Urban Research Using Census Records
Federal, state, and special censuses are productive genealogical tools, for probably no other records in existence contain more data about families. For additional help on census records, see the following: chapter 5, “Census Records,” of this book and chapter 9, “Censuses and Tax Lists,” by G. David Dilts, in Printed Sources: A Guide to Published Genealogical Records.
Several projects to index census records are now underway, and more electronic indexes are coming online or being published on CD-ROM almost every week. Increasingly, all persons are being indexed. As more indexes become available, many of the more traditional census search strategies in the next several paragraphs may become obsolete. Most of the indexes that are online require a personal subscription or access to a library that subscribes to one of the index services. Even when indexes are available, some names will inevitably be incorrect or will have been missed entirely, so do not conclude that an ancestor is not in the actual census schedule if his or her name is not found in an index. Because of this, it is may be necessary to rely on traditional strategies for searching censuses when indexes are not available. Searching through the census schedules for a metropolitan area often presents special problems.
The Soundex and other microfilmed and printed census indexes are helpful but somewhat limited, especially in city situations. Often, names were misspelled or completely omitted in transcription from the original schedules. For the 1880 census, only households with children ten years of age or under were listed in the Soundex index. When indexes fail or when they have yet to be created for a densely populated urban area, an educated approach is important. If you know the family’s makeup, you can go over a census line by line in hopes that names, ages, birthplaces, and other known facts will catch your eye, even though the name is misspelled or the page barely legible. However, in most cases, this can be extremely time-consuming. For the 1850 through 1870 censuses, the ward is the smallest division of the city. Space for the enumerators to identify street names and numbers did not appear on census forms until the 1880 enumeration.
The geographical arrangement of the census schedules makes finding aids vital when searching for urban residents, for in every census year some names were inadvertently left out of census indexes or were misspelled to the extent that they cannot be found in the index. Historian Keith Schlesinger devised a system to locate individuals overlooked by the Soundex. Schlesinger gleaned addresses from city directories, which he found both accurate and accessible, then plotted them on maps of census enumeration districts, which normally followed the boundaries of voting precincts in most cities. Figure 20-5 is a section of an 1880 Chicago map on which enumerator visitation dates are noted. By matching ward and visitation dates to census pages, it is possible to search for the nonindexed individual to one or two enumeration districts. By narrowing the search for the nonindexed individual to one or two enumeration districts, this scheme permits the historian to escape the confinement of the Soundex. A number of research institutions have acquired enumerator district maps and finding aids that trace the route taken during the census count. Additionally, enumeration district boundary descriptions are available on microfilm through the National Archives and Records Administration. The Newberry Library and Schlesinger have refined several methods for searching the 1850 through 1910 censuses for Chicago. See Keith Schlesinger and Peggy Tuck Sinko, “Urban Finding Aid for Manuscript Census Searches,” in National Genealogical Society Quarterly.29 The techniques, of course, are applicable to other cities as well.
For a number of years, several projects were conducted to index the 1880 federal census schedules. These efforts were made obsolete with the release of an every-name index by the Family Search program of The Church of Jesus Christ of Latter-day Saints. That index may now be searched online at either the Family Search website or from the census menu at Ancestry.com. In either case, search results are linked to images of the actual pages, but a subscription is required to view the images.
To even the casual researcher, the population explosion in the United States can be seen clearly by the increase in the number of census rolls filed for each succeeding census year. A Century of Population Growth 1790–1900, produced by the U.S. Bureau of the Census, reported that in 1790 there were but five cities having populations of eight thousand or more: Boston, New York, Philadelphia, Baltimore, and Charleston.30 In 1900 the number of cities with a population of eight thousand or more within the area enumerated in 1790 was 286, an increase of more than fifty-fold.
An excellent finding tool for thirty-nine cities in the 1910 federal population census, however, is 1910 Index to City Streets and Enumeration Districts; it has been reproduced on fifty sheets of microfiche.31 By determining the street address for an individual in a 1910 city directory, it is possible to search for the street name or number on microfiche, which is arranged alphabetically by name of city and thereafter in alphabetical order or numerical order of the street. Once the enumeration district number is determined from the microfiche, it is usually quick work to locate the actual address for the person or family on the actual census schedule. The city schedules were selected for indexing by the Census Bureau based on the frequency of requests for information. The records were originally in bound volumes but were unbound for microfilming. With the exception of several of the larger cities, the index for each city occupies a single volume. The original arrangement of the records has been preserved with the exception that the boroughs of Manhattan, the Bronx, Richmond (Staten Island), and Brooklyn are under the heading “New York City.” There is no index for the borough of Queens.
Cities indexed by street and enumeration district for the 1910 census are Akron, Ohio; Atlanta; Baltimore; Canton, Ohio; Charlotte, North Carolina; Chicago; Cleveland; Dayton, Ohio; Denver; Detroit; District of Columbia; Elizabeth, New Jersey; Erie, Pennsylvania; Fort Wayne, Indiana; Gary, Indiana; Grand Rapids, Michigan; Indianapolis; Kansas City, Kansas; Long Beach, California; Los Angeles and Los Angeles County; Newark, New Jersey; New York City (including Brooklyn, Manhattan, the Bronx, and Richmond); Oklahoma City, Oklahoma; Omaha, Nebraska; Patterson, New Jersey; Peoria, Illinois; Philadelphia; Phoenix; Reading, Pennsylvania; Richmond, Virginia; San Antonio; San Diego; San Francisco; Seattle; South Bend, Indiana; Tampa, Florida; Tulsa, Oklahoma; Wichita, Kansas; and Youngstown, Ohio.
