Monday, January 24, 2011

The KAIT Gf-Gc IQ test: An under-appreciated contribution to the evolution of CHC intelligence theory

I have been busy revising my 2005 book chapter on CHC Theory: Past Present and Future. It is now coauthored with Dr. Joel Schneider who took the lead and is now first author. Together Joel and I wrote WAY toooooooo much material, and we had to do some serious editing...dropping major sections that we thought were important. I have decided that some of those sections that I wrote would appear hear at IQs Corner, and possibly in future manuscripts yet to be determined.

Part of the chapter is a visual-graphic presentation, with narrative text, of the CHC Timeline project I've been working on for a few years. If you visit the timeline link you will also gain access, via links, to the original chapter, which will provide you more context.

Cut from the submitted draft were comments about the under-appreciated role of KAIT IQ test in the evolution of Gf-Gc/CHC assessment instruments. You can see this event in the timeline at the link above. Below is the text written to explain the small, yet important, contribution of the KAIT to the evolution of CHC theory. Although I wrote most of this section, there are a few well crafted sentences below that were written by Joel.

The contribution of the KAIT Gf-Gc test to the evolution of CHC theory and assessment

An under-appreciated contribution to the evolution of Gf-Gc intellectual assessment was the publication of the Kaufman Adolescent and Adult Intelligence Test (KAIT) (Kaufman & Kaufman, 1993), based on Cattell’s (1941, 1943) original gf-gc Theory instead of the Horn–Cattell (1966) Extended Gf-Gc Theory. The Kaufmans used the older theory not out of ignorance but for pragmatic reasons (and some admirably subtle theoretical justifications as well). However, the distinction between the older, broader, and deliberately multidimensional gf and the newer, narrower, and more unidimensional Gf (as reflected Extended Gf-Gc theory) was probably lost on most potential users of the test.

By self-admission, Kaufman (2009) noted that the KAIT “sadly, has been all but ignored in the United States” (p. 1). The failure of the KAIT to attract assessment professionals was most likely due to the dominance of the venerable adult Wechsler batteries and bad timing. The trend in test development was moving toward giving users the option to measure narrower abilities with more factor-pure subtests and the KAIT was out of step with this trend. Flanagan and McGrew’s (1998) joint KAIT/WJ-R CFA presented an eight-factor Extended Gf-Gc model (Gf, Gc, Grw, Glr, Gsm, Ga, Gv, Gs) and suggested that the KAIT Gf composite, by the standards of the Extended Gf-Gc Theory, was a mix of Gf, Gsm, Gv, and Glr. The decision to measure the older, broader gf was not wrong (indeed, other tests have composite scores with comparable ability mixtures such as the WJ-III Thinking Abilities composite, CAS Full Scale IQ, KABC-II Mental Processing Composite, DAS-II Special Nonverbal Composite, or the now-discarded PIQ from the Wechsler series). The problem was that only the Gf and Gc constructs were represented by at least two KAIT tests and thus there was no option to form reasonably unidimensional composite scores.

​The small KAIT blip on the Gf-Gc assessment radar screen was important for a number of reasons. First, Alan Kaufman’s opinion carries weight in this field. When he recognized the importance of Gf-Gc theory (first in the publication of the original K-BIT, which is now in its second edition), other test developers noticed. Post-KAIT publication Kaufman and colleagues began an active program of Gf-Gc research and publication (see Kaufman, 2009) that foreshadowed Kaufman’s embracing of CHC theory as the consensus model of intelligence for use in the development of intelligence batteries (Kaufman, 2009). Second, it was during their WJ-R/KAIT CFA study that Flanagan and McGrew (1998) recognized that all Gf-Gc assessment research up to that point in time had myopically focused exclusively on the broad Gf-Gc domains, ignoring the importance of understanding individual tests as per the narrow (stratum I) abilities they measured.

I would not be surprised if the KAIT, or parts of the battery, reappear in some form of cognitive assessment instrument(s) in the future.

- iPost using BlogPress from my Kevin McGrew's iPad

No comments: