Need for theory-based methods to test theory-based questions - Reply

Greenspan and McGrew have argued that confirmatory, rather than exploratory, factor analysis should be applied to Mathias and Nettelbeck's 1992 data in order to test Greenspan's model of adaptive intelligence. We do not agree with this argument and draw attention to inconsistencies in Greenspan and McGrew's reasoning, which, we believe, confirm our opinion that this is not a straightforward matter. Moreover; we question Greenspan and McCrew's assertion that their reanalysis provides strong support for Greenspan's model. an the contrary, although the factor structure reported by them is supportable, the extraction of a single factor representing general competence or general intelligence is, if anything, a more stable solution. Greenspan and McGrew have pointed to problems with the original data set, and we agree in part with their analysis of these shortcomings. Some difficulties are, however; inherent in the nature of the available test procedures. We believe that such difficulties are better addressed by further research rather than by applying different analytical procedures to existing data.