no. 2.] PSYCHOLOGICAL EXAMINING IN THE UNITED STATES ARMY. 307 



Test D. Maze test. — The more difficult mazes of the Porteus series ' were modified by 

 Yerkes. Data from Porteus have correlated well with Binet mental age. 



Test E. The Dearborn form board. 2 — This form board was selected because it offers a 

 graded series of problems with reasonably wide range. The instructions were prepared by 

 Yerkes. 



Test F. Stenquist construction. — This was not included in first series of tests. 



Test G. Orientational information . — This was included primarily for use with suspected 

 psychotics. It was arranged by Yerkes and Terman and was not intended as an intelligence 

 test. Four of the first five items are from Binet, the last five from Terman's series. 3 



Test H. Association. — Kent-Rosanoff series. 4 The instructions were prepared by Wells. 



Test J. Vocabulary. — Word lists as arranged by Terman. Those of the first series were 

 taken from the Stanford vocabulary test, 5 which had been made up by selecting every one 

 hundred and eightieth word in the Laird and Lee Vest Pocket Dictionary. The forty words 

 for series 1 of test J were selected from the hundred-word Stanford vocabulary test as follows: 

 In the case of 600 adults who had been given the Stanford-Binet, success in each of the 100 

 words was correlated with success in the scale as a whole. The 40 words selected for the present 

 purpose were those which had shown highest correlation with Stanford-Binet mental age. The 

 diagnostic value of this first list is accordingly very high. 



The other four series included in test J were selected thus: Four vocabulary lists of 100 

 words each were made up from the Laird and Lee Vest Pocket Dictionary by selecting every 

 one hundred and eightieth word, as in the case of the original Stanford vocabulary test. Each 

 hundred-word list thus obtained was reduced to 80 by omitting those which were either too 

 easy or too hard to have diagnostic value in the range of intelligence ordinarily found among 

 soldiers. The lists of eighty were then reduced to forty by dropping out alternate words. 

 Thus far the words of series 2 to 5 were in the main "unselected," in the sense that they were 

 selected by arbitrary rule. However, several changes were later made by substituting for words 

 which were deemed unsuitable other words judged to be equally difficult. By "unsuitable" 

 is meant words which tend to bring ambiguous responses, words which are more or less provin- 

 cial or obsolete, words with objectionable references, etc. 



The words of series 1 were arranged in order of difficulty as shown by tests of 600 adults. 

 Those of series 2 to 5 were arranged in order of difficulty as estimated by three judges. The 

 relative difficulty of the five series is not definitely known, but the differences are probably not 

 great. 



Test K. Letter line test. — From the Yerkes-Rossy Point Scale. 8 Yerkes credits this test 

 to the late E. B. Huey. The test was adapted for army use by Yerkes and Melville, who pre- 

 pared the instructions and the alternative series. 



Test L. Dissected sentences. — From Binet. Alternative series and instructions were 

 prepared by Melville. 



Test M. Absurdities. — From Binet. The five series and the instructions were prepared 

 by Terman, assisted by Goddard. The 50 absurdities of the five series have the following 

 sources: Binet or various editions of the Binet scale, 9; Simpson's series, 6; Goddard, 9; Terman, 

 24; unknown origin, 2. The 24 absurdities furnished by Terman consisted in the main of 

 verbal "translations" of the Terman and Williams absurd pictures. 7 



Test N. Controlled association (Rhymes). — From Binet. The instructions and the five 

 series were arranged by Melville. 



i Porteus, S. D. Mental Tests for Feeble-Minded: A New Series. J. Psycho- Asthenics. , vol. 19, No. 4, 1915, pp. 200-213. 



2 Dearborn, Walter F., Anderson, J. E., and Christiansen, A. O. Form Board and Construction Tests of Mental Ability. J. Educ. Psychol., 

 vol. 7, No. 8, pp. 448-449. 



3 Terman, L. M., and Chamberlain, M. B. Twenty-three Serial Tests of Intelligence and Their Intercorrelations. J. Applied Psychol., vol. 

 2, No. 4, 1918, pp. 343-344. 



« Kent, G. H., and Rosanofl , A. J. A Study of Association in Insanity. Amer. J. Insanity, vol. 47, 1910, pp. 37-96 and 317-390. 

 a Terman, I». M. The Measurement of Intelligence, 1916, pp. 224-231. 

 » Boston Medical and Surgical Journal, April, 1917, vol. 176, pp. 564-573. 



' Terman, L. M., and Chamberlain, M. B. Twenty-three Serial Tests of Intelligence and Their Intercorrelations. J. Applied Psychol., vol. 

 2, No. 4, 191S, pp. 347-349. 



