| The present study attempted to demonstrate the stability and construct validity of second-stratum factors of ability. A review of the major factor analytic theories of ability structure revealed that there is general agreement at the level of primary mental abilities, but that the number and nature of more general factors, such as Fluid and Crystallized Intelligence, are matters of some debate. By attempting to replicate the second-stratum factor structure reported by Hakstian and Cattell (1978) based on the Comprehensive Ability Battery and by relating the second-stratum factors to other tests purportedly assessing the same abilities, evidence concerning the stability and construct validity of the second-stratum factors was obtained. A sample of 272 first-year college students were given the 20-test Comprehensive Ability Battery as well as a second set of 10 measures that should have helped clarify the nature of the second-stratum factors. An analysis of sex differences on the Comprehensive Ability Battery clearly replicated previously reported results using the battery (Hakstian & Cattell, 1975), but a factor analysis of the battery arrived at a rather different structure than reported by Hakstian and Cattell (1978). Specifically, only five, rather than six, factors emerged from data from the present study, and the rotated factors showed rather low levels of congruence with the Hakstian and Cattell factors, with coefficients of congruence ranging from .651 to .867. In addition, a Procrustean rotation failed to achieve an acceptable match with the Hakstian and Cattell factors. The 10 additional tests, when projected into the five-factor space, produced loadings only moderately in line with hypotheses, and sex differences on the 10 tests provided only moderate support for the differences revealed on the 20-test battery. The results of the present study indicate that, while sex differences on primary mental abilities may be rather stable, evidence has yet to be presented that second-stratum, more general factors of ability are stable over different populations of persons or over different batteries of measures supposedly measuring the same factors. Some implications for theories of ability structure and suggestions for future studies were discussed. |
