62 



BASES FOR CURRICULUM MAKING 



Two separately ob- 

 tained and comput- 

 ed sets of data show 

 a remarkable con- 

 sistency. 



Probably the most convincing evidence that the data may 

 be taken as fundamentally reliable may be obtained from a study 

 of Table VI. This is a combination table including the data of 

 Table III as given on page 21 and a second set of data simi- 

 larly but independently obtained from the 

 four eastern states of New York, New 

 Jersey, Connecticut and Maryland. 42 A 

 casual inspection of these figures at once 

 shows their likeness. In no case is there 

 a wide divergence the widest occurring 

 in the inter-relationship between Field 



and Chore Skills and General Education, a difference of only 

 .119. In the main, the procedure in obtaining these two sets 

 of data were sufficiently alike to make it possible to compute 

 their proable error, which roves to be only .035. 43 



42. These data were obtained and worked up before 

 the Middle Western States' data were gathered. The latter 

 were much more carefully guarded and procedure more fully 

 standardized so it is felt that their probably increased accuracy 

 justified their use in the study in preference to that first ob- 

 tained. Business ability was not included in the earlier study 

 and a few inter-correlations were incomplete. 



43. Having these two sets of da- 

 ta obtained and computed in similar 

 ways but entirely independently the 

 probable error may be found as fol- 

 lows: Beginning with the r(fi) rela- 

 tion which has the two values .689 

 and .732 the difference, substracting 

 algebraically, between the two is 

 found to be .043. Continuing this the 

 difference for all the relations (with- 

 out repetitions) may be arranged in a 

 scale as given at the left. From this it is 

 readily seen that the first and third 

 quartiles fall at +.045 and at .045. 

 Adding these and dividing by two 

 gives the .045 which may be taken as 

 the PE of the differences. Using the 



2 

 2 

 3 

 6 

 10 



Scale Fre- To- 



quencies tals 



+.100 to +.090 1 1 

 +.090 to +.080 1 

 +.080 to +.070 

 +.070 to +.060 1 

 +.060 to +.050 3 

 +.050 to +-040 4 

 +.040 to +.030 3 13 

 +.030 to +.020 2 15 

 + .020 to +.010 1 16 

 -f .010 to .000 1 17 

 .000 to .010 2 19 

 .010 to .020 1 20 

 .020 to .030 1 21 

 .030 to .040 2 23 

 _.040 to .050 4 27 

 .050 to .060 1 28 

 .060 to .070 2 30 

 .070 to .080 1 31 

 .080 to .090 31 

 .090 to .100 2 33 

 .100 to .110 33 

 .110 to .120 1 34 



formula: PE dif= ^PEa 2 +PEb 2 and 



substituting we have .045=\/2PE 2 be- 

 cause the PE of the eastern data (a) 

 may be assumed equal to the PE of 

 the western data (b). Solving this 

 equation PE equals .0345. 



