Austria 102This takes some calculation, since languages differ and since IQ is generally normed to average 100. In the table above, Britain is normalized to 100 and the other countries are adjusted to that scale. I don't know how accurate the result is, but the differences are surprising. It would be interesting to relate this to the Flynn Effect-- that in the United States, the average absolute score has risen substantially over the past 50 years. Is India, for example, in a pre-Flynn state, so we can expect a 15-point increase with economic growth? Is IQ like height in this respect?
Britain 100
France 98
Ireland 93
Norway 98
United States 98
Canada 97
South Korea 106
Japan 105
China 100
Indonesia 89
Ecuador 80
Egypt 83
Nigeria 67
Uganda 73
India 81
I had another thought. It is hard to get a representative sample in big, diverse countries such as India and China. Would it be better to try to estimate the average by looking at the upper tail? If everyone in the top 1% is tested--for college admission-- and the distribution is truly normal (a crucial assumption), then the distribution of the top 1% will tell us where the mean score is.
That does get at something very important and quite neglected in discussions of test scores: the variance. I'd like to know if IQ scores are indeed normally distributed (perhaps this is a normalization, of course, by design) and how the variances differ.
Maybe the Lynn-Vanhanen book discusses this-- sometime I'll have to do more than just leaf through it.
[in full at 04.04.26a.htm]
To return to Eric Rasmusen's weblog, click http://php.indiana.edu/~erasmuse/w/0.rasmusen.htm.