A comparison of the factor structure of processing speed for younger and older adults: testing the assumption of measurement equivalence across age groups.

The purpose of this study was to investigate measurement equivalence of processing speed measures for different age groups. A structural equation modeling approach was used to investigate a measurement model and the factorial invariance between younger and older adults on speed measures. The analyses concurrently examined whether speed-related abilities dedifferentiate with increasing age. One hundred and forty-four younger and 105 older adults completed 9 measures designed to assess motor speed, alphanumeric speed, and geometric speed. Results indicated that although the number of factors and the factor loadings were invariant across age groups, the interfactor correlations, the variance-covariance matrices, and the unique variances differed across groups. Furthermore, a second-order speed factor seemed to explain much of the variance in the 3 first-order factors, although this higher order factor accounted for slightly more variance among the older group than among the younger group. The results suggest that there is sufficient evidence of measurement equivalence on the current speed measures across the 2 adult age groups and, in addition, provide evidence of dedifferentiation.