CHAPTER 10. 



THE ASSIGNMENT OF LETTER RATINGS. 



Section 1. — Examinations a and alpha. 



The question of the nature of ratings to be assigned was not considered at length in the 

 Vineland sessions of the committee on examination of recruits, although various suggestions 

 were made. Among these were: (1) Classification into deciles; (2) classification into five 

 groups — very superior, superior, average, inferior, and very inferior. After the unofficial 

 trial in July and August, 1917, it was decided to use both methods of classification, and space 

 was accordingly provided on the individual record card for recording the decile rank to which 

 a man belonged, and in addition a letter grade, A, B, C, D, or E. On the recommendation of 

 Thorndike it was decided to assign letter grades on examination a as follows: 



The percentile classification was never employed to any extent, chiefly because of the lack 

 of satisfactory norms. 



The above fivefold classification remained in use until March 1, 19 IS, and gave the follow- 

 ing distribution of grades for the 56,140 literate white men who were given examination a in 

 the first four camps : 



When the use of alpha was begun in March, 1918, the following sevenfold classification 

 was substituted for the previous fivefold classification: 



This basis of rating gave an unsatisfactory distribution of grades, as is shown by the fol- 

 lowing percentage distribution for 128,747 literate white men of several draft quotas. 



In June, 191S, the following rating system, based on raw scores, was tentatively adopted: 



This gave a much more satisfactory distribution of grades, as will be seen from the follow- 

 ing percentage distribution for the 128,747 literate men mentioned in the preceding paragraph. 



421 



