further that although in every State except Kentucky the percentage of Negro rejections for educational deficiency is higher than that of whites, in a number of the Northern States the percentage of Negroes rejected is appreciably less than that of whites in a number of the Southern States. For example, Alabama has a rejection rate of 8.5 percent for whites and 25.8 percent for Negroes, while the rejection rate for Negroes in Illinois is only 2.5 percent.

It is estimated that, as of September 1, 1943, 341,200 registrants had been placed in class IV-F because of educational deficiency. Negroes, with a rejection rate for this defect, about 6 to 7 times that of whites, accounted for slightly more than half of the total number of registrants rejected for educational deficiency.

TABLE III.-Rejection rates of white and Negro registrants in selected States due to failure to meet minimum "intelligence" standards, June-July 1943 12

1 Based on a sample of 1633 percent of Forms 221 received during June and July. Preliminary data. 2 States having at least 0.3 percent of total Negro registrants in first, second, third, and fifth and sixth registrations.

3 Less than 200 registrants represented in June-July sample of Forms 221.

4 Less than 50 Negro registrants represented in June-July sample of Forms 221. NOTE.-Above figures include local-board and induction-station rejections.

The parallel situation in World War I

The situation just described is strikingly similar to that which existed during World War I a quarter of a century ago. Then, as now, the Army testing procedure revealed that a large proportion of American men of draft age were at an extremely low mental and educational level. Then, as now, it was found that a disproportionate number of these submarginal men were in the Southern States. Then, as now, this retarded group was heavily weighted with Negro selectees. Then, as now, the Negroes in some Northern States were superior to the whites in some Southern States. In view of this parallel situation, we may profitably examine here the interpretations which were made of this phenomenon immediately following World War I.

In World War I mental testing gained a remarkable vogue and prestige attributable mainly to the group testing involving about 2,000,000 soldiers. Individual testing (Binet) had been known in this country since about 1909, but group mental testing was very largely the creature of World War I. The results of Army testing became a center of controversy. The naturenurture argument flared anew and raged for 2 decades. Only recently has it shown signs of abating. Mental testing began in France as a means of classifying mental subnormals. But within a year after its introduction in the United States, it was catapulted into American racialism. In Philadelphia testing was used to justify racial segregation in public schools, as early as 1910. By 1916 Terman had pronounced Negroes and Indians racially inferior stocks. A work in the early twenties by Brigham made the boldest of claims for superiority of the Nordics.1

1 Brigham, Carl C., A Study of American Intelligence, Princton: Princton Press, 1923.

Brigham believed that the mental tests tested inborn power except for experimental error: “* for we must assume that we are measuring native or inborn intelligence, and any increase in our test score due to any other factor may be regarded as an error." It should be said, however, to the credit

* *

of Brigham that he later pronounced this work useless as a study in racial differentials. The collapse, he thought, came from technical defects revealed by the factor analysis movement. Its collapse could well be justified on other grounds.

The claims of test extremists met sturdy objection from a number of workers. Among the earlier of these the best known was W. C. Bagley. He attacked a fundamental assumption of the mental tester; namely, that mental test scores are almost a single-valued function of physical inheritance. He insisted that mental scores are in reality a function of an undetermined number of functions of undetermined relative potencies and presented a mass of evidence to substantiate his contention. He studied by States the relations between test scores and magazine circulation, birth States of persons listed in Who's Who in America, birth States of persons committed to the State prisons of New York and California, frequency of homicides, venereally infected recruits, per capita of income tax, ratio of per capita savings-bank deposits to per capita income, and components of Ayers Index Numbers for State School Systems. Bagley found an average correlation of 0.85 (range 0.73 to 0.91) between A and B Alpha scores and Magazines. Leaders, Income and Schools. Between the same Alpha grades and venereal diseases, homicides, inmates in Federal prisons, he found correlations respectively of 0.74, 0.41, and 0.37.5 These factors also showed relatively high correlation with educational indices. The correlation between estimated percentage of Nordic blood and Alpha scores was negative, approximately 0.50.o A notable result of Bagley's work is the fact that the Southern States generally fell at or toward the bottom of the lists of States when ranked with reference to functions of the factors considered by him just as they do in the school indices. An impressive exception is the estimated percentage of Nordic blood. Bagley's work sharpened the controversy whose research sequels are well represented in the Year Book of the Society for the Study of Education for 1928 and 1939 and in the Journal of Negro Education, 1934. Though Bagley went beyond his data in some of his conclusions, his main thesis would seem markedly validated by later research.

