The key value required to convert SAT scores onto a scale analogous to IQ scores is a White population SD expressed in SAT points. The full list of key values needed to estimate the mean IQs of Asian and White candidates who sat the SAT in 1971-1972 is as follows: the percentage of the population cohort the candidate group represents; a correlation coefficient between taking the SAT and IQ; the mean IQ and SD of the population cohort or at least the general population. An example may illustrate how these values would help. Assume the top 27.5% of Whites took the SAT and that the correlation between taking the SAT and IQ was perfect. Then the mean IQ of the candidates would be 1.215 SDs above the population mean, or 18 IQ points above the population mean (1.215 x 15 = 18), or an IQ of 118 (100 + 18 = 118). The correlation is not perfect, of course. But with a correlation of say .60, multiplying that times 18 would give 11 points to be added to the population mean, and would give Whites who sat the SAT a mean IQ of 111. I will derive the SAT key values in the order in which they have been listed.
The first, White population SDs expressed in SAT points, comes from a representative sample of high school seniors who were given the SAT in 1966. They had SDs of 126 for the SAT-M and 120 for the SAT-V. However, these high school seniors were an elite group because at that time, only about 82% of the relevant age cohort reached the 12th grade. Computer projections suggest values for the general population of 139 and 129