Discover more from Just Emil Kirkegaard Things
Simple reaction times: decline since Galton, and consistenly favor men
Some time back I read this post http://charltonteaching.blogspot.dk/2012/02/convincing-objective-and-direct.html pointing out that simple reaction times have increased since the late 1800's when Galton first measured them. Since simple reaction times are known to be g-loaded, it would seem to indicate that g has fallen since then. This obviously fits with dysgenic ideas of Richard Lynn and others. So I came across:
Woodley, Michael A., Jan te Nijenhuis, and Raegan Murphy. "Were the Victorians cleverer than us? The decline in general intelligence estimated from a meta-analysis of the slowing of simple reaction time." Intelligence (2013). were-the-victorians-smarter-than-us
which discusses the idea. However, I also noted that men always averaged a lower RT than women. Possibly due to a higher male g. The study did not mention the SD's, so I looked up the cited meta-analysis from 2010:
IRWIN, W. SILVERMAN. "Simple reaction time: It is not what it used to be." American Journal of Psychology 123.1 (2010): 39-50. simple-reaction-time-not-what-it-use-to-be
And then I sent this email to the authors:
I read Were the Victorians cleverer than us? The decline in general intelligence estimated from a meta-analysis of the slowing of simple reaction time with interest, for I too had read Bruce Charlton's blogpost about the idea. From glancing the table of studies it occurred to me to check to see if one could see a sex difference. As far as I know, no one has used reaction time tests to examine sex differences in g. From looking at the tables in the paper, one can see that they are all in favor of men (men have lower reaction times). However, the SD's of the studies are not mentioned thus not permitting a test of significance.
I therefore looked up Silverman (2010)'s paper Simple reaction time: It is not what it used to be, which fortunately did mention the SD's of the studies, and even supplied 95%CI's. From glancing over them, one can see that in the 7 studies that include both men and women, in 6 of them do the 95%CI's not overlap. This seems therefore highly unlikely to be a chance finding, but would seem to have to result from either 1) inadequate representation (of bright women, or dull men), or 2) because men really do average lower reaction times.
Which brings me to the next question: is the relationship between g and reaction times the same for men and women? It might be that reaction time is a biased estimator for g with respect to men and women, in that a men with a reaction time of n and a women with reaction time m both correspond to the same g even while n<m. Does anyone know? Unfortunately we cannot ask Jensen anymore.
If it isn’t a biased indicator, it would seem to be new evidence that men average a higher g than women.