332 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vol. xv, 



Table 14. — Distribution of mental ages for groups used in correlating the tests of examination a with mental age. 



The data from all but the last group above were supplied by Terman and his students at 

 Stanford University. For the most part the Binet tests had been given from one to two years 

 previously, but the mental ages were brought down to date on the assumption that the ratio 

 of mental age to actual age (intelligence quotient) remains constant. The Stanford data were 

 based chiefly on form A of the army test, but in group 6 forms B and D were used. The pro- 

 cedure was regular except in the case of group 5, which had double time in tests 3 to 10. The 

 correlations of the total score and of the separate tests with mental age are shown in Table 15. 



Table 15. — Correlations of total score and separate tests of examination a with mental age. 



1 Combined. s Not computed for separate tests. 



The data of table 15 yield the values of rank order of the tests shown in table 16. 

 Table 16. — Rank order of tests of examination a for different groups. 



As a measure of the traits which enter into a Stanford-Binet mental age, test 6 is plainly by 

 far the best test in examination a. Tests 9 and 4 rank next, 7 and 5 are intermediate, while 

 2 and 10 are decidedly lower. 



