Still spurious: A comment on attempts to revive cognitive ability tilts

TL;DR

This paper argues that differences in cognitive abilities, like math versus verbal skills, are not meaningful when predicting other variables, as these correlations are misleading.

Contribution

The paper refutes claims that cognitive ability tilts have incremental validity by showing these findings are also spurious.

Findings

Correlations between ability tilts and other variables are spurious due to associations with the original measures.

Incremental validity of tilts is shown to result from the validity of the original measures, not the tilt itself.

Variance in 'super-residualized' tilts linked to shared environments is spurious due to adjustment for identical variables in twin studies.

Abstract

Ability tilts are within-individual differences between scores on two ability measures, e.g., math – verbal ability. We have shown in a series of reports that correlations between tilts and other variables are spurious consequences of associations to the constituent variables. Recently, Woodley of Menie et al. suggested that findings of incremental validity of tilts, over and above one of the constituent variables, refuted our claims of spuriousness. However, we show here that incremental validity of tilts are spurious consequences of incremental validity of the constituent variables. Moreover, Woodley of Menie et al. presented new results where so-called “tilt super-residuals” were attributable to shared environmental factors and they concluded that this finding confirmed a hypothesis that individuals specialize with respect to cognitive niches as an effort to adapt to stable environmental factors, alternatively do not specialize in the case of an unstable environment. However, we show that variance on “super-residualized” tilts attributable to shared environmental factors is a spurious consequence of adjusting for a variable (e.g., age) that is identical within twin couples. In summary, findings involving ability tilts still appear to be spurious.

Figures