300 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [vol. xv. 



of intelligence; that is, its correlation with other measures of intelligence of known validity. 

 (3) The range of intelligence measured. It was extremely desirable to secure tests which 

 would measure ability from the upper grades of mental deficiency to very superior levels of 

 intelligence. This criterion made it impossible to consider certain tests which have shown 

 themselves to be excellent measures of intelligence in somewhat restricted ranges. (4) Objec- 

 tivity of scoring. It was agreed that, if possible, the tests should be arranged so that the 

 responses could be scored by means of stencils. Otis at Stanford University and Thurstone at 

 the Carnegie Institute of Technology independently devised the stencil method of scoring 

 intelligence tests at approximately the same date (1915). Thorndike, however, had used it as 

 early as 1914 in the scoring of a reading test. Otis seems to have been the first to arrange a 

 battery of intelligence tests so that they could be scored exclusively by stencils. (5) Rapidity 

 of scoring. (6) Unfavorableness to coaching. It was proposed to select only tests which 

 could be made up in a large number of "forms" which would be entirely different in content 

 but equal in difficulty. (7) Unfavorableness to malingering. (8) Unfavorableness to cheat- 

 ing. (9) Independence of schooling. It was agreed that the aim should be to test native 

 ability rather than the results of school training. (10) Minimum of writing in response. The 

 aim was to secure tests which could be responded to in the main by underscoring, crossing out, 

 etc., as in the majority of the Otis tests. (11) Interest and appeal. Everything else being 

 equal, the more interesting test should be preferred. (12) Economy of time. 



Finally, 13 tests were selected which were regarded as capable of being adapted to meet the 

 above requirements more or less satisfactorily. These were as follows : 



(1) Vocabulary. — Three or four definitions stated for a given word; the subject to check 

 the best definition. 



(2) The Otis synonym-antonym test. — Words to be checked as same or opposite. 



(3) Opposites. — Writing the opposites of words, or selecting opposite from a set of response 

 words. 



(4) Analogies. — Otis, Bingham, and Thurstone form of the test, involving choice of four 

 responses. 



(5) Completion. — Trabue type of completion test; response either by writing the missing 

 part or checking one of three or four alternative responses. 



(6) Disarranged sentences. — The Otis adaptation of the Binet test. 



(7) Oral directions. — Adapted from Abelson, Otis, Woodworth, and Wells. 



(8) Information. — The subject to underscore the correct response, which occurs among four 

 alternative responses. (Form suggested by Wells, Bingham, and Whipple). 



(9) Practical judgment. — Adaptation of the Binet "comprehension question" test and the 

 Bouser "selective judgment" tests. The subject checks the best of four answers to a "com- 

 prehension question." 



(10) Arithmetical reasoning. — A series of problems ranging from easy to difficult, but in- 

 volving little beyond the four fundamental mathematical processes and simple fractions. 



(11) Number series completion. — This involves the completion of a series of numbers which 

 is made up according to some definite plan. The test was adapted from Miss Rogers' missing 

 number test. 



(12) Memory for digits. — The Otis form of the "memory for digits" test, arranged for 

 group use (each response number to be written in a square). 



(13) Number comparison. — Underscoring the largest and the smallest numbers in a column 

 of 10 numbers (proposed by Wells). 



Each of the above tests was rated on a scale of 1 to 5 for each of the 12 criteria by five 

 members of the committee (Haines, Terman, Wells, Whipple, and Yerkes). A composite 

 rating was then made for the 12 tests. The following tests were finally selected for trial: 



Test 1. Oral directions. 



Test 2. Memory for digits. 



Test 3. Disarranged sentences. 



Test 4. Arithmetical reasoning. 



