Friday, February 15, 2008

CHC interpretation of the KABC-II: Guest post by John Garruto


The following is a guest post by John Garruto, school psychologist with the Oswego School District and member of the IQs Corner Virtual Community of Scholars. John reviewed the following article and has provided his comments below. [Blog dictator note - John's review is presented "as is" with only a few minor copy edits by the blog dictator and the insertion of some URL links]


Reynolds, M.R., Keith, T.Z., Goldenring-Fine, J., Fisher, M.E. & Low, J.A. (2007). Confirmatory Factor Structure of the Kaufman Assessment Battery for Children—Second Edition: Consistency With Cattell-Horn-Carroll Theory. School Psychology Quarterly, 22(4), 511-539. [click here to view article]


Abstract:
  • The Kaufman Assessment Battery for Children-Second Edition (KABC-II) is a departure from the original KABC in that it allows for interpretation via two theoretical models of intelligence. This study had two purposes: to determine whether the KABC-II measures the same constructs across ages and to investigate whether those constructs are consistent with Cattell-Horn-Carroll (CHC) theory. Multiple-sample analyses were used to test for equality of the variancecovariance matrices across the 3- to 18-year-old sample. Higher-order confirmatory factor analyses were used to compare the KABC-II model with rival CHCmodels for children ages 6 to 18. Results show that the KABC-II measures the same constructs across all ages. The KABC-II factor structure for school-age children is aligned closely with five broad abilities from CHC theory, although some inconsistencies were found. Models without time bonuses fit better than those with time bonuses. The results provide support for the construct validity of the KABC-II. Additional research is needed to more completely understand the measurement of fluid reasoning and the role of time bonuses on some tasks.
Okay, I have to tie in the Super Bowl somewhere because the New York Giants won and I waited seventeen years for this to happen again. Like the Super Bowl, there are many practitioners who are interested in the end results (the score at the end of the game), not necessarily how one got there (the whole study) or the analysis of each play (the statistics). For the ease of readers, I’m going to jump to the score at the end of the game.

The Results: The K-ABC II is emerging as a serious contender among cognitive assessment batteries. I also want to say that from reading his posts on the CHC listserv, and now this article, I’m expecting to see some more good stuff from Matthew Reynolds (I’ve always been a fan of Tim Keith’s research and really like his stuff on the WJ-III).

This study seeks to analyze the K-ABC II from a CHC perspective, which is one of two theories the test is built upon (the other is the Luria-Das perspective where the original has its origins.) It’s worth mentioning that Kaufman is not new to CHC theory. The Kaufman Adolescent and Adult Intelligence Test (KAIT) also used Gf-Gc as its basis. The current study is an internal validity study using factor analysis methods. Several analysis were performed to determine best model fit with certain manipulations of the analysis.

First it bears mentioning that the test g (general intelligence) loadings are consistent with prior research (Gf and Gc being high g loaders-Gv as well interestingly!) Reynolds et al. sought to answer some interesting hypotheses regarding model fit as well as cross-factor test loadings. Here are some questions posed and answers provided:

  • Does Gestalt Closure measure Gc? The test requires subjects to look at “inkblots” (Gv) that resemble familiar objects (Gc). It was concluded that there was a Gc load on this subtest. My own thoughts…it might be neat to show the child a list of objects that represented the stimuli after the assessment is complete. From there, one could rule in or rule out Gc contamination. Nevertheless, the Gc load is important because Gv is often thought to be (or supposed to be) an area where less “acquired cultural knowledge” should impact performance.
  • Do Hand Movements measure Gf? This subtest measures a pantomime of repeated hand movements and is purported to load on Gsm. The authors note a relationship to Gf. The hypothesis was generated relating to strategy for success and working memory. My own thoughts…why isn’t anyone talking Gv? Sure this test requires motor planning (frontal activiation?), but I argue that success can result from remembering a visual sequence. Furthermore, Gsm has often been related to verbal prompt/auditory modality. Although the intertwining of working memory and fluid reasoning has been discussed..I’m not sure I see a huge component of either. The task appears very sequential to me. The visualization component is too hard to ignore. Given the lower load on Gsm I would be interested in looking at a Gv link.
  • Does Pattern Reasoning measure Gv? The analysis suggested loadings on Gv as well as Gf. I have lately found this link to be of interest. It seems pure measures of Gf are hard to find. Sometimes the comparison on the WISC-IV of Picture Concepts (Gf-I) and Matrix Reasoning (Gf-I) is interesting given that I see a major discrepancy in scores. Indeed the former requires more Gc and the latter more Gv-especiall if transformation of the stimuli is required in order to logically complete the puzzles. This becomes even more important if g=Gf...as some scholars have suggested. Under what presentation conditions then is Gf (g?) more successful? Could the learning modality trend be returning (just kidding-not touching that one!)
  • Does Story Completion measure Gc or Gv? The analysis suggested that the answer is "no." Story Completion appears to be a measure of Gf. It’s interesting because I remember reading that a similar test on the WISC-III (Picture Arrangement) had similar loadings on VCI as POI. I might have thought that there would be more of a Gc load on Story Completion than on Gestalt Closure.....but then Gc also requires verbal recall of names whereas this requires logical sequencing ability. I imagine there’s probably some Gc necessary, but not enough that having a lot of it will predict success (or too little will predict failure).
  • Do Rover and Block Counting measure Gf as well as Gv? The analysis suggested…definitely. Gv for Block Counting (which I would intuitively agree with) and the jury is still out for Rover. The deductive reasoning element with Rover is certainly apparent...but I think it’s important not to forget that Rover has some executive function elements to it (it’s not very much unlike Planning on the WJ-III). Right now though t seems Gv is present for Block Counting and Rover.
  • The Issue of Time bonuses: This research question was very important to me. I recall giving this battery to someone and finding out that the difference on one of the tests (whether timed bonuses were provided or not) resulted in a scaled score difference of almost two standard deviations! I followed the manual, which endorses using timed points unless there’s a reason not to. However, Reynolds et al. found a better model fit for no time bonus. This is not bad news. Sometimes we learn different practices after a test has been normed and published. I remember Kaufman’s book on the WISC-III where he indicated Symbol Search to be a higher ‘g’ loader than Coding and the informed practitioner may wish to have SS substitute for Coding as long as the decision was made in an apriori fashion. I guess for me, I do not want Gs contaminating a different factor I’m attempting to measure. I prefer to measure it far away from ‘g’ via a cross battery technique given that Gs has shown weaker relationships to ‘g’ but significant relationships to learning disabilities. Sometimes we learn new ways to practice as a result of follow-up research-this certainly fits that mold.
Overall Conclusion: I think K-ABC II is going somewhere. It is receiving some interesting recognition by some scholars and even is purported to have some utility with nonverbal, non English speaking, and/or autistic spectrum populations. Given the potential of this instrument in cognitive assessment, the research opportunities are certainly plentiful. I still see the Wechsler as my test of choice for Gc, the WJ-III as my favorite test to fill in the holes left by many cognitive batteries...but there certainly seems to be significant practical implications for the K-ABC II. Certainly the relationships to CHC theory are again very much substantiated. Certainly there are plenty of Patriots who want to view assessment from a traditional framework. Also, like the Patriots, there are those who are jumping on the loudest flavor of the month (RTI as being the only way to diagnose learning disabilities). However, the Reynolds et al. study continues to show that CHC theory has stood solid time and again as one of the "Giant" individual differences frameworks for use by school psychologists.

Technorati Tags: , , , , , , , ,

Powered by ScribeFire.

No comments: