no. 3.] PSYCHOLOGICAL EXAMINING IN THE UNITED STATES ARMY. 835 



In the case of both 24g and 6g we have significant differences in two of the three com- 

 parisons. In the case of 27h we have a significant difference between apprentices and journey- 

 men. In the case of 23t we find that none of the differences between any of the pairs of degrees 

 of skill are significant. It would seem that the function of intelligence plays a varying part 

 in its relation to degrees of skill as classified by trade tests. For example, in the case of 24g we 

 seem to find a difference only between the apprenticeship level and the higher levels of skill. 

 In the case of 6g we find the difference only between the expert level and the two lower levels. 

 In the case of 23t we find no significant differences between levels of skill, although intelli- 

 gence is demonstrably a factor in qualifying in the apprenticeship level of that trade. 



To secure a single measure representative of the degree of relationship between intelligence 

 ratings and degree of expertness as measured by trade tests Pearson's coefficient of mean square 

 contingency was utilized. Results were as follows: 



Occupation. 



6g 

 23t 

 27h 



Auto repairman 



Machinist 



Heavy track driver 

 Horse hostler 



0.265 

 .254 

 .130 

 .2766 



Number 

 of cases. 



451 

 644 

 296 



The above coefficients indicate that there is a slight degree of relationship existing between 

 intelligence and degree of expertness measured by trade tests, particularly oral trade tests. 

 The trade tests for heavy truck drivers is a performance test, and one would hardly expect 

 intelligence as measured by Army intelligence tests to show any relationship to a nonlanguage 

 performance test. It is not surprising, therefore, to find the coefficient to be only 0.13. How- 

 ever, to qualify in the trade does involve the intelligence factor. This is shown by the fact 

 that those passing the trade test for truck drivers are significantly more intelligent than those 

 who claimed trade ability and were so classified by personnel interview. 



Not enough data are available to determine in any satisfactory way the interrelations of 

 intelligence, skill, and expertness in various trades. That the relative importance of these 

 factors should vary from trade to trade is probably to be expected and is probably dependent 

 upon a wide variety of causes involving the nature of the specific trade and the types of people 

 who gravitate into that trade. 



Enough evidence is present concerning the influence of intelligence upon the results of 

 trade tests to warrant the use of intelligence tests in conjunction with trade tests. Such use 

 would be based on the principle that, assuming equal ability of two or more persons as measured 

 by a trade test, the more intelligent should be selected for placement. 



A further comparison between the results of the studies made in Camp Wadsworth, Camp 

 Devens, and the Surgeon General's Office brings out in detail the fact that the relative ranking 

 of various occupations by the alpha examination is very similar in the studies mentioned. 

 Tables 381, 382, and 383 give the several comparisons. To summarize the results of these 

 comparisons we have the following correlation values: 



Wadsworth study v. Devens study +0. 90 



Devens study v. Surgeon General's study + .88 



Wadsworth study v. Surgeon General's study + .98 



