Tackling Census

Posted on by


For most whose ancestry has been a while in the U.S., the census is the framework of a family history. It shouldn’t be the only source consulted, considering sometimes census shows ages off by more than just one year, but it does help document relationships and where families were residing at various times. The 1790 (first year U.S. took census) to the 1940 (the most recently released, April 2012) are available for public access. In 1921 there was a fire where the census was kept and most of the 1890 schedules were destroyed. It was decided that those 1890’s that survived would also be destroyed (typical government logic!) but even some of those survived the second destruction, and those are the ones available on microfilm today. From 1790 to 1840, only the names of heads of household are listed. All of the household members are listed from 1850 to present. The last released is 1940, but having filled the forms for the 1990, 2000, and 2010 Census, as well as my having worked in administration for the Census Bureau those years, I know that’s still the way it’s done.

The 21st Century has brought major breakthroughs in genealogical research. Especially when it comes to census research. One instance comes to mind, in the early 1980s and before Internet, when the Palm Springs Genealogical Society (of which I was charter member) had just concluded a meeting at the Palm Springs Library. After the meeting, quite a few of us stayed, to do research in the library’s Reference section. In my early 20s then, I remember a sixty-ish gentleman  telling us about some new technology that he’d read. That these new “personal computers” that were becoming popular would also be able to show census pages, without the requirement of a microfilm reader. A few years later, census “rolls” on CDs were made available for computers, and over a decade after that, I saw my first online census images.

Today, I have easy access to census records online, most times even indexed with convenient links to the specific census page. When I first started genealogical research, my methods in census research were similar to how census research was done in previous decades. That is, decide on a census roll that contained the year and area an ancestor lived. Next, obtain that roll by either visiting a National Archives location, an LDS library, borrowing through an inter-library loan from a lending-library or public library (taking about a month to receive!), or purchasing the roll by mail-order from either the National Archives or a company like AGLL. Once I got the roll, I would have to do a line-by-line search for the household(s) I wanted. If I were fortunate, I might have access to a census index book for the year and state I needed. Or, the Soundex/Miracode index for 1880, 1900, and 1910 census, which was only available on microfilm and covered just selected states. Yes, sometimes I had to order a Soundex/Miracode microfilm roll through inter-library loan. In the case of 1880 Soundex, it only covered households with children aged 10 and younger. Then, if I were lucky enough to find ancestor’s household in an index, I had to locate it on the actual census. Which wasn’t always easy because, depending on index, the household might be indexed by page number, sheet number, or household number.

Needless to say, it’s so much easier nowadays. However, there may come times when a household cannot be located, despite existence of census for the state and county in question and the convenience of an online census database. There is possibility of a household being overlooked by census-taker, or being between local moves and missed enumeration, or have moved out of the area. However, if all indications show the household should have been in a certain location that year, it is possible the name was misspelled by enumerator or wrongly guessed at by the indexer, that the household gets lost in the databases of sites such as  Ancestry, HeritageQuest, and FamilySearch.