8

Army testing was a release event for pro-Nordic propaganda in the United States which undoubtedly influenced the rise of Aryanism in Europe now best symbolized in nazism in Germany and in more recent fascism in Italy. Hitler is reputed to have said that the American treatment of Negroes served as a pattern for his persecution of Jews. Perhaps nothing so incensed the Anglo-Saxon peoples as Hitler's claim of native superiority of Germans and their Nordic allies including Italians and Japanese. We are now engaged in an unprecedented war presumably to liquidate a view nourished on our own soil. Bigotry got out of hand and turned upon its mentors.

We have shown here that the current situation parallels that which existed during World War I, a quarter of a century ago. Then, as now, large elements in our population were found to be unprepared for miliary service due to mental, educational, and physical disabilities. That little has been done on a national scale to alleviate this condition is nothing short of a national disgrace. Surely all thoughful persons in our society must agree that this problem shall not continue to be neglected.

Rehabilitation programs

The existence of a large number of men rendered unfit for military service by their educational deficiency obviously constitutes a challenging educational program. Especially is this true when we consider the fact that selectees constitute only a part of the total population and that they are the product of a situation of long standing. Can an educational program be devised which will effectively salvage persons who are below a minimum acceptable educational level? What will be the nature of such a program?

2 Brigham, Carl C., Intelligence Tests of Immigrant Groups, Psychological Review, 37: 158-65, 1930.

a W. C. Bagley, Determination in Education, Baltimore: Warwick & York, 1925.

4 Ibid., p. 81.

Ibid., p. 106.

Ibid., pp. 126-127.

The Twenty-seventh Yearbook of the Society for the Study of Education.

The Thirty-eighth Yearbook of the Society for the Study of Education.

Physical and Mental Abilities of the American Negro, Journal of Negro Education, July

1934.

FEDERAL AID FOR EDUCATION

The Army has recognized and attacked this problem. It has evolved a program through its special training units, which has serviced thousands of illiterate selectees and prepared them for effective military service. City and State school systems have likewise recognized and attacked the problem. Adult education programs, which have long been in existence, have in many instances, been geared to existing needs.

The total effectiveness of these programs, both military and civilian, and their possibilities in contributing to a solution of the problem of adult illiteracy, and thus increasing the social efficiency of submarginal population groups, are matters which require close examination. This matter is of especial importance

in view of our failure, after World War I, to effect in any fundamental way a solution of the problem.

The social implications of the situation

The chief conclusions to be drawn from the foregoing discussion are, that an inexcusably large proportion of the Negro population, and of the population of the Southern States, reveals an extremely low level of functional ability; that the situation is similar to that which existed during World War I; and that a fundamental solution of the problem of adult illiteracy is an urgent matter. This condition of adult illiteracy is a matter for national concern for a number of reasons.

First, it results, obviously, in a reduced reservoir of manpower for the armed forces; hundreds of thousands of physically fit men are disabled for military service because of educational deficiency.

Second, it results in an increased draft on other population groups for military service. The demands of the Army and Navy for men must be met. we have witnessed the drafting of fathers, of older men, of skilled workers in As a result essential industries. Thus, the deferment of a man in Georgia for illiteracy results in the drafting of a skilled worker or father in New York.

Third, it results in reduced efficiency of the military service. It follows that in any situation where a large proportion of the population fails to meet the bare minimum level of ability, a much larger proportion exceeds this level but still falls below any reasonably desirable level of ability. Thus the armed forces must recruit and train hundreds of thousands of men who are at a minimum level of acceptability.

Fourth, it results in an increased burden being placed on those States and communities which have provided good schools and other cultural facilities for their children. A large proportion of the American people have moved from their native State to live in another State, and the pattern of this migration is such that people tend to move from States of low educational and cultural level to States of higher educational and cultural level. persons are handicapped in their social participation, a handicap which is A large proportion of such reflected in the life of their adopted community.

