No abstract

Correlation coefficients ranging from -.47 to .92 have been reported. 449 450The Journal of Educational Psychology acuity of vision, etc. It depends, finally, upon the purpose for which the reading is done. For example, the reader may wish only to determine in general terms what the material is about. He may seek only to answer certain specific questions, or to locate certain specific facts. On the other hand, he may stop to ponder over the importance or the implication of the ideas presented. He may pause to memorize some of them. He may consider ways in which he can make possible use of these ideas. He may reorganize them to fit them better to his needs. He may think of new questions as he reads and re-read or continue reading with these questions in mind. What he does while he reads depends, then, on the purpose for which he undertook the reading in the first place or upon purposes which develop during the course of the reading. The nature of these purposes is conditioned by his educational background.Whether or not one considers all of these acts as a part of the reading process in the technical sense, or whether or not one considers all of the skills and the abilities exercised as reading skills, it must be acknowledged that these acts take place during reading and that a part of the time each individual spends in reading is given over to just such acts. It is apparent, then, that there is no meaningful 'single' reading rate (in words per minute) for any given individual, but that, instead, he reads at many different rates, each specific to a different purpose. It follows also, that for any given group of individuals, the degree of relationship between reading rate and any other variable may depend upon the purpose or purposes for which they read, that this relation will be most meaningful only if they all read for the same purpose, and that, accordingly, there may be many different degrees of relationship, each specific to a different purpose.It is, therefore, essential in a relationship study that the purpose of the reading be very carefully defined and controlled. This means, of course, that the purpose must be set in advance for all readers and must be clearly understood by them. Obviously, many of the purposes suggested above, especially those initiated during the course of the reading, can not be so defined and controlled. Hence, any rate-comprehension relationship study which is to yield meaningful results is limited not only by the fact that it must be based on reading done by a specific * The standard procedure for testing the significance of differences between pairs of related measures was employed (", p. 68f). ( -4.379 (<«t -2.576).

This article reports the data on sex differences in school achievement yielded by the Iowa Every-Pupil Testing Program, high school, for the years 1932 to 1939, and the Iowa Every-Pupil Basic Skills Testing Program (Grades III-VIII) for the year 1940. A brief review of representative articles dealing with previously published investigations of a similar character is also given.A REVIEW OF THE LITERATURE This brief review chiefly treats sex differences in achievement test scores, although attention is called at the outset to sex differences in school marks, promotion, acceleration, retardation, and similar evidences of school progress. In his Laggards in Our Schools, 1909, Ayres 2 concluded that "our schools as they now exist are better fitted to the needs and natures of the girl than of the boy pupils." He based this conclusion upon an analysis of the records of several hundred thousand pupils in various cities of the nation. In 7624 high schools in 1906-1907 there were 314,084 boys enrolled in comparison with 419,570 girls. In the elementary schools in fifteen cities, having an enrollment of 282,179 pupils, he found retardation among 37.1 per cent of the boys and 32.8 per cent of the girls. Approximately 23 per cent of the boys were repeating grades in comparison with 20.2 per cent of the girls. It is known that in recent years the number of boys in high school more nearly equals the number of girls.St. John's 16 data on retardation and acceleration make possible sex comparisons at comparable IQ levels. His investigation deals with the progress, over a four-year period, of about five hundred boys and four hundred fifty girls, Grades I to VI, chiefly I to IV, enrolled in the schools in a residential suburb of Boston. Table I shows the sex comparisons. St. John's data also show that correlations between IQ and achievement data were higher for girls than for boys. In marks of conduct and effort girls achieved a greater degree of superiority than in any of the other measures used.Johnson's 9 analysis of the records of the high-school pupils in St. Louis is to the same purpose. His data are shown, in part, in Table II.

No abstract