their school by holding back pupils. They will never abandon it under the present examination system; for under such a condition it is true; quantity of drill is a means of securing high standings in arbitrary examinations. Occasionally, a stupid boy who is misjudged by his teachers will be admitted to college, but the present system is a paradise for stupid boys -with clever tutors. A sagacious tutor can get a hundred boys into college, not one of whom he would be willing to certify to as fit to succeed there. Such a system is rational because it measures the ability of schools to fit for college, not their ability to forearm students against the twin cataclysms of preliminary and final examinations. It puts the premium on capacity and right habits of intellectual work, rather than on the mass of information held in solution at a given week. It avoids the dangers, possible under the ordinary certificating systems, of misjudgment of schools by inadequate or eccentric inspection. It measures directly and exactly the fact we wish to measure. Finally, such a system would be established thru a natural modification of the function of an already existing organ, thru an easy extension of the powers of the present board. No new machinery and only the simplest legislation is required. The only important change would be to add to the present duties and powers of the board the duty of rating schools by the success in college of the students they had vouched for and the power to accept from schools of a given rating a certificate that John Doe "is fit for college," and a certificate that John Doe "has done work equivalent to that recommended by the Middle States board for English I, English 2, History 1," etc., etc. The colleges which approved the system would vote simply to accept the board's examinations of schools as they now accept its examinations of individual students. The work of the board and of college admission committees would be lightened. Of the many administrative advantages of the plan, and of the possibility of unity of action amongst colleges thruout the country on the basis of a scheme so safe and yet so plastic, I do not care to speak, at least at this time. The system proposed is rational, just, and practical. It positively encourages the right students to go to college instead of making laborious, but futile, efforts to keep a few incompetents out. On these facts alone I rest my case. TEACHERS COLLEGE, COLUMBIA UNIVERSITY EDWARD L. THORNDIKE APPENDIX Tables I, II, III, and IV show for each individual the relation between entrance standing and college standing. Horizontal position denotes the rank in entrance (the median of the highest eleven marks obtained). Vertical position denotes the rank in college studies (the average of the five highest marks obtained)—in Senior year in Table I, in Junior year in Table II, etc. Each figure entered in the table means so many students. Thus in Table I the 1 at the upper left-hand corner means that one student scoring 60 in entrance scored 4 in the college work of Senior year. The other I in the same column means that one student scoring 60 in entrance scored 21 in college work. The 1 in the next vertical column means that one student scoring 61 in entrance scored 24 in college work. The vertical column under 70 would read: Of 10 students each ranking 70 in entrance examinations, one ranked 15 in the college work of Senior year, one 16, four 18, one 19, one 21, one 22, and one 27. The values 60, 61, 62, etc., up to 95 of the horizontal scale, are directly obtained from the entrance marks, which are given on the ordinary scale of from 100 down. The values 4, 5, 6 up to 30 of the vertical scale, are obtained from the college records of A B C D and F by taking A=6, B=4, C=3, D=1 and F=o (note). Thus 30 five As, 28=four As and one B, 27 four As and one C, 26-three As and two Us, 25 three As, one B and one C, or four As and one D, etc., etc. Tables VB and VIB show the facts of the relationship between entrance standing and college standing of Senior and Junior years still more clearly. In these tables a black rectangle about 2.5 by .5 mm. equals one individual. The measures used were in this case the median of all entrance marks and the average of all the college marks of the given year. Tables VA and VIA show how the black rectangles would be arranged if each student's relative position in college work were the same as in entrance examinations. NOTE-A=10, B=7, C=5, D=2, and F=0 would perhaps have been juster. |