Fifth, and in the long view most important of all, it results in the reduced social efficiency of large elements in the population and consequently of the Nation itself. People who are not sufficiently competent to participate in the war effort, are likewise unable to make their best contribution to a peacetime economy.

PROCEDURE AND SOURCES OF DATA

This study is concerned with two distinct but related problems. tion with the first problem-What basic factors contribute to the high rejection In connecrate of Negro selectees on account of educational or mental deficiency?-the committee utilized official statistics of rejections and other information furnished by the Selective Service System, and educational data largely from reports of the United States Office of Education. These data have been used to examine hypotheses set up to explain the differential rejection rates of white and Negro selectees. The relationships between rejections in the several States and various educational items such as per capita expenditures, average daily attendance, and percentage of population enrolled in high school have been examined, chiefly by the correlation technique.

For the rejection data most frequently referred to in this study-rejections for failure to meet minimum "intelligence" standards, June-July 1943-theSelective Service System provided information for the 29 States having at least 0.3 percent of total Negro registrants. sample of registrants examined during the 2-month period noted, but we are These data constitute a 16% percent assured by Selective Service officials that the data are typical of the entire period during which this plan of selection was in operation. Even minor errors

FEDERAL AID FOR EDUCATION

in the sample, however, would not appreciably affect the major findings of this study.

The The educational data used are chiefly for 2 years: 1939-40 and 1931-32. 1939-40 figures, the most recent year for which these data are now available, are representative of the condition of the schools when the younger men now The 1931-32 figures are repreeligible for military service were in attendance. sentative of the condition of the schools when the typical registrant was in attendance. In a few instances data for other years were used because they were readily available. It may be added, that, in view of the fact that the relative rank of the States changes but little over a period of years the relationships found would not be substantially different if other years were used.

In attacking the second problem of this study-How can the rejection rate of Negroes be reduced and what sort of program needs to be established to rehabilitate population groups now failing to meet reasonable standards of performance?-the committee made a critical analysis of the Army program for the rehabilitation of mentally retarded soldiers and of State and city programs for salvaging persons with educational and mental deficiencies. The data for these analyses were secured from published articles, interviews with key persons and, in the case of city and State programs, correspondence with persons associated with the programs,

The conclusions and recommendations which appear at the end of this report are based, of course, on the committee's interpretation of the findings of the study. We have attempted to be as objective as possible in the analysis of the data and it has been our aim that this report eventuate in a series of recommendations which will provide the basis for a practical program of action by the American Teachers Association and other organizations and individuals.

PART II. AN ANALYSIS OF THE BASIC FACTORS CONTRIBUTING TO THE REJECTION RATE OF NEGRO SELECTEES ON ACCOUNT OF EDUCATIONAL AND MENTAL DEFICIENCIES The statistics presented in part I, and especially the data of table III, portray an extremely serious situation with respect to the rejection of registrants for failure to meet the minimum "intelligence" standards established by the military services (the data of table III, p. 6, are shown graphically in fig. 3, p. 25). A number of questions immediately arise in connection with the pattern of rejections.

Why is there a great disparity in the rejection rate of white selectees in the several States? In North Carolina 10.7 percent of the whites were rejected while in Illinois the rate is only 0.3 percent.

Why is there a similar disparity in the rejections of Negroes? In South Carolina 43 percent of the Negro registrants were rejected for failure to meet minimum intelligence standards; on the other hand only 2.5 percent failed to pass in Illinois.

Why is there a difference in the rejection rates of whites and Negroes within the same State? Only in Kentucky, where 6 percent of the whites and 5.4 percent of the Negroes were rejected, is the rejection rate for whites greater than that for Negroes. In South Carolina 43 percent of the Negroes and 8.7 percent of the whites were rejected; in Illinois, 2.5 percent of the Negroes and 0.2 percent of the whites.

Why is the rejection rate less for Negroes in some States than for whites in other States? The percentage of Negroes rejected in New York City (2.4 percent) and in States such as Illinois (2.5 percent), Massachusetts (2.8 percent), Michigan (4.2 percent), is less than the percentage of whites rejected in States such as North Carolina (10.7 percent), Texas (10.4 percent), and Arkansas (9.8 percent). In fact the rejection rate for Negroes in Illinois (2.5 percent) is less than that for whites in 17 of the 29 States for which data are available.

