"Intelligent" intelligence testing with the W IV Tests of Cognitive Ability #3: Within-CHC assessment trees - a Gf "tease"

I have decided to temporarily skip the planned third installment in this series, and provide a "tease" for a small fraction of the "intelligent" testing material I will be positing in this series.  I will post an introduction to "intelligent" intelligence testing is (as per Kaufman and as applied to the WJ IV COG/OL) after this tease post.

One feature of Alan Kaufman's "intelligent" testing with the Wecshler series has been the provision of supplemental test groupings--groups of tests that may measure a shared common ability, but a group that is not one of the test's published clusters or indexes.

I have developed what I call "Within-CHC domain assessment and interpretation trees" for all 7 CHC domains in the WJ IV COG.  I developed these assessment trees by reviewing and integrating the following sources of information.

•Close examination of the CFA results in the WJ IV Technical Manual (TM)

•Close examination of the EFA, cluster analysis and MDS results in WJ IV TM

•Additional unpublished EFA, CFA, cluster analysis and MDS (2D & 3D) completed post-WJ IV publication (across ages 6-19)

•Review of supplemental/clinical groupings for the WJ, WJ-R and WJ III (e.g., McGrew, 1986; 1984--my two WJ COG books)

•Extensive unpublished “Beyond CHC”  analysis of the WJ III data

•Theoretical and clinical considerations

•Suggested groupings by Dehn (2015), Miller (2014) & Proctor et al. (2015, ASB6)

Below is the within-Gf assessment tree.  Click on images to enlarge for clear viewing.

(Note.  Since making this original post, I have now added a tabular version of the above information below.  Also, a clean PDF copy of both images can be found here.)

The dark arrows with bold font labels designate the Gf clusters provided by the WJ IV.  You will see Gf, Gf-Ext, and Quantitative Reasoning.  The dashed lines suggest other tests that might be important to inspect when evaluating a person's Gf abilities.  Note the line from Gf-Ext to the Visualization test.  It is labeled Gf-Ext 4/Gf+Gv hybrid.  This label is not in bold, indicating that it is not a cluster with score norms.  Close inspection of all data analyses of the WJ IV norm data found the Visualization test tending to "hang out" or near the primary Gf tests.  Also, as reported by Carroll (1993), sometimes Gv and Gf tests frequently would form a Gf/Gv hybrid factor (it is well known that some times factor analysis has a hard time differentiating Gf and Gv indicators).  This grouping  suggests that examiners should look to see if the Visualization test is consistent with the other Gf tests....which may reflect more shared Gf variance than anything specific to the Visualization test.

Also notice the Quantitative Reasoning-Ext (RQ) supplemental grouping,  This suggests that if the Quantitative Reasoning score is either high or low, on should inspect the Number Matrices and Applied Problems tests from the ACH battery---they, at times, will "follow" the scores on the Quantitative Reasoning  cluster.

Finally, one set of CFA models in the WJ IV TM suggested a possible Gf-Verbal vs Gf-Quantitative split.  The Verbal Reasoning supplemental grouping consists of the Concept Formation, Analysis-Synthesis, Oral Vocabulary, and Passage Comprehension tests.  Below the is a section of the CFA results that support the possible Gf-Verbal and Gf-Quantitative distinction.  This information is in the WJ IV Technical Manual.  This information suggests that the TM can be your "friend."  It contains considerable valuable information regarding tests that are not part of a cluster, but that showed evidence of some shared variance with a possible published cluster, or new clinical supplemental test groupings I will present.

Relevant Gf broad and narrow definitions are below:

Fluid reasoning (Gf): The use of deliberate and controlled focused attention to solve novel “on the spot” problems that cannot be solved solely by using prior knowledge (previously learned habits, schemas, or scripts).  Reasoning that depends minimally on learning and acculturation.

• Induction (I): The ability to infer general implicit principles or rules that govern the observed behavior of a phenomenon or the solution to a problem.  Rule discovery.
• General sequential reasoning (RG): The ability to reach logical conclusions from given premises and principles, often in a series of two or more sequential steps.  Deductive reasoning.
• Quantitative reasoning (RQ): The ability to reason, either with induction or deduction, with numbers or mathematical relations, operations and algorithms.

Given that I know that people tend to not to devour technical manuals like I do, my assessment trees are aids that incorporate all of this information in visual-graphic form--saving you from having to extract this potential interpretation-relevant information from the TM.

Stay tuned.  Some of the within-CHC assessment trees suggest many more test groupings to consider for clinical interpretation (than this Gf example.)

I, Kevin McGrew, am solely responsible for this content.  The information presented here (and in this series) does not necessarily reflect the views of my WJ IV coauthors or that of the publisher of the WJ IV. 

Click on images to enlarge