Validity of Greenspan's models of adaptive and social intelligence.
Greenspan’s social-intelligence pyramid wobbles when you switch from exploratory to confirmatory stats—check the math before you bank on it.
01Research in Context
What this study did
The team tested Greenspan’s idea that social smarts sit in a pyramid. They gave teens with intellectual disability a set of tasks. Then they ran a factor analysis to see if the scores stacked the way Greenspan drew it.
What they found
The numbers did not line up with the pyramid. One factor called Practical-Interpersonal Competence showed up, but its meaning stayed shaky. The study called the model “not supported.”
How this fits with other research
Four years later Festinger et al. (1996) re-ran the same numbers with a stricter test. They say the two-factor model fits just fine, so the 1992 “no support” verdict may be too harsh.
Wang et al. (2010) and Chiu et al. (2017) show the tool matters. When they used confirmatory factor analysis on quality-of-life scales in ID samples, the hierarchy held.
The clash is about method, not truth. Exploratory analysis says “no structure,” confirmatory says “structure possible.” Pick your stats, pick your story.
Why it matters
Before you buy an IQ or social-skills test that claims “hierarchical,” flip to the stats page. If the authors only ran exploratory factor analysis, take the pyramid picture with a grain of salt. Ask for confirmatory evidence or collect your own. Your assessment choice shapes goals, funding, and where kids get placed.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pull the last adaptive test you used and see if the manual lists “confirmatory factor analysis” in the validity section—if not, plan a pilot with a different tool.
02At a glance
03Original abstract
Study 1 assessed the construct validity of Greenspan's 1979 and 1981 models of adaptive and social intelligence. Seventy-five adolescents with mental retardation completed four measures of conceptual intelligence, a general measure of adaptive behavior (practical intelligence), and seven measures of social intelligence. A factor analysis of all measures yielded three factors that were obliquely rotated and labelled Practical-Interpersonal Competence (defined by a subset of social intelligence and adaptive behavior measures). Verbal Intelligence, and Accuracy of Inference. An analysis of the seven measures of social intelligence found no evidence to support the hierarchical organization of variables that is proposed in Greenspan's model of social intelligence. The criterion validity of the Practical-Interpersonal Competence construct identified in this study was then assessed in Study 2 (N = 20). Factor scores were calculated from nine measures that were selected to represent the Practical-Interpersonal Competence, Verbal Intelligence, and Accuracy of Inference factors, and correlated with three criterion measures of practical and interpersonal competence. Two of the three criterion measures validated both the Practical-Interpersonal Competence and Verbal Intelligence Factor scores but did not discriminate between the two. The criterion validity of the Practical-Interpersonal Competence construct, therefore, has yet to be established.
Research in developmental disabilities, 1992 · doi:10.1016/0891-4222(92)90019-3