708 



MEMOIRS NATIONAL ACADEMY OF SCIENCES. 



[Vol. XV, 



centage of men taking alpha, beta, or alpha and beta who received a D— rating, but no fur- 

 ther examination. In other words, these figures represent the percentage of men receiving 

 their filial rating on a group examination who should have been recalled for further examina- 

 tion. It will be observed that the percentages are large. In all 39.8 per cent of the men ex- 

 amined rated D— on group examination and were not recalled. Men recalled from alpha for 

 beta numbered 1,216, and the total number of men given individual examinations was 825. 

 This gives a total of 2,041 men who were recalled for further examination by reason of fadure, 

 or 8.6 per cent of the total number examined. In brief, then, about half of the negroes were 

 rated D— on the group examinations, and of those so failing about one-fifth were recalled 

 and four-fifths allowed to go without further examination. 



Table 224. — Number and percentage of negroes rating D— on group examinations who were not recalled for further exami- 

 nation (Groups IV and V). 



Camp. 



Custer 



Pevens 



Dix 



Dodge 



Funston . . . 



Gordon 



Grant 



Lewis 



Meade 



Pike 



Taylor 



Travis 



Upton 



Wadsworth 

 All camps.. 



Number. 



Alpha 

 only. 



(855) 



37 



(67) 



1 



(1, 372) 



582 



(521) 



84 



(1,264) 



664 



(327) 



206 



(884) 



32 



(68) 



(188) 



49 

 (160) 



58 

 (827) 

 206 

 (410) 



63 

 ,159) 



3S 

 (327) 

 176 

 ,429) 

 ,196 



Beta 

 only. 



(799) 



285 



(31) 



8 



(1,371) 



732 



(296) 



27 



(1,592) 



840 



(1, 769) 



919 



(1,084) 



495 



(29) 



(1,816) 



972 



(629) 



30S 



(1, 786) 



1,148 



(296) 



142 



(326) 



44 



(1,310) 



1,103 



(13,134) 



7,023 



Alpha- 

 beta. 



(157) 

 27 

 (14) 



(1) 



1 



(14) 



1 



(64) 



(95) 

 15 

 (226) 

 18 

 (15) 



(45) 



5 



(11) 



1 



0) 



1 



(374) 



43 



(119) 



(1,216) 



119 



Individ- 

 ual. 



(25) 



"(3i' 



(74) 

 "(6)' 



(219) 

 "(74)" 



(53) 

 "(295)' 



(71) 

 "'(h)' 

 "(825)' 



Total. 



(1,836) 



349 



(115) 



9 



(2,744) 



1,315 



(831) 



112 



(2, 920) 



1,509 



(2,191) 



1,140 



(2,268) 



545 



(118) 



(2, 268) 



1,026 



(874) 



367 



(2, 667) 



1,355 



(1,375) 



248 



(1,755) 



84 



(1,642) 



1,279 



(23,604) 



9,338 



Percentage. 



Alpha 

 onlv. 



Beta 

 only. 



Alpha- 

 beta. 



1.7 



100 

 7.2 

 7.8 

 15.8 

 8.0 



11 



9.1 

 100 

 11.5 



1.0 



Total. 



19.1 



7.8 

 47.4 

 13.5 

 51.6 

 52 

 24 



45.2 

 42.1 

 51 



18.5 

 4.8 

 78 

 39.8 



A comparative study of the negroes by camp was one of the objects of the Hollerith analysis. 

 The results of the sortings are given in tables 225 to 244. These tables give the figures for 

 Group IV and Group V and also for Groups IV and V combined. The results are given by letter 

 grade in tables 245 and 246 for Groups IV and V, respectively. In each table the first colu mn 

 gives the number of cases and the following ones the percentage of this number making the 

 different letter grades. The superiority of the camps of Group V over the corresponding camps 

 of Group IV is at once apparent. Group V includes only northern negroes, Group IV the general 

 run. When Group IV is considered alone, it seems that here the northern camps are somewhat 

 superior to the southern ones in negro intelligence. If we compare the percentage distribution 

 of each camp with the total percentage distribution, we may call a camp "better" if it has a 

 larger percentage of superior and a smaller percentage of inferior men than the total, and 

 "poorer" if the percentage of superior men is less than the total and the percentage of inferior 

 men greater. If such comparisons are made, we find that Camps Custer, Devens, Dodge, Grant, 

 Lewis, and Upton have distributions which may be considered "better" than the total distri- 

 bution; Camps Taylor and Travis have distributions which approximately correspond to the 

 total distribution; while Camps Dix, Funston, Gordon, Meade, Pike, and Wadsworth have 

 distributions which may be considered "poorer" than the total distribution. 



