The census did not include a question on educational attainment (e.g., in the form of years of school completed, or diplomas and degrees) until 1940. Prior to that, the only questions on education from 1840 to 1930 were on the ability to read and/or write (in any language), and these questions were limited to the free population prior to 1870. Based on the data collected, information on illiteracy can be presented in the form of the percent illiterate (i.e., the percent of persons who were not able to both read and write). Data for 1840 are not shown here because the form of data collection (tallies of the number of household members rather than data for each individual) raises issues of comparability and because the data are not available by sex. In addition, information collected on illiteracy in 1850 and 1860 is not totally comparable with data collected in subsequent censuses; however, data for 1850 and 1860 are included here with the assumption that all slaves were illiterate.
Data on educational attainment were collected in the census starting in 1940. From 1940 to 1980, the data were collected in the form of years of school completed. In 1990 and 2000, data on the highest level of school completed were collected in the form of diploma (e.g., high school) or degree (e.g., college), with highest grade completed for those without a high school diploma. Thus data on educational attainment for 1980 and 1990 are not totally comparable. The summary measures of educational attainment shown are the percentages with a high school diploma (or years of high school completed) and with a bachelor’s degree (or 4 or more years of college completed). It is unfortunate that data on educational attainment were not collected in the census until 1940; however, the educational attainment of the population 25 years and over in previous censuses can be estimated based on 1940 data for the population 35+ in 1930, 45+ in 1920, and 55+ in 1910, although the estimates are subject to some error for various reasons, including the potential effects of immigration and mortality on population composition. The following estimates were developed by John K, Folger and Charles B. Nam (1967), Education of the American Population. Their estimates of the percentages with less than 5 years of school, 4 years of high school or more education, and 4 or more years of college, respectively, are as follows: for 1910 — 23.8, 13.5, and 2.7; 1920 — 22.0, 16.4, and 3.3; and 1930 — 17.5, 19.1, and 3.9. The corresponding percentages for 1940, as shown in Figure 10-3, are 13.7, 24.5, and 4.6.