Tuesday, April 6, 2010

Court Decision: Johnston v Florida appeal denied--interesting expert opinions re: old vs new IQ scores

Last month I passed along an urgent plea re: an appeal of the non-MR/ID determination of Johnstone (Johnstone v Florida, 2010).  Yesterday the judge ruled against Johnstone an indicated that the burden of proof that he was MR/ID was not meet.

Although I have only glanced at the ruling (it was just yesterday), it provides some very interesting insights into how multiple IQ test scores across many years and different instruments are treated by different psychological experts and the court.  An interesting twist in the appeal was an attempt to make the case that scores from a new version (WAIS-IV) of a previously administered WAIS-III constituted "new evidence" that should be used in the MR/ID determination.  This brought up the issues of practice effects and the Flynn Effect...and the very interesting issue of how to treat scores on a new revision of a previously administered (but older) IQ test.  The discussion of how to evaluate the old and new WAIS scores (and old vs new norms) by the different experts is interesting, and in some places troubling.

I find it amazing that two of the psychological experts did not know the correlation between the WAIS-III/WAIS-IV as reported in the WAIS-IV technical manual.  One admitted not knowing the correlation and another suggested it was probably in the "mid .8."  The correct value, as reported in the WAIS-IV technical manual is .94----a point accurately made by the states psychological expert.  One would think that psychological experts testifying in Atkins cases would review the latest available research re: new instruments about which they are to testify...esp. the information in the technical manuals.  On this point the states expert was much more credible.

Also of interest was the introduction of the controversial 2009 Flynn article where Flynn suggests that the WAIS-III test norms are flawed.  Weiss (2007) has challenged the basis of Flynn's claim, and in an article I have "in press" (McGrew, in press; The Flynn Effect and Its Critics: Rusty Linchpins and “Lookin’ for g and Gf in Some of the Wrong Places”, Journal of Psychoeducational Assessment---special issue on the Flynn effect due to be published Dec 2010 or early 2011) I concur with Weiss's arguments.  Issues surrounding the Flynn effect are starting to heat up among IQ scholars.

Finally, the defense argues that the newer norms of the WAIS-IV make it a more accurate measure.  On this point I agree.  They then further argue that the change in the factor scores from the WAIS-III/WAIS-IV, which results in a different interpretation system of major component scores, makes the WAIS-IV a better indicator of intelligence.  Although I agree with the fact that the WAIS-IV factor-based index score system is a more valid system for interpreting cognitive strengths and weaknesses, this argument overlooks the more important central IQ score comparison issue..which is at the core of MR/ID Dx in Atkins cases.  Namely, as I have reported previously, despite the new organizational structure of the WAIS-IV, the ability composition of the FS IQ score (which is the best indicator of general intelligence -- baring concerns for the validity of the FS score and the need to look at part scores) is nearly identical across editions.

Using the consensus CHC model of intelligence as the cognitive ability taxonomy for understanding what IQ tests measure, CHC analysis of the FS IQ scores reveals that the WAIS-III/WAIS-IV FS IQ scores are not dramatically different in CHC ability representation (Gc- 31.8 v 30 %; Gv - 22.7 v 20 %; Gs - 18.2 v 20%; Gq - 4.6 v 5 %; Gsm - 13.6 v 15 %; Gf - 9.1 v 10%).  The high .94 WAIS-III/WAIS-IV correlation reported in the TM is not surprising given this CHC content comparative analysis.

Lots more to think about and discuss in the examination of all the IQ scores.  Maybe a future post...or, if anyone is interested, a good opportunity for a guest post.

I think there is a good chance that this case may try to get before SCOTUS   Correction----this case needs to go to the FL Supreme Court first.

Technorati Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,