The entire 1930 census produced thirty-three volumes of data, including six volumes of population data and two of unemployment information. The number of published pages amounted to 35,700. The bulk of this was published within the decennial period, which extended from January 1930 through December 1932, although several volumes were printed in 1933 and 1934. At least one volume of 1930 data was published as late as 1937.38

As one might expect, many of the demographic trends revealed by the 1930 census were continuations of those that had made the reapportionment fight over the 1920 census so bitter and protracted. The proportion of the population that lived in cities of 100,000 people or more increased from 25.9 percent to 29.6 percent between 1920 and 1930. The total urban population grew from 51.4 percent to 56.2 percent.39 The fastest growing region of the country was the West. The big winner in the reapportionment that followed the 1930 census was California, which gained nine seats in the House of Representatives, nearly doubling the size of its congressional delegation.40

The Last "Traditional" Census

In several ways, the 1930 census can be thought of as the last traditionally conceptualized and organized census.41 Beginning in 1940, a census of housing was added to the census of population in order to determine the extent of the nation's housing needs in the wake of the Great Depression. Millions of people had been forced out of their homes by mortgage foreclosure, repeated crop failures, and forced migration. Millions more lived in dwellings that lacked heat, electricity, or running water. The housing census helped clarify the scope and geographic distribution of this problem and forced public officials to develop ways to alleviate it.42

The 1930 census was the last to ask virtually all questions of the entire population. The inclusion of probability sampling in the census was another innovation for 1940. Probability sampling could not have been included prior to 1940 because the methodology of applying it to human populations was developed during the 1930s. By incorporating sampling methodology into the census, it became possible to obtain accurate, detailed information, with known margins of error, without unduly increasing respondent burden and at relatively little added cost.43

The 1940 census was the first to include a question on income. Many observers believe that one of the causes of the depression of the 1930s was the maldistribution of income. They believe that an economy that produced more goods and services than its people could consume or than could be exported was structurally imbalanced. Although the initial effort to include income in the census was relatively limited (asking only for the amount of money wages or salary up to five thousand dollars and whether the respondent received more than fifty dollars from sources other than wages or salary) and met with substantial resistance from legislators and portions of the population, the income question has become a standard inquiry in censuses ever since.44 For those not willing to tell the enumerator their income, the 1940 census included a confidential income report form that allowed respondents to mail their answers to the income question to the Census Bureau.45 To reduce opposition further, the question was moved from the 100-percent part of the census to the sample in 1950.

Finally, the 1930 census was the last one that did not include an effort to obtain quantitative assessments of how comprehensively the census had counted the population. The initial quantitative evaluation of the 1940 census consisted of comparing aggregate data on the number of men who registered for the draft during World War II with 1940 census data on the number of men in the same age group. The study concluded that the census contained an undercount of males of draft age and that the undercount was significantly higher for African Americans than for Whites.46 Beginning with the 1950 census, all American population and housing censuses have included quantitative evaluations of census coverage.

In spite of all the improvements that have taken place in American census taking since 1930, the process of obtaining information from people who do not respond voluntarily remains fundamentally the same. It involves identifying, testing, hiring, training, and sending tens of thousands of enumerators onto the streets of the cities, towns, and hamlets of the United States to knock on doors, ask the questions on the report form, record the responses, and send the completed questionnaires to facilities where they are aggregated, tabulated, and released to the public to be used and misused in the innumerable ways that Americans have devised since the data from the first U.S. census were published in 1791.

Notes

I would like to thank Connie Potter and Bill Creech of the National Archives for helping me locate the records used to prepare this article. Bill Maury, chief of the History Staff of the Bureau of the Census, supported this project, carefully edited the draft, and redistributed office work to give me time to complete this article. The Census Bureau's reviewers included Tom Jones, Jerry Gates, Campbell Gibson, and Tommy Wright; I want to thank them for giving of their time and expertise. Michele Freda and Cathie Childress of the Census Bureau's Public Information Office helped me find and duplicate the photographs reproduced in this article. My colleague, Shannon Parsley, helped with all aspects of this work including brainstorming, editing, and photo selection, and contributed mightily to its completion.

This paper reports the results of research and analysis undertaken by the U.S. Census Bureau Staff. It has undergone a Census Bureau review more limited in scope than that given to official Census Bureau publications. This report is released to inform interested parties of ongoing research and to encourage discussion of work in progress.

1 W. C. Bailey to Director of the Census, June 16, 1930, Entry 215, box 231, Record Group (RG) 29, National Archives Building (NAB), Washington, DC. Many complaints dealt with pay. A number of letters to Bailey echoed the comment of Oliver Butterfield, who wrote, "The pay we will receive for this work is not in keeping with the efforts the enumerator must use to complete it; why our government expects a complete census for such a small wage is beyond me." Butterfield to Bailey, Apr. 16, 1930, Entry 215, box 231, RG 29, NAB.

2 S. J. Norton to Bailey, Apr. 19, 1930, ibid.

3 Anna M. Pevoto to Bailey, Apr. 19, 1930, ibid.

4 Bessie L. Erickson to Dr. Bailey, May 13, 1930, ibid.

5 The literature on the causes of the depression of the 1930s is vast and contentious. A good, recent overview of the depression in the United States is David M. Kennedy, Freedom From Fear: The American People in Depression and War, 1929 - 1945 (1999), pp. 10 - 103. A broader account that discusses both the European and American contexts and that emphasizes multiple causation is Charles P. Kindleberger, The World in Depression, 1929 - 1939 (rev. ed. 1986).

6 See Margo J. Anderson, The American Census: A Social History (1988), chap. 6. The text of the law authorizing the 1930 census is in U.S. Department of Commerce, Bureau of the Census, Census Bureau Legislation, comp. Robert H. Holley (1936), pp. 29 - 37.

7 U.S. Bureau of the Census, Annual Report of the Director of the Census to the Secretary of Commerce for the Fiscal Year ended June 30, 1921 (Annual Report, 1921) (1921), p. 30.

8 For the 1910 census of population, Census Day was April 15; between 1830 and 1900, Census Day had been June 1.

9 April 1 has remained Census Day since it was first designated as such in 1930.

10 Annual Report, 1927, p. 2.

11 Annual Report, 1928, p. 2; Annual Report, 1929, p. 2; Joseph A. Hill, "Progress of Work in the Census Bureau," Journal of the American Statistical Association (December 1927): 511. For some cities no maps were available; bureau officials had to obtain descriptions of enumeration districts from local residents. See Annual Report, 1928, p. 2. By mid-1929, the Census Bureau had contacted about 85,000 people and received information useful for delineating enumeration districts and for establishing local enumerator pay rates.