Another extensive study explored enrollment rates of children of various ages, races, and socioeconomic categories using 1960 census data. The study included also an investigation of school dropouts using the October 1965 Current Population Survey of the Bureau of the Census. This survey uses a carefully selected sample of 35,000 households. It was a large enough sample to justify reliable nationwide estimates for the Negro minority but not for other minorities. In this section the word "white" includes the Mexican American and Puerto Rican minorities.

According to the estimates of the Current Population Survey, approximately 6,960,000 persons of ages 16 and 17 were living in the United States in October 1965. Of this number 300,000 (5 percent) were enrolled in college, and therefore, were not considered by this Census Bureau study. Of the remaining, approximately 10 percent, or 681,000 youth of 16 and 17, had left school prior to completion of high school.

The bottom line of table 17 shows that about 17 percent of Negro adolescents (ages 16 and 17) have dropped out of school whereas the corresponding number for white adolescents is 9 percent. The following table 18 shows that most of this difference comes from differences outside the South; in the South the white and Negro nonenrollment rates are much the same.

Table 19 is directed to the question of whether the dropout rate is different for different socioeconomic levels. The data suggest that it is, for whereas the nonenrollment rate was 3 percent for

families, it was more than four times as large (13 percent) in the case of those from other than white-collar families (where the head of household was in a blue-collar or farm occupation, unemployed, or not in the labor force at all). Furthermore, this difference in nonenrollment by parental occupation existed for both male and female, Negro and white adolescents.

The racial differences in the dropout rate are thus sharply reduced when socioeconomic factors are taken into account. Then the difference of 8 percentage points between all Negro and white. adolescent dropouts becomes 1 percent for those in white-collar families, and 4 percent for those in other than white-collar families.

Table 20 breaks the data down by metropolitan and nonmetropolitan areas as well as by South and non-South. The largest differences between Negro and white dropout rates are seen in the urban North and West; in the nonurban North and West there were too few Negro households in the sample to provide a reliable estimate. In the South there is the unexpected result that in the urban areas, white girls drop out at a greater rate than Negro girls, and in the nonurban area white boys drop out at a substantially greater rate than Negro boys.

Relation of integration to achievement

An education in integrated schools can be expected to have major effects on attitudes toward members of other racial groups. At its best, it can develop attitudes appropriate to the integrated society these students will live in; at its worst, it

can create hostile camps of Negroes and whites in the same school. Thus, there is more to "school integration" than merely putting Negroes and whites in the same building, and there may be more important consequences of integration than its effect on achievement.

Yet the analysis of school factors described earlier suggests that in the long run, integration should be expected to have a positive effect on Negro achievement as well. An analysis was carried out to seek such effects on achievement which might appear in the short run. This analysis of the test performance of Negro children in integrated schools indicates positive effects of integration, though rather small ones. Results for grades 6, 9, and 12 are given in table 21 for Negro pupils classified by the proportion of their classmates the previous year who were white. Comparing the averages in each row, in every case but one the highest average score is recorded for the Negro pupils where more than half of their classmates were white. But in reading the rows from left to right, the increase is small and often

those Negro pupils in classes with only a few white pupils score lower than those in totally segregated classes.

Table 22 was constructed to observe whether there is any tendency for Negro pupils who have spent more years in integrated schools to exhibit higher average achievement. Those pupils who first entered integrated schools in the early grades record consistently higher scores than the other groups, although the differences are again small.

No account is taken in these tabulations of the fact that the various groups of pupils may have come from different backgrounds. When such account is taken by simple cross-tabulations on indicators of socioeconomic status, the performance remains highest in those schools which have been integrated for the longest time. Thus, although the differences are small, and although the degree of integration within the school is not known, there is evident, even in the short run, an effect of school integration on the reading and mathematics achievement of Negro pupils.

Tabulations of this kind are, of course, the

simplest possible devices for seeking such effects. It is possible that more elaborate analyses looking more carefully at the special characteristics of the Negro pupils, and at different degrees of integration within schools that have similar racial composition, may reveal a more definite effect. Such analyses are among those that will be presented in subsequent reports.

Case studies of school integration

As part of the survey, two sets of case studies of school integration were commissioned. These case studies examine the progress of integration in individual cities and towns, and illustrate problems that have arisen not only in these communities but in many others as well. The complete case studies are maintained on file at the Office of Education. In addition publication of all or some of the reports by their authors will be carried out through commercial publishers.

In the main report, excerpts from these case studies are presented to illustrate certain recurrent problems. A paragraph which introduces each of

these excerpts is given below, showing the kinds of problems covered.

Lack of racial information.-In certain communities, the lack of information as to the number of children of minority groups and of minority group teachers, their location and mobility, has made assessment of the equality of educational opportunity difficult. In one city, for example, after a free transfer plan was initiated, no records as to race of students were kept, thereby making any evaluation of the procedure subjective only. Superintendents, principals, and school boards sometimes respond by declaring racial records themselves to be a mark of discrimination.

A narrative of "the racial headcount problem" and the response to the search for a solution is given in the excerpt from the report on San Francisco.

Performance of minority group children.— One of the real handicaps to an effective assessment of equality of education for children of minority groups is the fact that few communities have given

of racial isolation while, at the same time, many white families have moved to the suburbs and other families ha ve taken their children out of the public school system, enrolling them instead in private and parochial schools. Small towns and medium-sized areas, North and South, on the other hand, are to some extent desegregating their schools.

In the Deep South, where there has been total school segregation for generations, there are signs of compliance within a number of school systems. The emphasis on open enrollment and freedom-ofchoice plans, however, has tended to lead to token enrollment of Negroes in previously white schools. In school systems integrated at some grade levels but not at others, the choice of high school grades rather than elementary grades has tended further to cut down on the num ber of Negroes choosing to transfer because of the reluctance to take extra risks close to graduation.

The move toward compliance is described in the excerpt from the report of one small Mississippi

town.

A voluntary transfer plan for racial balance in elementary schools. The public schools are more rigidly segregated at the elementary level than in the higher grades. In the large cities, elementary

schools have customarily made assignments in terms of neighborhood boundaries. Housing segregation has, therefore, tended to build a segregated elementary school system in most cities. in the North and, increasingly, in the South as well, where de facto segregation is replacing de jure segregation.

Various communities have been struggling to find ways to achieve greater racial balance while retaining the neighborhood school. Bussing, pairing, redistricting, consolidating, and many other strategies have been tried. Many have failed; others have achieved at least partial success. In New Haven, Conn., considerable vigor has been applied to the problem: Whereas pairing was tried at the junior high level introducing compulsory integration, a voluntary transfer plan was implemented at the elementary level. Relief of overcrowding was given as the central intent of the transfer plan, but greater racial balance was achieved since it was the Negro schools that were overcrowded. With the provision of new school buildings, however, this indirect stimulus to desegregation will not be present. In New Haven the transfer plan was more effective than in many other communities because of commitment of school leadership, active solicitation of transfers by