The top correlates of high-range I.Q.

© October 2019 Paul Cooijmans

Introduction

In the fourth column is the weighted average of that correlate's correlations with high-range I.Q. tests (in all cases significant). The correlates are ranked by the absolute values of their average correlations. One may take the square root of this value as an indicator of the correlate's g loading.

For better understanding, one should realize that, a priori, there is no reason whatsoever why these correlations should deviate significantly from zero (but they do), and that for high-range tests themselves, this value — that is, the weighted average of the test's correlations with other high-range I.Q. tests — is mostly around .5 or higher.

The correlates

Rank Correlate n r
1 Observed associative horizon 421 .498
2 Observed behaviour 940 .357
3 P.S.I.A. Neurotic - Revision 2007 389 .314
4 P.S.I.A. Rational - Revision 2007 389 .269
5 P.S.I.A. True - Revision 2007 389 .229
6 P.S.I.A. Ethics factor - Revision 2007 389 .223
7 Educational level 3348 .196
8 Disorders (own) 3298 -.163
9 Sex (female = 0, male = 1) 4887 .155
10 Disorders (parents and siblings) 3251 -.143
11 P.S.I.A. Cruel - Revision 2007 389 -.139
12 P.S.I.A. Cold - Revision 2007 389 .130
13 Gifted Adult's Inventory of Aspergerisms 779 -.127
14 P.S.I.A. System factor - Revision 2007 389 .108

The personalia with correlations between .1 and -.1 are left out here; this concerns year of birth, mother's and father's educational level, Cooijmans' Inventory of Neo-Marxist Attitudes, and the remaining P.S.I.A. scales.