Commentary on "Construct Maps as a Foundation for Standard Setting"

The authors of this article comment on "Construct Maps as a Foundation for Standard Setting," by Adam E. Wyse (this issue) in which Wyse presents construct maps, a visual display of a variety of sources of evidence that support standard-setting decisions, and shows how this approach could be used with a variety of existing standard-setting approaches. Wyse argues that the process of setting cut-scores is complex, and it is often difficult for panelists to fully grasp how cut-score decisions relate to various other pieces of information provided during the process. By combining data into one conceptual framework, construct maps help panelists visualize the relationships between performance-level descriptors (PLDs), cut-scores on the latent construct, and various pieces of test data. The authors of this article strongly agree that this is a powerful approach. Wyse also argues that construct maps may provide solutions to issues with some standard-setting methods. The authors state that while they do not disagree, they feel that best practices in implementing these various methods could minimize or avoid these specific issues, even without the use of construct maps. In this article, the authors address several misconceptions regarding the Body of Work method, issues associated with multidimensionality, and issues associated with the choice of an item response model.