no. 2] PSYCHOLOGICAL EXAMINING IN THE UNITED STATES ARMY. 541 



Table 197. — Rank order of tests of examination a according to average intercorrelation with other nine tests. 



Test. 



Unsclected groups. 



Dix 



unselected 

 (895 cases). 



Dix Field 



Artillery 



(172 cases). 



National 



Guard i 



(313 cases). 



Combined 



order for 



unselected 



men. 



Fort Snell- 

 ing reserve 

 office rs' 

 training 

 camp (213 

 successful 

 candidates.) 



High-grade groups. 



University 



of 

 Minnesota 

 (201 male 

 freshmen). 



University 



of 

 Minnesota 

 (209 female 

 freshmen). 



Combined 

 order for 



high-grade 

 groups. 



1. Oral directions 



2. Memory span 



3. Disarranged sentences 



4. Arithmetical problems.. . 



5. Information 



6. Synonym— antonym 



7. Practieal judgment 



8. Number series completion 



9. Analogies 



10. Number comparison 



6 

 10 



7 



3 



1 



2 



4.5 



9 



4.5 



4 

 10 

 6.5 

 2.5 

 1 



2.5 

 5 

 9 

 6.5 



4.5 

 10 

 7 

 3 

 1 

 2 

 6 

 9 

 4.5 



7.5 

 7.5 

 2 

 6 

 3 

 4 

 10 

 5 

 1 



7 



8 



4 



5 



3 



2 



9.5 



6 



1 



9.5 



i Reported on p. 316. 

 Table 198. — Correlation of each test of examination a with total weighted score. 



A. DRAFT GROUPS UNSELECTED. 



B. SPECIAL GROUPS, SUPERIOR TO DRAFT GROUPS. 



Section 4. — Practice effects in examination a. 



The problem of the effect of practice in connection with the Army examinations has re- 

 ceived little attention. A certain amount of data was available on examination a, secured for 

 the most part either through chance or in connection with the problem of reliability, but none 

 were obtained on alpha and beta. For the Army situation the question of practice effect was, 

 in fact, very nearly negligible, since pressure of demands made it imperative that there be the 

 minimum of duplication of work. Occasionally men received from other camps or transferred 

 within a given camp were reexamined, but this was not a sufficiently usual occurrence to affect 

 the mass results and the comparison of groups. Moreover the results of the first examination 

 were entered on the man's qualification card and frequently on his service record, and were 

 therefore available. 



The chief difficulty arose in connection with officers and candidate officers. The former 

 were fairly frequently reexamined on transfer to a new camp; the latter were even more likely 

 to be given a second examination, receiving their first with the group of enlisted men and their 

 second with the officers' training camp group. In both cases it was possible to obtain the man's 

 first record from his qualification card. It was only a question of taking the time to look this 

 up, which was a fairly time-consuming task. The data presented herewith show that it was a 

 matter of importance that second examination be not used interchangeably for first. The 

 sources of data available on the effect of repetition of examination a are as follows: 



Cases. 



Enlisted men, Devens and Dix 190 



Third officers' training camp, Devens 331 



Third officers' training camp, Taylor 225 



Officers, Devens 202 



School children, Oakland, Calif 324 



