398 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vol. xv. 



Table 58. — Showing correlation between -mental age and six tests of the individual examination series (r=0.75). 



For the first 83 of the above 114 cases the following Pearson correlations were found: 



Mental age and total score of A, B, C, D, E, G 0. 727 



Mental age and total score of A, C, D, E 634 



Mental age and test A (cube construction) 363 



Mental age and test B (clock test) 699 



Mental age and test C (cube imitation) 535 



Mental age and test D (maze) 582 



Mental age and test E (form board) 563 



Mental age and test G (orientation) 620 



Group examination a and total of A, B, C, D, E, G 715 



Group examination a and total of A, C, D, E 620 



It will be noted that the above correlations (all Pearson) are high except that for cube 

 construction. 



The method of combining the tests of the individual examination series into a total score 

 was arranged by Otis and is set forth in full in a supplement prepared by him for the Examiner's 

 Guide (see part I, p. 149.) It was not ready for use until the latter part of November — a fact 

 which strongly militated against the use of these tests in the critical period of the official trial 

 of methods. 



Section 2. — Adoption of the Y erlces-Bridges point scale and the Stanford-Binet scale. 



The correlations yielded by the tests of the individual examination methods indicate that 

 the 20 tests of the series could probably have been made the basis of at least two mental age 

 scales — one for literate Americans, the other for illiterates and foreigners. In fact the method 

 employed by Otis for combining the scores of the separate tests made it possible to use them 

 in any combination, and mental-age standards for special combinations could have been 

 established after a few months' use. The majority of examiners, however, preferred to use 

 scales for which norms were available. Moreover, experience showed that the danger of coach- 

 ing, which had been one of the chief reasons why special individual examination methods had 

 been prepared, was practically negligible for the class of subjects taking such examinations. 

 It was accordingly decided to substitute for the specially prepared series of tests the Stanford- 

 Binet scale, the Yerkes-Bridges point scale, and a series of performance tests for foreigners and 

 illiterates." 1 



The arrangement of these three scales for Army use was begun early in January, 1918, 

 by Dr. Bridges. Abbreviated instructions for the use of the point scale and the Stanford- 

 Binet scale were prepared, which consisted mainly of the actual words to be used by the exam- 



1 The Stenquist test was retained in the revised Examiner's Guide for use in special cases as a test of skill to supplement the regular individual 

 examination. It was not used, however, and there is nothing further to report regarding it. 



