Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Educational Attainment and Economic Status

by Edmund Des. Brunner - 1948

As a result of the 1940 Census it is now possible to throw some light on the relationship between economic status and the educational attainment of children seven to seventeen years of age. This article summarizes one aspect of these data, which are based on a 5 per cent sample of all non-farm children of the stated ages enumerated in the 1940 Census.

IT has long been assumed that the quality of housing in any given locality is related to a number of social factors such as morbidity and death rates, juvenile delinquency, crime, extent of activity by character-building agencies, and the like. A considerable number of studies have documented this assumption for particular cities. Some of these have inferred a causal connection, though this has been difficult to demonstrate.

As a result of the 1940 Census it is now possible to throw some light on the relationship between economic status, as measured by the actual rent of tenant-occupied homes and the estimated rent value of owned homes, on the one hand, and the educational attainment of children seven to seventeen years of age, as measured by number of grades of schooling completed, on the other. The data cover the urban and rural non-farm population only. For obvious reasons the farm group could not be included. However, native white and Negro children can be separately studied. This article summarizes one aspect of these data,1 which are based on a 5 per cent sample of all non-farm children of the stated ages enumerated in the 1940 Census.

The margin of error is probably quite small. This conclusion is supported both by the statistical formula for determining probable error and by the comparison of the actual total number of children for each year of age enumerated in the Census with the total for each age calculated from the findings on the 5 per cent sample. For whites, the difference between the two totals exceeded .9 per cent in only one age group, that of the native-born females sixteen to seventeen years old, for which the difference was 1.2 per cent. For the Negroes, the largest difference was 2.8 per cent. Obviously the possible errors increase in size as the number of cases in any given category declines. This would affect particularly the highest rental categories in the rural non-farm population.

Before the data are given, certain cautions should be indicated. The categories in which the Census presents the material are crude in that very large populations are used. The rural non-farm category, for instance, includes all incorporated places with less than 2,500 population. While most of these are the village centers of farming communities, this category includes also some suburban places and some one-company industrial villages where the industry owns all houses and where rent concessions may be part of the indirect wages. Furthermore, rents vary considerably among the regions of the United States and within a region among the several cities. Seventy-five dollars a month will buy better housing in Mobile, Alabama, than in Wilmington, Delaware, but both cities fall within the Southern region. Moreover, as the 1940 income data show, some high-income families live in low-rent homes. The subtle factor of family values enters in here. None the less, the data show certain definite trends that appear to have significance for educators.

In national terms the data indicate the expected relationship between economic status, measured by rental value of the home, and educational attainment. In other words, for each year of age, the higher the rent the higher the median years of schooling completed. As would be expected, the actual differences are least in the lower ages and grades. The child of seven in a city of over 250,000 population, living in a home costing under $10 a month, had finished 1.3 years. In all rental groups from $30 to $49 or over for males and from $20 to $29 up for females, this figure was 1.7 years. In cities of 25,000 to 250,000 population the spread for the seven-year-olds was from 1.3 to 1.6 years; in cities of under 25,000 it was from 1.2 to 1.6. Only the rural non-farm seven-year-olds showed a narrow spread, from 1.4 to 1.6 years.

Apparently all groups from the lower middle class, to use loose categories, on up the economic scale enter their children in school as early as permitted, and these children make normal progress. Below that level, either the children enter later, or difficulties in progress develop in a sufficient number of cases to affect the median, or both causes operate.

Though the difference between low and high rental groups for age seven is statistically small, a spread of nearly half a year suggests that difficulties showing up later may have begun quite early and that therefore psychological services should by no means omit attention to the early grades.

By age seventeen the spread between the lowest and the highest rental groups had widened to 2.1 years in the largest cities and to 2.4 in the smallest cities. For each rental group except the lowest, under $10, there was an average increase of half a year of schooling at age seventeen over the previous rental category, until the $30 to $49 and upward categories were reached. Thereafter differences, while positive, were only .1 or .2 of a year. Here, obviously, is an indication of the influence on such medians of those who drop out of school as soon as the law permits their employment. Thus, among the high rental groups, the seventeen-year-old has completed 1.8 years more of schooling than the fifteen-year-old. In the lowest rental group, the difference is only 1.1 years.

One item of some interest develops from these national figures. It has been well known, from field studies, school records, and prior Census data, that girls are more likely to finish school or to go further in school before dropping out than boys. This fact is true for the three size groups of cities and for the rural non-farm population, with a single exception: the lowest rental group in cities of 250,000 or more. Here the boys of seventeen have completed 9.4 years, the girls 9.2. Moreover, the difference between the boys of sixteen and seventeen in this rental group is .8 of a year, and among the girls only .2 of a year. What psychological motivations and economic pressures produced this result is a matter for interesting speculation. Almost a quarter of a million seventeen-year-olds, 14.9 per cent of this age, including slightly more girls than boys, are involved in this lowest rental group.

