;so. 2.1 PSYCHOLOGICAL EXAMINING IN THE UNITED STATES ARMY. 



489 



in the total, whereas in table 145 only those men who actually took examination a are included. 

 The two tables should not be considered without reference to each other. They serve to 

 illustrate quite clearly the type of problem which arises because of the elimination of illiterates, 

 and the necessary consequent adjustments in the treatment of data. 



Figure 2 shows graphically the data of table 144 summarized by letter grades. 



ALL CAMPS 



13.9% 



20.4 %> 



\.Ai>ni 



52.4% 



1 1.0% 70% 



DEVENS 



66% 33^ 9G% 



59.3% 



12.5% -6-9£ 



DIX 



17.7 % 



91% 16.5% A5.&% 



67% 36% 



LEE 



74% 5 0% 14.4% 



56.6%= 



9.SJ. 4.9% 



TAYLOR 



ILLIT 



□ 



Fig. 2. Comparisonofcamps.showingpereentageineachlettergrade. 

 Dix, 10,936; Lee, 26,640; Taylor, 16.753. 



Examination o. White enlisted men: Total, 65,267; Devens, 10,9 



Table 144. — Percentage distribution of scores by camps — examination a — enlisted v\en (white), all cases, including those 



unable to take examination a because of illiteracy .' 



Devens, 

 Seventy- 

 sixth Divi- 

 sion. 



Dix, Sev- 

 enty-eighth 

 Division. 



Lee, Eight- 

 ieth Divi- 

 sion. 



Taylor, 

 Eighty- 

 fourth 

 Division. 



Allfour 

 camps. 



350-414 



300-349 



250-299 



200-249 



150-199 



100-149 



50-99 



0-49 



Eliminated as illiterate. 



Number of cases 



Median (Md.) 



Upper quartile (Qs) 1 . . 

 Range from Q3 to Md.' 

 Percentage A or B 



B 



C+ 

 C 



c- 



D 



E 



1.5 

 5.5 

 11.0 

 15.2 

 19.6 

 17.6 

 7.7 

 1.4 

 20.4 



10,926 



157 



227 



70 



18.0 



0.9 

 6.0 

 12.3 

 18.1 

 21.8 

 19.4 

 9.6 

 3.3 

 8.6 



10,936 



171 



234 



63 



19.2 



0.5 

 3.1 

 6.7 

 11.1 

 16.0 

 18.7 

 16.5 

 9.7 

 17.7 



26,640 



116 



189 



73 



10.3 



0.5 

 4.4 

 9.5 

 15.7 

 21.6 

 21.3 

 14.4 

 5.0 

 7.4 



16,765 



154 



216 



62 



14.4 



0.7 

 4.2 

 9.1 

 14.2 

 19.0 

 19.3 

 13.4 

 6.1 

 13.9 



65,267 



143 



212 



>3 



14.0 



1 Medians are computed on the assumption that the cases eliminated as illiterate would have appeared in the lower half of the distribution; 

 upper quartiles on the assumption that they would have appeared below the highest quarter. 



121435°— 21 32 



