What Is in the Corn Judge's Mind?

In the spring of 1916, Prof. H.D. Hughes, of the Iowa Experiment Station, asked a number of experienced corn judges to score some five hundred ears of corn on the basis of what they thought the relative yields would be. These five hundred ears of corn were field run, varying from only three or four inches in length to more than ten inches. The variety was the college strain of Reid. In addition to the scoring, complete measurements were taken of each ear. Among other things, there were determined the length and circumference of ear, weight of kernel, filling of the kernel at the tip (tip of kernel, not tip of ear), blistering of kernel, and starchiness. These ears were planted, an ear to a row, and in the fall of 1916, yields were secured. The experiment was repeated in 1917. The method of correlation coefficients is admirably adapted to interpreting data of this sort to discover just what is in an experienced corn judge’s mind. It was found that the typical judge’s score was correlated with various factors as follows: length of ear .7, circumference .4, weight of kernel .5, filling of kernel at tip .4, absence of blistering of kernel .2, absence of starchiness .3. When these results were obtained, it was determined to make out the score card which really existed in the judges’ minds. The method used was the method of path coefficients as described in the January 3, 1921, issue of the Journal of Agricultural Research in the article “Correlation and Causation” by Sewall Wright. In using this method, it is necessary to have not only the correlation coefficients between the judge’s score and the various ear and kernel characteristics, but also the inter-correlations between the various characteristics. The correlation between length and circumference was found to be .3, between length and weight of kernel .3, between length and filling of kernel at tip .2, between length and absence of blistering, .2, between