the Standards for Educational and Psychological Testing (American Educational Research Association, et al., 2014).