The 1920 census, though lacking some information categories that some previous census schedules included, has Soundex indexes for every state and territory. The 1920 census consists of 2,076 rolls of population schedules and 8,585 rolls of Soundex. Between 1938 and 1940, the WPA prepared approximately 51 million Soundex index cards, based on surname, for all states and territories enumerated in the 1920 census. The indexes were generated to assist the Census Bureau in searches for individuals who needed official proof of age from a period before all states had a uniform system of registering births.
The 1920 Soundex index contains approximately 107 million names and resembles the 1910 Soundex/Miracode in format. For the 1920 Soundex, however, related members in a dwelling were enumerated on a “family card,” while boarders, servants, and the like were listed on a separate card.
In the official Instructions to Enumerators, issued by the Bureau of the Census on 1 January 1920, item 68 provided the method for canvassing a city block:
- If your district is in a city or town having a system of house numbers, canvass one block or square at a time. Do not go back and forth across the street. Begin each block at one corner, keep to the right, turn the corner, and go in and out of any court, alley, or passageway that may be included in it until the point of starting is reached. Be sure you have gone around and through the entire block before you leave it.
There is no way of knowing how precisely enumerators followed the fifty pages of small-print instructions, but a diagram was provided with item 68 to show exactly how to proceed on a block divided with alleys and courts.
The 1930 census was released to the public by the National Archives in 2002. There are 2,667 rolls of microfilm. The attached image is an example of a 1930 census schedule for a portion of Los Angeles. The basic information categories are similar to those for the 1920 census; a complete listing is at <http://1930census.archives.gov/FAQ.html>. There are Soundex indexes for only twelve southern states: Alabama, Arkansas, Florida, Georgia, Kentucky (only counties of Bell, Floyd, Harlan, Kenton, Muhlenberg, Perry, and Pike), Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Virginia, and West Virginia (only counties of Fayette, Harrison, Kanawha, Logan, McDowell, Mercer, and Raleigh). Ancestry.com has prepared an every-name index, which is available to subscribers or through participating libraries. The National Archives website has step-by-step instructions for geographic searches at “How to Research the 1930 Census Microfilm”.
To assist in geographic searches, the National Archives collected city directories for years around 1930 and deposited microfilm copies in research rooms in the National Archives Building and the various regional archives. They are not microfilm publications and are not copied by the National Archives. The agency has issued three microfilm publications related to enumeration districts. Written descriptions are in Descriptions of Enumeration Districts, 1830–1950 (156 rolls) and are arranged alphabetically by state, county, and city and thereunder by supervisor’s district.32 The descriptions for 1930 are on rolls 61 through 90. Maps showing the boundaries and number of each district are in Enumeration District Maps for the Fifteenth Census of the United States, 1930 (36 rolls).33 Street addresses are cross-referenced with enumeration districts for over fifty cities in Index to Selected City Streets and Enumeration Districts, 1930 Census (7 rolls).34
The attached image is part of a page from the Census Enumeration District Descriptions (1910). It shows Boston’s twentieth ward. These finding tools are especially useful for large cities and when institutions, even if enumerated at their street addresses, were recorded at the end of the schedules for an enumeration district, as they frequently were.
Establishing the whereabouts of your own urban ancestor does not exhaust the possibilities of census schedules. The microfilm contains the raw data necessary to understand the life of a neighborhood. What, for example, was the ethnic and occupational makeup of the street? Did the family settle with or near relatives? When combined with historical or sociological studies, a census can provide a better insight into what it meant to live in a nineteenth-century city.
State censuses, mortality schedules, and other special enumerations should not be overlooked as potential city sources, though few of them are indexed. The New York state census for 1855, for example, is far more important than the federal census because it includes information on the value and construction of the dwelling; the number of families occupying it; household members by name, age, sex, and relation to the head of the family; state or country of birth for each; marital status; profession, trade, or occupation; number of years resident in the city; voting status; and literacy of adults. The 1855 census-takers took the local election districts in each city ward for their districts, enabling the searcher to focus on a desired household faster than in a corresponding federal census, which was organized by ward. While this census does not specify bounds of election districts for New York City, on 7 November 1854 the New York Times published the polling places of each of the 128 election districts in the twenty-two wards for the previous election. A number of state and other special censuses have been microfilmed by the Genealogical Society of Utah. The listings for the New York state census for the city of Rochester are shown in the attached image. Other states took statistics that are particularly valuable to genealogists. The 1925 Iowa state census lists name, place of abode, relationship to head of household, citizenship, education, names of parents (including mother’s maiden name), nativity of parents, place of marriage of parents, military service, occupation, and religion. There is a separate index for the 1925 Iowa census, but many state and other special census schedules are not indexed.