|reply to grafenberg |
Actually, measured (by some test standard) intelligence of large groups fits the Normal distribution. The results are scaled so that the mean is 100 and the standard deviation is 15. In order for even the range between 70 and 130 to have meaning, thousands of samples are required. In this real case, half the population have an IQ of 100 or less, by definition.
The wider extremes are more difficult to evaluate. For example, Marilyn vos Savant's purported IQ of 190 or thereabouts is six standard deviations above the mean. There aren't enough samples tested by any standard test to establish that value from the test population. I suspect her IQ was extrapolated by scaling MENSA test results from its members IQ distribution.