It is our purpose, in this section to examine the factors contributing to this condition and to ascertain, if possible, the answer to the questions raised above. Although we are concerned about the total rejection picture, the focus of our inquiry is the Negro selectee.

Any one of, or a combination of, the following factors might account for the phenomenon and these will be examined to the extent that available data will permit

1. That the tests used are not valid for use with persons from subcultural environments.

2. That there are defects in the administration of the tests.

73384-45-pt. 1-22

3. That Negro selectees malinger in the test situation in an attempt to escape military service.

4. That there is a close relationship between educational opportunities available to Negroes in the several States and the rate of rejections for educational and mental reasons. And, as a corollary of this, that Negro rejectees in Northern States are heavily weighted with migrants from the Southern States.

ARE MENTAL TESTS VALID FOR USE WITH PERSONS FROM A SUBCULTURAL ENVIRONMENT?

10

What do mental tests measure? It seems now well established that a very large component of mental test results is general culture. No mental test thus far devised, or that promises to be devised measures raw mental capacity. This seems so obvious that one wonders how any scholar could have believed otherwise. A significant error of earlier tests was the hypothesis that non verbal tests could be counted upon to measure more closely native capacity of the culturally deprived than can be done through verbal tests. In the Army report pains are taken to credit Negroes with relatively high verbal ability for this was thought to emphasize their inferiority. We know that this was a mistaken view. Culturally deprived subjects in the American milieu actually make a poorer showing on nonverbal than on verbal tests. This has been the experience with the California mental maturity test which provides for both nonverbal and verbal I. Q.'s. It comes out especially clearly in a recent work by Machover," and we are informed that it has been validated in current Army testing experience. We may lay it down that apart from consideration of culture, in the larger sense, native intelligence is not isolable. Below a certain critical level of culture, normative measurement of intelligence is simply impossible.

12

This principle has especial bearing upon the oft-repeated observation that in childhood Negroes and whites are alike in intelligence, but differences widen with increase of age. It is well established that the I. Q.'s of children in deprived environment decline rather rapidly with age. This tendency is clearly revealed among rural children. One of the writers has studied this trend in Wheeler's east Tennessee mountain children and in Negro children in Washington, D. C., and finds that the decline of I. Q.'s among the white mountain children is more rapid than among Washington Negro children." It appears that the crux of this matter is that I. Q.'s on the average require a certain level of cultural media in order to be sustained." It follows then that in a relatively poor environment, given reasonable schooling, children at, say, the third-grade level may measure normal because the functions measured by the intelligence tests at this level are fairly well represented in the third-grade cultural environment. On higher grade levels, the functions measured are not provided in such an environment. The cultural environment simply does not sustain I. Q.'s based on higher cultural levels of society. If this hypothesis is established, it will greatly clarify certain social misconceptions more or less current with reference to race and social class.

ARE THERE DEFECTS IN THE ADMINISTRATION OF THE TESTS?

The members of the committee had no opportunity to observe the administration of the tests to selectees. We are assured, however, by persons intimately connected with the program that the tests are administered under exactly the same conditions to white and Negro selectees. It appears that, in some instances, the test situation is not satisfactory. Men may enter the test situation without adequate orientation, or without sufficient rest after a long journey to the induction center. But it is probable that these conditions are not greatly different for the white and Negro selectees. We should draw the conclusion, admittedly on meager evidence, that the differential rejection rate of whites and Negroes cannot be accounted for by defects in the administration of the tests.

10 Robert Yerkes et al. Psychological Examining in the United States Army, Memoirs of the National Academy of Sciences, vol. XV, 1921.

11 Solomon Machover, Cultural and Racial Variations in Patterns of Intellect, 1943. 12 L. R. Wheeler, "The Intelligence of East Tennessee Mountain Children," Journal of Educational Psychology, 33: 321-34, 1942.

13 Unpublished data.

14 Howard H. Long, "The Intelligence of Colored Elementary Pupils in Washington, D. C.," Journal of Negro Education, 3: 305-22, April 1934.