no. 2.i PSYCHOLOGICAL EXAMINING IN THE UNITED STATES ARMY. 367 



the tests did not receive a trial. The following, however, were included in the beta trial series 

 though in modified form: Detection of similarities, mutilated pictures (picture completion), 

 the imitation test, cube analysis, digit symbol, and the position test (modified into the spot- 

 pattern test). 



Camp Devens. — The problem of devising a group test for illiterates and foreigners was 

 even more systematically studied at this camp than at any of the others. Three investigations 

 were reported. 



Investigation 1 : Three performance tests were devised and given, in addition to the Sten- 

 quist test, to 48 feeble-minded subjects in the State School for the Feeble-Minded, at Waverley, 

 Mass., and to 69 soldiers of less than fourth-grade literacy. The three tests were memory span 

 for digits (visual exposure) ; arithmetical reasoning (problems similar to those in the individual 

 series, test T, original examiner's guide) ; memory for designs (including the diamond, figures 

 a and b of test P, original examiner's guide, and the two Binet designs). 



The rank difference correlation between mental age and the three supplemental tests was 

 0.65 for the feeble-minded subjects. With the same subjects the best correlation that could 

 be secured for mental age and the Stenquist test, however scored, was 0.32. It was evident 

 that these three brief tests, requiring 15 or 20 minutes, were very much more useful than the 

 Stenquist test in identifying the feeble-minded. 



Investigation 2 : The same three supplementary tests and two additional ones were given 

 to 117 literate men of three companies. The subjects were mostly of superior ability, only 7 

 of the 117 testing below 100 in examination a. The two additional tests included were the 

 dynamometer test and a motor coordination test. In the latter a sheet containing a large 

 number of rather small squares was placed before the subject. The task required was to draw 

 a small circle in each square without touching any side of the square. Immediately after the 

 men had been given the above five tests they were given examination a. The correlations 

 with examination a were as follows: Dynamometer, 0.30; circles, 0.24; memory span, 0.43; 

 arithmetical reasoning, 0.62; designs, 0.55; composite of memory span, arithmetical reasoning, 

 and memory for designs, 0.68. The dynamometer and circle tests, as might have been expected, 

 proved to be unsatisfactory measures of intelligence. The other three tests again make a 

 reasonably satisfactory showing. 



Investigation 3 : A more important experiment than either of the former was arranged by 

 C. R. Brown. It will be noted that the tests just described, while perhaps suited to illiterate 

 English-speaking subjects, were not suitable for foreigners. Brown prepared a series of five 

 group performance tests of the pencil and paper variety which could be given without the use 

 of language and which required no writing in the response. The tests follow: 



Test 1. Maze test: Essentially the form of the maze test included in the final beta scale. 

 Time, 2£ minutes. 



Test 2. Picture sequence: Essentially as is shown on pages 242 and 243. Six out of 

 the 12 sets of pictures were contained in Brown's original series. Response by drawing 

 arrows to indicate the proper sequence of pictures. Time, 3£ minutes. 



Test 3. Series completion: This is in principle the same as the X-0 series test included 

 in the final beta scale. The problems, however, of which there were 11, were not quite identical 

 with those finally adopted. It will be observed that the test is an attempt to translate test 8 

 of examination a into terms that would make it serviceable for illiterates. Response, by mak- 

 ing crosses. Time, 2 minutes. 



Test 4. Line length: An adaptation of the Huey-Yerkes letter line test for group use. 

 Instead of letters, however, it utilized 10 pairs of geometric figures. The task was merely to 

 decide which figure of a pair required the greater length of line. Response, by checking. 

 Time, 1 minute. 



Test 5. Pictorial analogy: This test resulted from an attempt to translate test 9 of exami- 

 nation a into terms suitable for illiterates. Of the 18 sets of figures included in Brown's origi- 

 nal series, 17 sets were included practically unchanged in the beta trial series (Part I, pp. 244 

 and 245). Response, by checking. 



