Just what these degrees of correlation mean is made somewhat more apparent if we treat the data in another way. If instead of computing coefficients of correlation we divide each class of men into four quartiles, and determine the average salaries of the men in these quartiles, we get very definite results. The upper quartile or group will now contain that fourth of the class whose grades were highest. The second, third and fourth quartiles will in turn represent decreasing degrees of academic proficiency. If the average salaries are the same for all quartiles, this will mean that there is no relation between salary and school grades. But if the salary varies with the grades, this will be a significant result. The actual data are as follows:

TABLE 19

Rice's Data Presented in a Revised Form

Average Salaries of the
Class and YearCases 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile
Mechanical '07 35 $1800 $1675 $1362 $1387
Mechanical '08 41 1450 1512 1512 1275
Mechanical '09 39 1375 1262 1313 1137
Electrical '07 26 1750 1675 1675 1412
Electrical '08 36 2147 1437 1262 1262
Electrical '09 41 1462 1212 1387 1200
Averages $1664 $1462 $1418 $1279
Percentages 100% 87% 85% 76%

If the separate classes be now considered the results are seen to be more or less irregular, although the general tendency is apparent. If the average results from all six classes are considered the results are more reliable as well as more uniform. The average salary varies in the same way as do the grades. If the average salary of the men of the first quartile ($1,664) be taken as a basis of comparison and considered one hundred per cent, then the salaries of the men in the second, third and fourth quartiles are respectively only eighty-seven, eighty-five and seventy-six per cent of this amount. In general terms, the salary of the men in the lower or poorest quarter of the class, from the point of view of school grades, will be only three-fourths the salary of the men in the upper or best quarter. The two middle quartiles will differ but little from each other, although the second has the advantage, by two per cent, or $44, over the third quarter.

If the class be divided into a better and a poorer half, then the average salary of the men in the upper half is seen to be $1,563, while that of the men in the lower half is only $1,348. The men in the upper half earn $215 more in a year than the men in the lower half. This way of expressing the results is both clearer and more concrete than the mere statement of the coefficient of correlation.

Interesting data on all three of these preceding questions are to be found in A. Lawrence Lowell's study of the academic careers of students in Harvard College, Law School and Medical School. This investigation included an examination into the college entrance examinations, the records attained during the college course, the subjects elected in this course, and the subsequent achievement of the men in the professional schools of law and medicine. The statistics cover the cases of all men who took the degree of A. B. at Harvard and then graduated from the two professional schools connected with Harvard. Only men who had taken at least three years of college work in residence were included. The records for the Law School cover the twenty years from 1891 to 1910. Those for the Medical School cover the sixteen years from 1895 forward.

The college gives degrees indicating four grades of distinction on the basis of scholarship. These are indicated as "plain," "cum laude," "magna cum laude" and "summa cum laude." The two professional schools grant degrees with two grades of distinction, viz., "plain" and "cum laude."

Lowell assumes that the grade attained on the college entrance examinations indicates with a certain degree of correctness the natural scholarly abilities of the student. The course of studies elected during college reflects roughly the general interests of the student at that time. The college records indicate his ability in the pursuit of those studies, including under ability such things as persistence, patience, fidelity, zeal, as well as native intelligence. The records in the professional schools are taken as indicating quite approximately the student's real ability to achieve success in the particular professional work of the technical sort.

All students are consequently classified according to these various factors. The entrance examinations are divided into "clear" and "conditioned." The college degrees and the professional degrees are classified on the basis of the degree of distinction awarded. All students are also classified on the basis of their election of the four possible college courses: (a) literature and languages; (b) natural sciences; (c) history and political science; (d) philosophy and mathematics. The relations between these various classifications are then presented, and analyzed in various ways.