58 BASES FOR CURRICULUM MAKING 



(e) Study of the business essentials in general farming. 



(f) Managerial rating sheets or score cards. 



(g) Scales for measuring skills, managerial ability, etc. 



(h) Use of case, unit, project, problem and other meth- 

 ods in various aspects of the instruction work. 



(i) Standard record cards for grades and junior high 

 schools for vocational direction data. 



The Reliability of the Data. 



So far it has been assumed that the ratings as given on 

 the various Form B's (the original data of the study) were al- 

 ways correct. As a matter of fact, how- 



The method of rank- ever, it is probable that there are many 

 ing individuals for misplacements of men in the groups. If it 

 certain qualities or had been possible to obtain four, five or 

 characteristics may six separate student ratings on each 

 be open to question group it is probable that there would have 

 as to its reliability. been differences more or less marked. 

 Moreover, the ability of students to act as 



judges may be questioned. It may be very legitimately con- 

 tended that the only way to tell how thirteen farmers should be 

 arranged in order of financial success from best to poorest would 

 be on the basis of information obtainable by the usual farm man- 

 agement survey methods. Even this could be criticised and is 

 being criticised, especially if the criterion used be the labor in- 

 come criterion, which has been the basis for most surveys so 

 far undertaken. 



But, granting the validity of the labor income criterion, 

 what would be the possibility of its use in such a study as this? 

 At first plans were made to tie up the determination of the char- 

 acteristics and qualities desired to the groups of farmers in vari- 

 ous states who had been subjected to labor income surveys by 

 the usual farm management methods. With random selections 

 from such lists it was hoped to have several different judges 

 (county agent, banker, high school agriculturist and others) 

 rank the same group and from these data obtain the desired 

 correlations. But this was found to be an almost impossible 



