When one finds a difference between the school environments; of minority pupils and majority pupils, the question immediately arises: What is responsible for this difference? An important aspect of this question concerns residence: Does the difference arise because of differential allocation of resources within the same locality, or does it arise beca use minority children live in different localities from those of the whites, with different resources? If a difference is the result of the first of these two causes, elimination of the difference requires attention to the local distribution of reresources between minority and majority pupils, and to possible sources of discrimination; if it is the result of the second, this implies that there are geographic "pockets" of school deficiencies. The second means that certain localities-those in which a high concentration of minority students is found-have fewer resources devoted to schools. Elimination of the deficiency depends upon infusion of resources into the schools in localities, probably by means of an infusion of resources into the locality itself.

To distinguish between these two sources of inequality, two procedures have been used. One, used only for Negroes, is to show results separately for eight geographic and metropolitan-nonmetropolitan strata. Comparisons of Negro and white school environments within the strata show inequalities within these regional strata; comparisons for either race between different strata show inequalities that characterize the stratum as a whole. For example, table 2.24.2* shows that in the nonmetropolitan South, 15 percent of Negro elementary pupils and 11 percent of white elementary pupils are in a school with a remedial reading teacher, while in the nonmetropolitan North, 37 percent of Negroes and 46 percent of whites are in such a school. In one region, Negroes are slightly more often than whites in schools with remedial reading teachers; in the other, they are somewhat less often in such schools. However, it is in the comparison of regions that the major difference lies, with the nonmetropolitan South having a lower frequency of such teachers. This fact, taken together with the fact that there are many Negroes in the nonmetropolitan South, indicates a dis*Page 100.

by Negro and white alike in the nonmetropolitan South.

These regional comparisons, however, show the effect of locality of residence only very broadly, by large geographic regions. To examine the same question for much more restricted localities, an additional datum is presented in the tables and discussed in the text. It is ordinarily referred to as "whites in the same county," and in the tables is labeled "W(N)" for whites in the same county as Negroes (and for other minorities, W(), with initials of the minority inserted). Its construction is described in appendix 9. In brief, this measure weights the schools attended by whites in each county proportionally to the numbers of Negroes in that county.

It allows two comparisons. First, differences between Negroes and whites in the same county show differences within the same counties (or the standard metropolitan areas, for the metropolitan strata). Comparisons between whites in the same county and all whites in the region show differences between the counties or metropolitan area in which minority children are most heavily concentrated, and those in which whites are most heavily concentrated. An example may be seen in the same table referred to above, 2.24.2, where 37 percent of Negroes in the nonmetropolitan North are in a school with a remedial reading teacher, and 46 percent of whites are in such a school. To see the source of this 9-percent difference, we compare Negroes with the W(N) column, and find that 40 percent of the latter have remedial reading teachers. Thus, they are 3 percent different from Negroes in these counties. The other comparison, of W(N) with whites in the region, shows that whites in the same counties where Negroes are concentrated are 6 percent less often in schools with remedial reading teachers than whites in the region as a whole. Thus, in this case, the larger part of the overall difference between Negroes and whites is accounted for by the lower frequency of remedial reading teachers in the counties where Negroes are most heavily concentrated.

In nonmetropolitan strata, the locality for which this comparison is made is the county; for metropolitan strata, it is the standard metropolitan area, including the city and suburbs. The inclusion of suburbs with the central city in metropolitan areas means that the comparison between Negro and W(N) is in part a comparison of the schools in the cities, where most Negroes live, and the schools in

the suburbs, where many of the whites live. To have restricted the range to the central city alone would have obscured these city-suburb inequalities that exist within metropolitan areas.

2.13 Distribution of Negro children

Besides being classified by race, data for Negro and white children are classified by whether the schools are in metropolitan areas or not. The definition of a metropolitan area is the one commonly used by Government agencies: a city of over 50,000 inhabitants including its suburbs. All other schools in small cities, towns, or rural areas are referred to as nonmetropolitan schools.

For most tables, data for Negro and white children are classified by geographical regions. For metropolitan schools there are usually five regions defined as follows:

Northeast-Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont, Delaware, Maryland, New Jersey, New York, Pennsylvania, District of Columbia

Midwest-Illinois, Indiana, Michigan, Ohio,

Wisconsin, Iowa, Kansas, Minnesota, Mis-
souri, Nebraska, North Dakota, South
Dakota

South-Alabama, Arkansas, Florida, Georgia,
Kentucky, Louisiana, Mississippi, North
Carolina, South Carolina, Tennessee, Vir-
ginia, West Virginia

Southwest-Arizona, New Mexico, Okla-
homa, Texas

West-Alaska, California, Colorado, Hawaii,
Idaho, Montana, Nevada, Oregon, Utah,
Washington, Wyoming

The nonmetropolitan schools are usually classified into only three regions:

South-As above Southwest-As above

North and West-All States not in the South and Southwest

Data for minority groups other than Negroes are presented only on a nationwide basis because there were not sufficient cases to warrant a breakdown by regions. The following table using 1960 census data shows the distribution of Negroes age 5 to 19 among the eight strata defined above.

