College freshmen evidently are, as they should be, a highly selected group in regard to intelligence. The results obtained at different colleges differ somewhat, and the figures here given represent an approximate average of results obtained at several colleges of high standing. The median [{281}] score for freshmen has varied, at different colleges, from 140 to 160 points.
[Footnote: The "median" is a statistical measure very similar to the average; but, while the average score would be obtained by adding together the scores of all the individuals and dividing the sum by the number of individuals tested, the median is obtained by arranging all the individual scores in order, from the lowest to the highest, and then counting off from either end till the middle individual is reached; his score is the median. (If the number of individuals tested is an even number, there are two middle individuals, and the point midway between them is taken as the median.) Just as many individuals are below the median as above it. The median is often preferred to the average in psychological work, not only because it is more easily computed, but because it is less affected by the eccentric or unusual performances of a few individuals, and therefore more fairly represents the whole population.]
It will be noticed in the graph that none of the freshmen score as low as the median of the drafted men. All of the freshmen, in fact, lie well above the median for the general population. A freshman who scores below 100 points finds it very difficult to keep up in his college work. Sometimes, it must be said, a freshman who scores not much over 100 in the test does very well in his studies, and sometimes one who scores very high in the test has to be dropped for poor scholarship, but this last is probably due to distracting interests.
No such sampling of the adult female population has ever been made as was afforded by the draft, and we are not in a position to compare the average adult man and woman in regard to intelligence. Boys and girls under twelve average almost the same, year by year, according to the Binet tests. In various other tests, calling for quick, accurate work, girls have on the average slightly surpassed boys of the same age, but this may result from the fact that girls mature earlier than boys; they reach adult height earlier, and perhaps also adult intelligence. College women, in the Alpha test, score on the average a few points below college men, but this, on the other hand, may be due to the fact that the Alpha test, being prepared for men, includes a few questions that lie rather outside the usual range of women's interests. On the whole, tests have given very little evidence of any significant difference between the general run of intelligence in the two sexes.
Limitations of the Intelligence Tests
Tests of the Binet or Alpha variety evidently do not cover the whole range of intelligent behavior. They do not test [{282}] the ability to manage carpenter's or plumber's tools or other concrete things, they do not test the ability to manage people, and they do not reach high enough to test the ability to solve really big problems.
Regarding the ability to manage concrete things, we have already mentioned the performance tests, which provide a necessary supplement to the tests that deal in ideas expressed in words. It is an interesting fact that some men whose mental age is below ten, according to the Binet tests, nevertheless have steady jobs, earn good wages, and get on all right in a simple environment. There are many others, with a mental age of ten or eleven, who cannot master the school work of the upper grades, and yet become skilled workmen or even real artists. Now, it takes mentality to perform skilled or artistic work; only, the mentality is different from that demanded by what we call "intellectual work".
Managing people requires tact and leadership, which are obviously mental traits, though not easily tested. It is seldom that a real leader of men scores anything but high in the intelligence tests, but it more often happens that an individual who scores very high in the tests has little power of leadership. In part this is a matter of physique, or of temperament, rather than of intelligence, but in part it is a matter of understanding people and seeing how they can be influenced and led.
Though the intelligence tests deal with "ideas", they do not, as so far devised, reach up to the great ideas nor make much demand on the superior powers of the great thinker. If we could assemble a group of the world's great authors, scientists and inventors, and put them through the Alpha test, it is probable that they would all score high, but not higher than the upper ten per cent, of college freshmen. Had their IQ's been determined when they were children, [{283}] probably all would have measured over 180 and some as high as 200, but the tests would not have distinguished these great geniuses from the gifted child who is simply one of a hundred or one of a thousand.