42 



MASS. EXPERIMENT STATION BULLETIN 310 



A large proportion of the experimental children were from foreign homes and 

 started to school with a limited knowledge of English. The difificulty which 

 some of the children had to even understand the directions given was a handicap, 

 particularh' in the first test. At the beginning of the second year some had not 

 >et mastered the language so that they read as well as the English-speaking 

 children, but at the end of the year the difference was less obvious. It is probable 

 that the greater increase in the scores of the first- and second-grade children who 

 received the lunches was in part due to the overcoming of the language difificulty 

 of these foreign children. However, this does not appear to be the only expla- 

 nation, since the mean increases in the scores for the first and second grades in 

 school D during 1932-33 when they received a lunch were higher than those for 

 the same grades in the same school the preceding year when the children received 

 no feeding. It may be that these little folks were less fatigued and therefore able 

 to appK' themselves better as a result of the mid-morning lunch. 



Table 14. — School Progress from October to May as Indicated by the 

 Mean Increase in Total Scores from Achievement Tests 



Classified bv School Grades 



Grade 



Experimental 

 Group 



Control 

 Group 



Mean Mean 



Number Increase Number Increase 



Scores 



Scores 



Classified bv Mental Levels 



Ranee 



of 

 I. Q.-s 



Experimental 

 Group 



Control 

 Group 



Mean Mean 



Number Increase Number Increase 



Scores 



*Otls Group Intelligence scale used. 



tPintner-Cunninghani mental test used for first, and New Stanford achievement tests for second 

 to eighth grades. 



It was possible that children at difTerent mental levels might not react to the 

 tests in the same manner, the brighter ones showing more improvement at the 

 end of the year. The children were therefore divided into groups according to 

 their I. Q.'s irrespective of age or grade. Some difference in the mean increase 

 in total scores was found at most of the mental levels in favor of the experimental 

 group, but was so slight, particularly when the achiexemenl tests were used, as 

 to be only suggestive. (Table 14.) 



