It is to be noted that changes so small as from one quintile to the immediately adjacent one are not taken into account in this table. The figures show only those who were in precisely the same quintile all the way through. The indication is then that a student's performance in the first high school year is very significant of what his performance will be through the rest of the high school course, and also of significance with respect to what he will do in his university work. The significance of the early work, as has appeared in other studies also, becomes less and less the farther through the course one goes, so that in the senior year in college there is approximately a chance distribution with reference to the work of the first year high school. Smith also presents his results in the form of coefficients of correlation between various rankings. The following are the most interesting in the present connection: TABLE 15 CORRELATIONS (SMITH) H. S. Average and Univ. Freshman Average .48 H. S. Average and Univ. Sophomore Average .39 H. S. Average and Univ. Junior Average .47 H. S. Average and Univ. Senior Average .28 1st and 2nd Year High School .77 1st and 3rd Year High School .67 1st and 4th Year High School .66 University Freshman and Sophomore .73 University Freshman and Junior .61 University Freshman and Senior .45 These figures of course indicate the same facts as those derived from the previous methods of expressing the data. The high school (H. S.) average correlates throughout with the college ranking, the correspondence becoming less apparent in the later college years. Similarly, the good students in the first high school year are the good ones all through the high school course, and the able college freshmen are able as sophomores, juniors and seniors. But both in high school and in college the significance of early standing becomes less and less as the years progress. A. L. Jones[13] compared college entrance examinations with work done later in the college course, in the freshman and sophomore years. Two hundred men from the entering classes of 1907, 1911 and 1912, in Columbia College, were selected for study. These men were arranged in four groups, fifty in each group, on the basis of (a) their marks in entrance examinations, (b) their college marks in the first and second college years. Group I contains the best fifty individuals, Group II the fifty next best, etc. The following compiled table shows where the members of each group in entrance examinations stood in their college work: TABLE 16 SHOWING RELATIONS BETWEEN ENTRANCE RECORDS AND COLLEGE STANDING (JONES) See Text for Explanation