UuRlNG THE LAST THREE YEARS, published research on the applications of intelligence tests has been very much along the usual lines. With few excep tions, the results of these researches were not unusual or surprising. A large portion of the published studies were concerned with conditions and predic tions of scholastic achievement at all levels. Applications and usefulness of the 1937 Stanford-Binet, in particular, were the subjects of an appreciable number of researches; and Thurstone's "primary mental abilities" were given close scrutiny in a number of situations. Constancy of IQ (or equiva lents), community differences, socio-economic status, intelligence of nonwhite groups, atypical groups, sex differences, reading disability in rela tion to test performance, and several other miscellaneous problems received more or less attention. For the most part, students of these problems will find that the results of research of the three years covered herein largely confirm earlier findings.
Intelligence Tests and Educational AchievementElementary school-Cohler (20), studying a group of pupils in Grades VI-VIII, all having IQs of 120 or higher, explored differences between "achievers" (those working up to expectation) and "nonachievers." He found the expected inverse relationship between achievement and IQ. The difference in favor of the "achievers" was greatest in arithmetic and least in reading comprehension; and on the whole, "achievers" were about onehalf semester ahead of "nonachievers" in school achievement. For achieve ment age and MA, the r was +.58.Feinberg (24), studying IQ and EQ of children referred to a mental hygiene clinic, reported correlations between the two quotients varying from -.37 to +.76. The former was found for those having IQs between 44 and 69; the latter, between 120 and 192. The correlations were negative for the IQ-groups below 90, but positive for all groups above 90 IQ. The author concluded that EQ cannot be substituted for IQ, as some have sug gested, in dealing with children who present clinical or school problems. Thorndike, Woodyard, and Weingart (76) argued, however, that if school progress correlates .80 or better with intelligence tests, such progress may be treated as having the same value as intelligence tests, since coefficients of correlation for two different tests are rarely much over .80. The authors correlated IQ and age of children in Grade VI of seventeen cities. The me dian r in northern cities was .79; in southern cities it was .85.Benson (8) reported on the scholastic survival of 1680 children of dif