August 25, 2005

Sex differences in progressive matrices

A new article by Paul Irwing and Richard Lynn suggests that men score approximately 4.6 IQ points higher than women in the Progressive Matrices test. From the article:
It has frequently been asserted that there is no sex difference in average general intelligence but that the variance is greater in males. In this paper we examine these two propositions by a meta-analysis of studies of sex differences on the Progressive Matrices among university students. We find that both are incorrect.


There are five points of interest in the results. First, the present meta-analysis of sex differences on the Progressive Matrices among university students showing that men obtain significantly higher means than females confirms the results of our meta-analysis of sex differences on this test among general population samples (Lynn & Irwing, 2004). The magnitude of the male advantage found in the present study lies between 3.3 and 5 IQ points, depending on various assumptions. Arguably the best estimate of the advantage of men to be derived from the present study is .31d, based on all the studies and shown in the first row of Table 2. This is the equivalent of 4.6 IQ points and is closely similar to the 5 IQ points found in the meta-analyses of general population samples previously reported.


Second, the Progressive Matrices is widely regarded as one of the best tests of Spearman’s g, the general factor underlying all cognitive abilities... Now that we have established that men obtain higher means than women on the Progressive Matrices, it follows that men have higher general intelligence or g.


Third, the finding that males have a higher mean reasoning ability than females raises the question of how this can be explained... Hence, the larger average brain size of men may theoretically give men an advantage in intelligence arising from a larger average brain size of 0.78 multiplied by 0.40, giving a theoretical male advantage of .31d = 4.7 IQ points. This is a close fit to the sex difference obtained empirically in our previous meta-analysis of the sex difference of 5 IQ points on the Progressive Matrices in general population samples, and of 4.6 IQ points on the Progressive Matrices, in the present meta-analysis of the sex difference in college student samples.


Fourth, a number of those who have asserted that there is no sex difference in intelligence have qualified their position by writing that there is no sex difference ‘worth speaking of’ Mackintosh (1996, p. 567), ‘only a very small advantage of boys and men’ (Geary, 1998, p. 310), ‘ no practical differences in the scores obtained by males and females’ (Halpern, 2000, p. 90), ‘no meaningful sex differences’ (Lippa, 2002), and ‘negligible differences’ (Jorm et al., 2004, p. 7)... These different proportions of men and women with high IQs are clearly ‘worth speaking of’ and may go some way to explaining the greater numbers of men achieving distinctions of various kinds for which a high IQ is required, such as chess grandmasters, Fields medallists for mathematics, Nobel prize winners and the like.


Fifth, the finding in this meta-analysis that there is no sex difference in variance on the Advanced Progressive Matrices and that females show greater variance on the Standard Progressive Matrices is also contrary to the frequently made contention, documented in the introduction, that the variance of intelligence is greater among males. This result should be generalizable to the general population of normal intelligence. The greater male variance theory may, however, be correct for general population samples that include the mentally retarded... The issue of whether there is greater male variance for intelligence in general population samples needs to be addressed by meta-analysis.
British Journal of Psychology (preprint)

Sex differences in means and variability on the
progressive matrices in university students: A meta-analysis

Paul Irwing and Richard Lynn

(no abstract)


No comments: