Haggerty: Studies in Arithmetic 



69 



The Length of Term. — Table II shows that the shorter term 

 rej>;ulars exceed those of the longer term in 8 out of 14 determining 

 cases in Wabash county, and in 8 out of 12 in Huntington county. 

 The same explanation applies here as in the previous paragraph. 

 This condition also is probably due to the teachers being "crowded 

 down" to the older subjects in the curriculum. 



County Superintendents Devricks and Funderburg have both 

 offered further explanations which have had important local 

 application in our inter-township studies.^ The two most im- 

 portant of these are the influences of close supervision and local 

 school interest. 



Influence of the Quality of Gradatioyi on the Class Score. — The 

 good gradation that results from efficient supervision is effective 

 not only in obtaining better results from the individuals, but may 

 affect the score of a class without influencing the scores of the 

 individuals in the class. Suppose, for instance, that several 

 children whose scores were equal to the fifth grade standard were 

 wrongly placed in the sixth grade. Here these scores would fall 

 below standard and would lower the sixth grade score. Were this 

 incorrect gradation a common thing thruout a school the grade 

 scores would be low. Were the gradation corrected the grade 

 scores would rise. An investigation showed this to account for 

 more than one-half the difference between two sets of scores, in 

 one system tested. This principle probably helps to account for 

 high scores in the eighth grade in some instances, since the 

 passing of the children from that grade is the only place the 

 superintendent gets a hand in gradation in many county systems. 



Supervision. — Applying the principles developed in the pre- 

 ceding paragraphs to the comparison of the results from Wabash 

 and Huntington counties we find two conditions in each county 

 that tend to make it stand the higher. For Huntington county 

 these are the shorter term and the greater per cent of district 

 schools. For Wabash county they are the better gradation re- 

 sulting from the township supervision system, and the definite 

 standards set for attainment in arithmetic from the first grade up. 

 Reference to Table II shows that the Wabash county regulars 

 exceed in 14 out of 16 determining cases, while Figure I shows that 

 the totals exceed in all 15 determining cases and that it has less 

 variability in 19 out of 29. We may conclude, then, that the 



'For the Inter-township study of Wabash county see Wabash County Teachers' 

 Manual, 1915-16, pp. 47-62. 



