PAPERS OX PSYCHOLOGY AND EDUCATION 347 



Inasmuch as the quartile which learned the most 

 learned about two and one-half times as much as the 

 quartile which learned the least, the advantage is clearly 

 and decidedly with the more rapid learners. If it be 

 thought that the differences in the retention ratios of the 

 quartiles is significant, the idea is soon dispelled, for 

 repeating the experiment live times, each time with a 

 new group, showed very small average deviations for 

 each quartile from the average of all 24 ratios. The 

 average of all the rati' ; the average deviations of 



the four quartiles from this average is +2.3 for the 

 first quartile, — 0.3 for the second, for the third, and 

 —0.8 for the fourth. 



With the same group of subjects a much more difficult 

 selection, "The Policies of Labor Unions," was next 

 tried. It dealt with the theory of labor unionism, and 



aken from Eeadings in Economics by C. J. Bullock. 

 It contained about 900 words. The first ten lines of the 

 selection used are : 



~!ective Bargaining. 1. Its purpose. If the whole 

 body of workers of a given kind can be brought into the 

 union, so that the union can meet the employers as the 

 representative of the whole, the position of the worker 

 will be greatly strengthened. The fear that if he re: 

 to accept certain terms, another man will be employed 

 in his place is removed. His ignorance of the market 

 conditions will be partly remedied both thru the com- 

 bination of the knowledge of all the workers of the union, 

 and in some cases, by the broader outlook which the 

 union officials, partly or wholly exempted from daily 

 application to manual work, may be able to obtain. The 

 whole matter of bargaining can be put into the hands 

 of the most — 



Two changes were made in method. First, on account 

 of the greater length of the selection, the subjects an- 

 swered questions about it instead of reproducing it. 

 Secondly, each subject determined his own length of 

 learning time, knowing that the length of the time he 

 took was an element in detennming his score. The learn- 



:ores and retention scores of different subjects were 

 made comparable by calculating the amount learned and 



