Statistical inference and spatial patterns in correlates of IQ
What is it about?
A number of previous studies had found apparent relationships between international variation in IQ between countries and the characteristics of those countries. Some relationships were fairly innocuous (e.g temperature) but others were more provocative (e.g. race). A recent study had suggested that international variation in IQ might be driven by parasites, with countries with more parasites having a lower average IQ due to the strain that parasites place on brain development. However, a major problem with these kinds of studies is that you have a lot of countries which are treated as though they were completely separate when they are in fact very similar, especially if they are very close together. This can introduce a problem known as "spatial autocorrelation", which interferes with statistical analysis. We applied some common controls for this problem and produce a new set of results. These results suggest that parasites are one of the leading contenders as a potential cause of variation in IQ, but we can only demonstrate correlation and not causation without further research.
Why is it important?
We noticed a fundamental problem with statistical analyses in a previous study of geographical variation in IQ. Correcting for this problem allowed us to test more rigorously the findings of a previous analysis and propose our own interpretations for the patterns observed. While we have demonstrated correlation, we cannot comment on causation apart from to say that a plausible mechanism exists by which disease could affect cognitive development.
The following have contributed to this page: Dr Christopher Hassall