Since World War I there has occurred in this country a vast regional redistribution of the Negro

population through internal migration.* In 1910, 89 percent of the Negro population was concentrated in the South. By 1960, the proportion of Negroes living in the North and West had reached 40 percent, almost four times the proportion in 1910. The great change in the concentration of the Negro population is even more dramatic in their exodus to the metropolitan areas, and particularly to the large central cities of the Nation. In 1910, 73 percent of all Negroes lived on farms and in rural areas with less than 2,500 inhabitants. A complete reversal had occurred by 1960 with 73 percent of the Negroes living in urban areas, with 65 percent being in the largest metropolitan areas. Moreover, it is in the central cities of the metropolitan areas where the bulk of these Negroes reside. In 1960, 51 percent of all Negroes in the United States lived in the central cities; this urban residential pattern is much more characteristic of the North and West than it is of the South.

Table 2.13.1.-Distribution of Negro children
age 5-19, 1960

Furthermore, because the Negro population in this country is younger than the white population, the trend to a large concentration of Negroes in the cities is even more dramatic for the school-aged population. In 1960, 68 percent of Negro children of school age in the United States lived in the metropolitan areas. This 50-year pattern of

*For a more complete account of the changes in the distribution of Negroes throughout the Nation, see Philip M. Hauser, "Demographic Factors in the Integration of the Negro," Daedalus fall 1965, pp. 847-877.

increase the proportion of Negroes in the urban North and West.

2.14 Racial composition of schools attended by whites and minorities

The survey sample was not designed for the purpose of obtaining precise estimates of the racial compostion of schools attended by each of these groups of children. Despite this, for large regions of the country, it is possible to estimate the distribution of each of these groups of children in schools of varying racial composition. The regional groups for Negro and white children consist of clusters of the eight strata described in section. 1.2, as follows: the metropolitan North (Northeast, Midwest, and West), metropolitan South (South and Southwest), nonmetropolitan North (North and West), and nonmetropolitan South (South and Southwest). Graphs are presented (figs. 2.14.12.14.28)† to show estimates of the racial composition of schools at the beginning of school, for 1stgraders, and at the end of school, for 12th-graders, in each of these groups.

Each graph shows the percent of pupils in a given group in schools with a racial composition of 0-10 percent, 10-20 percent, 20-80 percent, and 90-100 percent pupils of that same group. For example, figure 2.14.1 shows that in the metropolitan North, 27.8 percent of Negro first-graders are in schools that are 90-100 percent Negro, 16.4 percent are in schools 80-90 percent Negro, an average of 8.0 percent are in schools in each 10percent interval from 20-80 percent Negro, 4.5 percent are in schools 10-20 percent Negro, and 3.1 percent are in schools 0-10 percent Negro. (The data for 20-80 percent are averaged over this range, because the low frequency gives unreliable estimates for any 10 percent interval in this range.)

The figures for first-grade Negro children show that their segregation is greatest in the metropolitan South, next greatest in the nonmetropolitan South, and least in the nonmetropolitan North. The first-grade white children are in schools that are even more completely limited to their own race. Almost no whites are in schools that are predominantly nonwhite. Among the other minorities, only Indian American children are located in any numbers in schools where their racial group is predominant.

*See, for example, Bureau of the Census, Supplementary Reports PS (SI)-49, Nov. 16, 1965.

Figures appear at end of this section except four which appear in sec. 1.1.

duced somewhat for Negroes. More of them are in schools of intermediate racial composition, and a higher percentage of them are in schools that are predominantly white. The change is least in the South (figs. 2.14.16 and 2.14.18), and greatest in the nonmetropolitan North. However, this latter region contains less than 5 percent of the Nation's Negroes, so that this integration represents only a small number of Negroes (see table 2.13.1).

Whites at grade 12 show almost identical concentration in predominantly white schools as they do at grade 1. Although more Negroes at grade 12 are in predominantly white schools, whites remain about as racially concentrated, though the category 90-100 percent obscures the fact that at grade 12, there is a higher percent of schools that have a few Negroes in them (though they remain in the 90-100 percent white category). The other minorities at grade 12 are, except for Oriental Americans, uniformly in schools where they are a smaller minority than are the grade-1 pupils.

race.

A slightly different way of looking at the data on racial composition is to find the percent of students of a race in schools in which they are the majority Almost all white pupils in both grades 1 and 12 are in schools where they are a majority. The percentage rises to within a fraction of 100 percent in grade 12 of the nonmetropolitan North and only goes as low as 94 percent in grade 1 of the nonmetropolitan South.

Table 2.14.1.-Percent of white and minority pupils in schools in which they are in the majority, fall

Negroes, although much less numerous than whites, are almost as likely to be a majority in the schools they attend in the South, especially in metropolitan areas. The greatest amount of racial diffusion, by this measure, for Southern Negroes is in grade 12 of the nonmetropolitan South, where 85 percent of the Negro students are in predominantly Negro schools. Negroes in the North are less racially isolated, especially at the 12th grade. At grade 1 in both the metropolitan and nonmetropolitan North, about 70 percent of the Negro

children are in schools in which they are a majority. At grade 12, 35 percent of the Negroes in the metropolitan North are in predominantly Negro schools, but only 8 percent of the Negroes in the nonmetropolitan North are in predominantly Negro schools.

A substantial number of Indian American and Mexican-American first-graders are in schools in which they are the majority group. This is not true at the 12th grade.