yields a significant negative correlation with age (rg,age= –.14), consistent with
the existing literature on age and intelligence.
For the most part, the knowledge domains were well represented in the
academic world, and thus could be considered to give an advantage to
younger adults, in comparison with middle-aged adults, who would be fur-
ther removed from the academic environment. Nonetheless, scores on only
three knowledge domains were significantly negatively related to age—they
were all in the sciences domain (physics, chemistry, and biology). Ten of the
remaining domain knowledge tests were significantly positively related to
age, meaning that middle-aged adults performed, on average, better than
the younger adults. If we average across all of the 18 knowledge domains,
there remained a significant positive correlation between age and perform-
ance (raggregated domain knowledge,age= +.19). To be consistent with the notion that
typical performance (reflected by Gc and domain knowledge scores) is a
more important component of adult intelligence than Gf (or intelligence-as-
130 ACKERMAN AND KANFER
FIG. 5.2. Scatterplots of data from Ackerman (2000). The scatterplots each show
age (the abscissa) andz-score (ordinate) for fluid intelligence (Gf ), crystallized intelli-
gence (Gc), an aggregated score across all of knowledge domains, and an equally
weighted composite of Gf, Gc, and the aggregated knowledge score. Symbols repre-
sent each participant.N= 228.