Response to Mathias and Nettelbeck on the structure of competence: need for theory-based methods to test theory-based questions.
Use confirmatory factor analysis, not exploratory, when you want to test a theory about how intelligence or skills are organized.
01Research in Context
What this study did
The authors re-examined an earlier claim about how intelligence is organized in people with intellectual disability.
They used confirmatory factor analysis to test Greenspan’s two-factor model against the single-factor view.
What they found
The two-factor model fit the data about as well as the single-factor model.
In plain words, both stories are still in the running.
How this fits with other research
Koegel et al. (1992) tried the same test earlier and could not back Greenspan’s hierarchy. The new paper says the method, not the idea, may be at fault.
Chiu et al. (2017) and Wang et al. (2010) later showed that confirmatory factor analysis can uphold layered models in ID samples when the right steps are followed.
Together, these studies point to a simple rule: pick the tool that matches your question. If you want to test a theory, use confirmatory methods.
Why it matters
When you choose an IQ or adaptive test for a client with ID, you are betting on a model of how skills hang together. This paper reminds you to check whether the test’s structure was proven with confirmatory methods. If it was not, keep a open mind about what the scores mean and look for newer tools that pass the stricter test.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pull the manual of your main assessment and check if confirmatory factor analysis backed its structure—if not, plan to supplement results with other data.
02At a glance
03Original abstract
In this paper, we respond to a 1992 study by Mathias and Nettelbeck in which, using exploratory factor analysis, they argued that the structure of intelligence in persons with mental retardation deviates from the model previously proposed by Greenspan. Applying the LISREL method of confirmatory factor analysis to Mathias and Nettelbeck's original correlation matrix, we found the Greenspan model (in which social and practical intelligence form separate factors) to be an equally plausible interpretation of their data as the single (Interpersonal Competence) factor found by Mathias and Nettelbeck. The fundings are discussed with respect to (a) the importance of using theory-based methods when addressing theory-based questions and (b) the role of social intelligence in the ongoing controversy over the meaning of mental retardation.
Research in developmental disabilities, 1996 · doi:10.1016/0891-4222(95)00043-7