Cox's Study of 300 (301) Eminent Geniuses born from 1450 to 1850, including Flynn Effect Calculations, listed alphabetically and by descending IQ

The Cox IQ data were taken from column 25 "Corrected IQ estimate"; "AII" of Table 12A in Volume II: The Early Mental Traits of Three Hundred Geniuses by Catharine M. Cox from Genetic Studies of Genius edited by Lewis M. Terman. Copyright 1926, Stanford University Press. (Data taken from the third printing January, 1959. I have seen no evidence that the Cox study was ever revised past 1926.)

The Intelligence Quotient scores are on the Stanford-Binet scale. The scores listed are based on biographical data (including school rankings, anecdotes, works written, etc.) from data up to 26 years of age (and corrected to counter a regression towards the mean).

According to Cox, "The correction attempted in the present report is a crude approximation...: it indicates a point below which the true IQ probably did not fall." (p. 52) and "The resultant approximations are probably in most cases still too low, and perhaps, in a few cases, a trifle too high. The final correction is thus no more than an approximation to a true score." (p.83).

With a study this old, the Flynn Effect has to be taken into account if one wants to realistically compare one's IQ score with that of the people on the list. IQ tests have had to be revised several times. The average score is always supposed to be 100. But as time passes, new generations start to get higher scores on the older tests. There are various hypotheses to explain this including improved nutritional and environmental factors. The Flynn Effect numbers are given on a 15 point standard deviation scale (the Stanford-Binet uses 16 S.D.). Depending on the developed country, there is a steady IQ gain of 5 to 25 points per generation (30 years), with a median of about 15 points. (James R. Flynn, Massive IQ Gains in 14 Nations: What IQ Tests Really Measure, Psychological Bulletin, 1987, Vol. 101, No. 2, pp. 171-191. For more details click here.) This means that perfectly average modern kids have gotten 'genius' scores on old IQ tests.

The gains are not just for average people. For example, looking at a Dutch 1952 IQ test given to 18 year old military recruits, 0.04% got scores of 150 and above in 1952, whereas in 1982, 2.27% achieved that level. The ratio is 57 to 1.

Now, specifically for the Stanford-Binet Test (in the U.S.), there has been a gain of about 0.3 IQ points per year. This is lower than the approximately 0.5 median gain for other IQ tests in other developed countries. Converted to the corresponding 16 S.D., the gain is 0.32. To make the correction we can calculate the effect from 1916 to the year 1986 (years of the original and current versions of the Stanford-Binet test). So the calculation is (1986-1916)*0.32 =22 IQ points, if rounded.