So far as the detailed data are concerned, an illustrative sample only can be given. The reason lies in the plenitude of material the Census offers. Three major regions are used, North, South, and West. Three urban size categories and the rural non-farm population are separately tabulated. Seven rental classifications are employed. The data on the number of years of schooling completed are then given for each year of age separately by sex, in each of the categories mentioned, for both whites and Negroes. Thus the range of figures is wide— for age fifteen and over, the years of schooling range from none to college.

This article centers attention on the native white twelve-year-olds. The child of that age who progresses normally through a school system has finished the sixth grade. In systems that use the 6, 3, 3 plan he has completed elementary school. Moreover, this age is one which in every state is less than the legal school-leaving age, a fact which makes some of the data disturbing. It should be pointed out that the complete data as published by the Census permit any school to compare its rate of retardation, normality, and acceleration with the national or regional record for other communities of its size group for the years covered by compulsory school attendance laws. This article is presented in the hope that it will lead to further analysis both regionally and locally.

The data for the twelve-year-olds are presented in the Tables 1,2, and 3 herewith. The percentages have been calculated from the total numbers given in Table 1 of the Census monograph noted in footnote 1.

It is apparent that the proportion of children who are retarded decreases as the economic status of their families, measured by rental paid, increases. Conversely, the higher the rental the higher the proportion of children who are accelerated. This is true in all three regions and, with slight exceptions, for both urban and rural non-farm children.

In comparison with that of the North and West (except for the Western rural non-farm group) the proportion of retarded children, that is, those who at age twelve have completed four years or less of school, is high in the South. Conversely, except for the highest rental rural non-farm group, which is numerically very small, the proportion of children who have completed seven or more years of school at this age is significantly higher in the North and West than in the South.


In all regions, in the two lowest rental groups the proportion of children retarded sharply exceeds the proportion accelerated. In the three highest rental categories, the reverse is true.

It is important to realize, however, that in every region there are those in the lowest economic status categories who have completed eight or more grades. Again, conversely, the highest rental categories also contribute to the group which has completed only three grades or less. Economic status, then, is not a final determinative factor, but it may serve either as an impetus or as a severe handicap to educational progress. The data raise the unanswerable question as to the amount of social loss resulting from failure to compensate for this handicap.

In all regions and almost all rental categories there is less spread, that is, a greater concentration around normal accomplishment, for the rural non-farm children than for the urban. The rural group tends to show less extreme retardation or acceleration than the urban. This fact may be related to the type of school in rural villages or to a tendency in some parts of the nation to start a child to school at six years or later rather than at the earliest possible moment, five and one-half years. It could also be related to the lack of kindergartens in rural non-farm areas.

Table 2 as presented conceals some striking differences between boys and girls. The latter make by far the better record. They furnish a smaller proportion of the retarded and a larger proportion of the accelerated. In the lower rental categories, the differences are as high as 30 to 50 per cent. In the two highest rental groups they are much smaller.

With only minor exceptions, the generalizations presented hold true also for Negroes. They have, naturally, a larger proportion of children in the lower income categories than the whites and also a higher proportion who have failed to complete six years of school at age twelve. Indeed, as in most of the rental categories for whites in the^ South, five years of school is modal for age twelve. But Negroes who have completed eight or more years of school also are present in all economic status groups.

Following this illustration of the type of data available in this Census monograph, attention will now be given very much more briefly to two other ages, the fourteen and seventeen-year-olds, using total figures only and thus combining urban and rural non-farm. The former age is chosen because it normally marks the end of the traditional grammar school education; the latter, because it is the last year for which information is given in the Census, and is a voluntary school attendance age in most states.

For the nation as a whole, the largest single group of native white fourteen-year-olds has, as was just indicated, finished eight years of school. The next largest groups in order have completed seven or six years. The fourth group has finished the first year of high school. Again the girls make a better record than the boys. At this age there is i per cent more girls than boys in the population, but 20 per cent more girls than boys have finished eight years of school and about 25 per cent more have completed the first year of high school. Indeed, completion of this year ranks third with the girls by a considerable margin over the completion of six years.

When this fourteen-year-old group is considered by rental value of the home, the trends are quite similar to those already described for the twelve-year-old group. Some results of this trend are, however, quite striking. More often than any other group, boys coming from homes renting for $10 a month or less in 1940 had completed the seventh grade—one grade less than the total group had most frequently completed. The number of those who had finished only six years of schooling was larger than of those who had completed eight by almost 4 per cent. Among the girls as a whole, the ranking of years completed followed that for the total group, but with only a slight difference between the number of those finishing eight or seven years. For the boys, one year of high school became the second most frequently completed year in the rental category of $50-$74 a month. For the girls, this happened with the $30-149 rental group.