  • Finding a Page on the Census:If you’re using the census index for the first time, the page numbers take a while to get used to. In fact, I didn’t have them figured out until after 20 years of research! There are two sets of numbering on the pages. The first set is the one that is most of the time referenced in census indexes prior to 1880. These page numbers are rubber-ink-stamped. This would have been done at the census office, to number them by county. (I personally call this “county numbering.”) Every other page is marked with an stamped page number, though prior to 1860 this may be in large thick handwriting rather than stamped. The stamp is usually found in the upper right corner, but pre-1850 is sometimes found stamped vertically at the bottom right. Though it isn’t written on the pages, think of the stamped page as [page number] “A” and the next (unstamped) page as [page number] “B” throughout the numbering system. This is how you will locate the page you want, when you find a page number in a published census index. However, the index doesn’t reference the exact “A” or “B” page, so if you don’t find it on the stamped numbered page, go on to the next (unstamped) page, which shares the same page number.The second numbering, which I personally call “township numbering,” shows up beginning with the 1870 census, in addition to the stamped paging. In 1870 and 1880 these were referred to also as “page numbers” and numbering ran consecutively for each township/district, but beginning 1900 they were called “sheet numbers” and the numbering ran with an “A” and “B” suffix. For example, 9A and 9B, 10A, 10 B. (The “A” sheet is usually the same page as the stamped number.) These were numbered by the enumerator before the census office stamped the schedules with its county numbers. Each township or division began with “sheet number 1,” which differed from the stamped numbering that ran consecutively for the whole county. The stamped numbering is usually found in the top middle of every other page (although I have found some instances where the stamp was towards the right corner). Township numbering was done in the enumerator’s hand, in the category provided at the upper right corner page (or left hand corner in 1870 and 1880). The township page numbers are the easiest to read, compared to the stamped numbers that didn’t always impress all its ink. As of 1880 “Enumeration Districts” were introduced and along with this “ED” number, sheet numbers began to be used as reference for the soundex/miracode indexes rather than the stamped numbers. As in the case with the pre-1880 stamped number references, a sheet number reference could refer to either the “a” or “b” pages of that census schedule.THE EXCEPTIONS: In some 1860 census I have seen where there are no a and b pages, but all stamped numbers run consecutively. The odd numbers were stamped and the even number pages were unstamped. Most census indexes use only these stamped numbers for reference, though I’ve found some cases the 1850 or prior will list household/visitation numbers or even line numbers instead of page numbers. In 1880 and after, the “township numbering” used a pre-printed suffix A, B, C, and D. Sometimes the rubber stamped numbering is indexed with the C and D, too, like the “township numbering” instead of restarting A and B. So how do you find a listing, with all these ways of indexing? Basically, through trial and error, because the one doing the indexing could choose however he wanted to refernce a census page. But at least with these guidelines you can pick one of the other “systems” if you can’t find the household corresponding to the number the index gives.
  • Other Census:Besides the Federal census, many states had their own census between the decades the Federal census was taken. For example, Illinois had census for 1855 and 1865 (and I think also 1845?). Some states and territories had census taken, available on microfilm, as late as the 1880s and 1890s, while some were done as early as 1809 (Mississippi Territory, which includes modern-day Alabama).
  • Where Do I Find Census Records?:Most libraries have census microfilm available through Interlibrary Loan. Check with your librarian about borrowing procedures and loan fees, as it differs by library. In the 1980s I used to pay $1 for each roll borrowed (two week loan). The current fee is $3 per roll. The drawback on Interlibrary Loans (at least for Southern CA libraries) is the waiting period, which took me anywhere from one to three months to receive requested microfilm! I later began going through AGLL (American Genealogical Lending Library) that charged a membership fee but sent the film directly to me, gave me one month of viewing for my rental fee, and let me deduct rental price from cost of film if I chose to purchase it. Although I’ve never attempted an Interlibrary Loan through them, I’m aware the LDS “Family History Centers” (located in towns throughout the U.S.) can also obtain microfilm for viewing at their center. If the center is large enough, it may stock the rolls on site and you don’t have to wait to borrow them. Some smaller centers have a partial stock of census microfilm while the larger ones, such as in Phoenix and largest one in Salt Lake City, have the complete stock. The National Archives, a branch of the U.S. Government, has all the census available at its depositories scattered across the U.S. I used to visit the one in Laguna Niguel, CA. And, last but not least, is Ancestry.com’s online census subscription. This is the one I use now and it’s been a real timesaver. Not only are their census schedules indexed (1790-1870 and 1920-1930), but you can do a nationwide search instead of a state-by-state that regular soundexes offer. As the Family Search site offers a free nationwide 1880 census index search, most of the available census years are covered by index between these two sites. Just with the exception of the 1900 and 1910 for now. Various years of census have been published by genealogical societies and other genealogical groups. From 1790 to 1830, most have been published by state. From 1840 to present, most have been narrowed down to a specific county rather than the whole state. These transcriptions are a good reference, but I suggest that the researcher double-check the published census listings with the original microfilm if at all possible. Typos are easy to make, not to mention the different ways of deciphering illegible handwriting. One genealogical society transcribed my great-great-great grandfather as “Josiah” Loy rather than “Joseph” because the tail of the “p” was fainter than the rest of the writing. In the same household, his son (my grandmother’s grandfather) John was listed as a 7 year old female while his sister Rachel Elizabeth was listed as a 12 year old male. The names were just mixed–he was the 12 year old male and she the 7 year old female. (I won’t go into detail that most everyone listed that year was a year older than they SHOULD have been that fall of 1850.) The transcription “corrected” the gender of the children but the ages should have been switched back as well!