Next, we must realize that, since 1918, the vast improvements in the social and economic status of the Negro have not changed his relationship to the whites regarding capacity for education. This is not to say that this relationship cannot be changed; it says merely that it has not been changed....

Thirdly, as far as our knowledge of the problem goes, the improvements in the social and economic opportunities have only increased the differences between Negroes and whites. This is because such improvements have been given to both racial groups—not only to the Negro—and the whites have profited the more from them. This serves to emphasize the former statement that a fruitful approach to racial equality cannot follow the lines of social and economic manipulation. There is something more important, more basic, to the race problem than differences in external opportunity.

Dr. McGurk’s conclusions, it should be said in fairness (even in this partisan brief), have been widely denounced by his equalitarian colleagues. Following publication of his 1956 statement in U. S. News & World Report, eighteen social scientists united in a rebuttal assertion that “given similar degrees of cultural opportunity to realize their potentialities, the average achievement of the members of each ethnic group is about the same.” And in the Spring 1958, issue of Harvard Educational Review, William M. McCord, an assistant professor of sociology at Stanford University, and Nicholas J. Demerath, III, of Harvard, a senior student, returned to the attack on McGurk.

In my own view, the rejoinders of McCord and Demerath are remarkably feeble. The investigations they cite, in an effort to refute McGurk’s conclusions, provide no refutation at all. Their own study of “predelinquent” and normal boys in Cambridge-Somerville, Mass., is so affected by subjective evaluations that it contributes little to an objective appraisal of conditions that confront school administrators elsewhere. (They attempted to establish a correlation between the boys’ intelligence and their social class, parental education, “home atmosphere,” and “personality of the boys’ fathers”; other factors dealt with the subjects’ homes—cohesive, quarrelsome, quarrelsome-neglecting, or broken—and whether the boys’ fathers were loving, passive, cruel, neglecting, or absent.) In any event, most of their elaborately tabulated findings tend merely to support McGurk’s own conclusion that at the lowest social levels, white and Negro test scores are not significantly different.

The evidence put together by Shuey and McGurk is solid, dispassionate, unbiased, overwhelming; it cannot be disregarded—not, that is, if one wishes to gain any real understanding of the problems that confront local school boards over much of the South. To pull the general figures down to a single, specific case study, consider the findings of some tests administered in Dallas in 1954-55. There more than 1600 Negro pupils and almost 5700 white pupils were tested in the first grade on their general readiness for learning—on their ability to pay attention, follow directions, handle crayons and pencils, understand and use language, and so on. These were the findings:

Number ofPer centNumber ofPer cent
NegroNegroWhiteWhite
ChildrenChildrenCategoryChildrenChildren
15.92Superior57610.14
1056.47High Normal1,50326.50
29918.43Average1,81431.96
67741.71Low Normal1,39124.50
52732.47Poor Risk3926.90

In sum, 37 per cent of the white first-graders scored in the “high normal” and “superior” groups, against only 7 per cent of the Negro first-graders. At the other end of the scale, 31 per cent of the white pupils scored in the “low normal” and “poor risk” groups, against 74 per cent of the Negro pupils.

For another specific example, consider the findings in Virginia among pupils of an older age group. Over a period of five successive years, between 1949-50 and 1953-54, the State Department of Education administered the Iowa Silent Reading Test to all eighth-graders in the Virginia public school system. This is a standardized achievement test in reading, employed by school systems throughout the country to discover certain facts of immediate, practical importance to classroom teachers: How well do the children read? How well do they understand? The tests in Virginia were given in May of each year, when all of the children had a grade placement of 8.8 (eighth year, eighth month). Scores on the Iowa test are calibrated to match the grade placement, so that a pupil who scores a reading-grade equivalent of 8.7 would be one month retarded in achievement, and a pupil who scores a reading-grade equivalent of 8.9 would be one month advanced in achievement.

This is what the Virginia tests found in May 1954, the month of the Brown decision (findings were not significantly different in the four preceding years): The median white child in the county schools was about half a year behind the achievement level he should have reached; he was reading at a level of 8.3 (eighth grade, third month). But the median Negro child in the county schools was reading at a level of 6.2 (sixth grade, second month). The top one-fourth of the white children (75th percentile) were reading at a level of the tenth grade, third month, or better; but the top one-fourth of the Negro children were not even at the 8.8 level—the 75th percentile among the Negro pupils was found at 7.5.

Scores on the Virginia tests were higher in the city schools, but among the Negro pupils, not much higher. In the cities, the median white eighth-grader was found to be reading at a level of the ninth grade, second month; the median Negro eighth-grader scored 6.5. In less statistical language, this means simply that in terms of reading skills, which are the foundation of all other academic skills, Virginia’s white eighth-graders as a group were found in 1954 to be from two years to nearly three full years ahead of the Negro eighth-graders as a group. Subsequent tests, administered on a more limited scale since 1954, have shown no material change.