Among the fourteen-year-olds, the marked differences in the numbers and proportions completing specified grades indicated in the national figures disappeared in the $15-119 a month rental group for the girls and in the $20-$29 group for the boys. It seems probable that the better record made by the girls in most age and economic status groups is an indication of the fact that they are held more closely to the home and are perhaps more amenable to discipline than boys.

The trends for the fourteen-year-olds can be indicated by comparing the proportion of functional illiterates, using the army standard of four years or less of completed schooling, with the proportion who have finished nine or more years for the lowest and second highest rental groups.


The data for the seventeen-year-olds are also worth discussing briefly, since at this age in most states school attendance is voluntary. By large margins for both sexes and on a national basis, the largest group is that which has completed three years of high school, including over half a million persons all told, approximately one third of the total number of this age. It should be remembered that since the 1940 Census was taken as of April 1, a considerable majority of these boys and girls were in their fourth year of high school but had not completed it. Had the Census been taken three months later, the number who had completed high school would be much higher.

Again, economic status seems to be influential in determining grade status. For the boys, it is not until the $20-$29 rental group is reached that the number of years completed reach the national average of three years of high school. Below that rental level, those who had finished two years of high school are the leading group. The girls from the lowest rental group on reflect the national trend. The difference here is doubtless caused in part by the tendency of boys in the lower economic groups to seek gainful employment soon after they are legally permitted to do so.

Again, the trends for this age group can be indicated by using the national distribution for seventeen-year-olds and comparing it with those of the lowest and second highest rental groups.


It should be reiterated, and indeed the data prove, that economic status, or at least the rental value of the homes America's children live in, does not determine the educational status. For every age and both sexes, children in every rental group span the whole range of years possible, from lowest to highest. But that housing conditions have an influence on school status is also clear. The data presented were gathered before the war and before the development of the current housing shortage, foretold by sociologists fifteen years ago. Educators can hardly view without concern, therefore, the fact that of the approximately 16.5 million native white non-farm children in the United States in 1940, about one sixth lived in homes with a rental value of under $10 a month, and nearly two out of five in homes with a rental value of under $20 a month. Even granting that the same number of dollars buy better housing in rural non-farm areas than in urban, and that therefore a disproportionate number of these low rental values are in such areas, the picture is none too encouraging. There can be, and are, slums even in rural villages, and the housing in many an industrial village is not the illustration of the American standard of living to be shown to a foreigner. The forbidding columns of figures in the Census report used in this article are, then, a documentation of the educator's, yes, and society's, stake in adequate housing for American families.

As already explained, the Census presents no data for the farm population, since the farm home is not rented separately from the acres of the farm itself. It is quite possible, however, to secure for any county in the United States data on the proportion of children of any given age who are in school. When this is related to the farm income or per farm value for any given county, a measurement of school attendance (rather than grades completed) and economic status can be developed. There is some evidence, both in papers by the author's students and in a few publications, that the trends would not be wholly similar to those noted in this article for the urban and rural non-farm populations. Thus, in Minnesota 75 per cent of the boys and 82.8 per cent of the sixteen- and seventeen-year-old girls, respectively, are in school in the top county, economically measured, and 24.2 and 32.7 per cent are in the county rated lowest financially. Not only is this a tremendous variation, but the counties ranking highest in farm income and value of land and improvements per acre tend to rank in the two lower quartiles of Minnesota counties in the school attendance of this age group.2

The situation in this prosperous income state is not so favorable as in some. Clearly, however, economic status is not the sole determinant of educational status in such farm groups. This raises serious questions regarding the functional value of the education offered farm youth.

1 Bureau of the Census, Educational Attainment of Children by Rental Value of Home. Washington, D. C., 1945. The Bureau announces that unpublished figures of areas large enough to be reliable can be procured for the cost of transcribing or reproducing them. Presumably this would apply to most states and to large cities.

2 Milo Peterson and Douglas Marshall, Are Minnesota's Farm Youth in School? Bureau of Educational Research and Agricultural Experiment Station, University of Minnesota, April, 1947. See also Margaret Bright and C. S. Lively, Farm Youth in Missouri, pp. 10-13. College of Agriculture Experiment Station, Bulletin 504. June 1947.

Cite This Article as: Teachers College Record Volume 49 Number 4, 1948, p. 242-249
https://www.tcrecord.org ID Number: 5399, Date Accessed: 5/20/2022 11:57:33 PM

Purchase Reprint Rights for this article or review
Member Center
In Print
This Month's Issue