Why is Cognitive Diagnosis Necessary? A Reaction

Click to increase image sizeClick to decrease image size

Notes

¹See Davidson (2000) for more on statistical determinism.

²To see this in action, we need only visit some sample items from the Army Alpha Test, for example, those at the Microsoft Encarta Web site, and ponder what these items are measuring and what role (if any) they have in predicting success or failure as a soldier: http://encarta.msn.com/media_461550852_761570026-1_1/u_s_army_alpha_intelligence_ test_sample_questions.html

³For a wonderful example of this risk in language testing, see Peirce (1992).

⁴They are not alone in expressing this regret. Jang says virtually the same thing at page 233, but she makes the further salient point that such retrofitting is particularly problematic if the test was not designed with diagnosis in mind, from the outset.

⁵(I am grateful to Charles Stansfield for pointing this out to me some years back.) There is no IHTA (International History Testing Association) or IAPTA (International Anatomy and Physiology Testing Association), and so forth, but we do have ILTA and we have the yearly LTRC—for which the April 2010 meeting will be the 32nd. Our discipline is not only established, but growing.

⁶With a nod to Ron Howard's brilliant movie Apollo 13 .

⁷Many of them are authors in this special issue.