The world’s leading expert on the Method of Correlated Vectors, Jan te Nijenhuis, has argued that this method can be applied to all sorts of measures of cognitive ability, including achievement tests. As such, I decided to extend my recent investigation by looking at the Jensen Effect on shared environment for achievement tests. Below is my quick analysis of Samuelsson et al., (2007). As can be seen, in this sample of 809 twin pairs across three different countries there was a substantial positive Jensen Effect on shared environment. Interestingly, consistent with virtually every other analysis done by the present author there was a strong anti-Jensen effect on e^2. Now, one would want to look at numerous studies before drawing any conclusions. That said, as noted to Meng Hu, repeated findings of a positive Jensen Effects on c^2 are rather problematic for many of the inferences, with respect to genes, drawn based on MCV. With regards to the race-IQ-genes debate, in this situation, a Jensen Effect on the group differences evidences either genetic or shared environmental differences. To distinguish between genes and shared environment one would have to resort to biometrically informed structural equation modeling where genes and shared environment are pitted against each other, for example, as done by Rowe and Cleveland (1996). Another way of putting this is that, insofar as c^2 correlates with g-loadings, biometric modeling is not redundant, since it tells us something that MCV can’t — that the magnitude of a group difference correlates specifically with the genetic loadings of measures.
Samuelsson, S., Olson, R., Wadsworth, S., Corley, R., DeFries, J. C., Willcutt, E. & Byrne, B. (2007). Genetic and environmental influences on prereading skills and early reading and spelling development in the United States, Australia, and Scandinavia. Reading and Writing, 20(1-2), 51-75.