Validate High Stakes Inferences by Designing Good Experiments, Not Audit Items: A Comment on “Self-Monitoring Assessments Educational Accountability Systems”

Click to increase image sizeClick to decrease image size

Notes

This is a commentary on “Self-Monitoring Assessments for Educational Accountability Systems,” by Daniel Koretz and Anton Béguin.

¹It is unfortunately outside the scope of the present comment for me to detail my objections to Koretz and Béguin's conceptualization of the target domain for inference as somehow being synonymous with the measurement of a latent construct. These are not the same thing (c.f., Borsboom, 2005) and to use the terms interchangeably can invite confusion. With this in mind I consciously avoid use of the terms measurement and construct in my discussion of the Koretz and Béguin SMA approach.