Why does it follow that this is the upper limit (of how variation in environmental factors shape scores on IQ tests)?  
 
Anyway, I'm confused, because first you argued that the study involved contained numerous methodological flaws--lack of randomization, endogeneity issues, and so forth--and then you assert that this is the upper limit of environmental influence. How does that follow?

So a less snide comment: 6-17%, particularly at the upper range, is still a fairly significant change in cognitive performance as measured by the tests in question. Given that the debate is not, at least among sane people, about whether cognitive capacity is heritable, but about the degree to which environmental factors can influence cognitive performance, this still seems like a fairly significant rebuttal of strong (in these relative terms) determinism.

Many, many apologies. I didn't hit the "read the rest of the post" link.

The study you link to suggests that Head Start is of limited effectiveness because other environmental factors eventually swamp its influence, therefore supporting David Kirp's argument that IQ manifestation is heavily influenced by socio-cultural, rather than innate, variables. Do you have a criticism of the study's research design or some other reasons why we should discount it?