Genealogy Workbook: US Census

Genealogy Workbook: US Census

Your complete guide to finding and using your ancestors' US census records in your genealogy research.

Imagine someone showing up at your American ancestors’ door every 10 years and taking a group photo of the household. You’d want all of those pictures, wouldn’t you? They’d show a time-lapse biography of your family. You’d study each person’s age progression, how hairstyles and fashions changed, who appeared or disappeared (and when) and how the backdrop changed over time.

The US census is that decade-by-decade picture—albeit in words. Every decade, census-takers compiled a list of nearly all American households. As time went on, they gathered slightly different—but increasingly more—information. By the late 1800s, censuses could tell us who lived with whom and how they were related, how well they lived, what work they did, and sometimes when or where they were born, married and died.

Censuses are so data-rich that they’re among the first places you should look for your family history—and you should keep coming back to them. It’s difficult to absorb everything on the first pass. Revisit your ancestors in censuses to find clues you missed, interpret new finds or confirm you’re on the right track. We’ll show you what’s in census records, how to find your ancestors’ records, and how to extract every possible clue.

A national head count

The US Constitution calls for a nationwide population tally every 10 years for congressional representation purposes. The first was in 1790; the most recent was in 2010. Additionally, the federal government picked up part of the tab for an 1885 census of some states and territories, so these count as federal censuses, too.

Census-takers were to visit each household. It might have taken weeks or months to visit everyone, but all information collected was supposed to reflect the household as it was on a specific date. The official census date (see a list of official US census dates) is important when you’re calculating household members’ birth years based on the age in a census, or if someone was born or died around the time a census was taken.

Over time, standardized questions, census forms and procedures developed. Enumerators received instructions about the direction to proceed around neighborhoods, the intent behind questions and how to code a variety of answers. For some censuses (see below), enumerators asked a set of supplemental questions of those in certain population segments, such as veterans or those with disabilities, to capture social, economic and public health information.
Tip: Look up every ancestor in every census and note each piece of information you find on a census extraction form.

Questions, questions

Genealogists might cluster US censuses into three groups: the least informative (1790-1840), fairly informative (1850-1870) and genealogical jackpot (1880-1940). Those early censuses name only heads of household and count others. The 1830 census was the first to use a standard preprinted form. Here’s a breakdown:

• 1790-1810: name the head of household with tallies of other household members by age group, gender, race (white, black or Indian), and free or slave status
• 1820: same as above, plus tallies of aliens and workers in agriculture, commerce and manufacture

Sample Record: 1820 Census
Citation for this record: Connecticut, New London, Bozrah. 1820 US census, population schedule. Digital image. accessed 15 January 2014.  
1. The earliest censuses didn’t use preprinted forms. This page doesn’t have column headers. Using a blank census form, as shown here, makes the census easier to read.
2. Without a preprinted form, missing columns are possible. When you extract data onto a blank census form, note any missing columns and take this into account when interpreting data.
3. Only head-of-household names appear in this census. Use the tally marks to identify likely household members based on what you know from later censuses and other sources. Remember, not all may be relatives.
4. This list is in alphabetical (not household) order, a rearrangement occasionally done by early census takers. This makes it easier to spot others of the same surname. If you’re looking for the neighbors, check maps, land records and other sources that would show who lived where.
• 1830: drops the industrial categories and added tallies of the deaf, “dumb” and blind, segregated by race and age
• 1840: reintroduces employment tallies for various industries; adds military veterans receiving pensions; and tallies those in school, illiterate white adults, and “insane” and “idiots” cared for at public expense


Beginning in 1850, every free household member was named, and homes and families were numbered in order of visitation.

• 1850: name, age, sex and color of each free person; value of real estate owned; place of birth; whether married or attended school within the year; whether an illiterate adult or “deaf and dumb, blind, insane, idiotic, pauper or convict;” and occupation of males over 16
• 1860: adds value of personal property
• 1870: adds more-detailed age, race and literacy descriptors. This census adds indicators for parents of foreign birth; month of birth or marriage if occurring within the year; voter eligibility; and whether a victim of voter discrimination. Names the formerly enslaved for the first time.
Beginning in 1880, the level of family history detail in the census rises to a new high. Home addresses, family relationships and more put these among the richest existing sources for American genealogy.
• 1880: adds street address, relationship of each person to head of household, marital status, everyone’s profession, months unemployed, illness or disability on date of visit, parents’ birthplaces
• 1890: adds more detailed capture of race, household structure, naturalization status and disability/illness; number of children borne/living and years in the United States; months attending school; parents’ birthplaces; whether a homeless child, convict, prisoner or Civil War veteran/widow; native tongue (if not English). Sadly, all but a handful of schedules for this census were ruined after a Census Bureau fire.

• 1900: removes native tongue, adds month and year of birth, whether an English speaker, years in present marriage, year of immigration/number of years in the United States, homeownership/renter status.
Sample Record: 1900 Census
Citation for this record: Missouri, Camden, Osage. 1900 US census, population schedule. Digital image.; accessed 27 March 2014. 
1. Note the actual date the census was taken—June 11. The information recorded should reflect the official census date, June 1, but sometimes births or deaths occurring in the interim weren’t accurately logged.
2. Later censuses like this one collected a variety of data. This one requests birth month, year and place for household members, and birthplaces of their parents. Extract all the information (including what you already knew) onto a blank census form.
3. Households are numbered in the order the census taker visited them. Browse through a few pages for familiar names and neighborhood makeup, such as a dominant ethnicity and common occupations.
4. The occupation column may include nonworking terms such as at school (for students), pauper, pensioner and at home (for homemakers). These clues may point you to school, poor relief, pension and other records.

• 1910: removes birth month and year, adds occupational and unemployment details, Civil War veteran status, and language of non-English speakers. Less detail requested on schooling and disability.
• 1920: removes childbearing data, adds year of naturalization, mother tongue for self and both parents, whether living on a farm
• 1930: adds value of home/rental, radio ownership, age at first marriage, mother tongue of foreign born, English language ability, unemployment, military veterans mobilized (and for which war)

• 1940: adds informant to the census taker; highest school grade completed; place of residence on April 1, 1935; and details on employment status, occupation, involvement in public emergency work, income amount and source.
A sample was also asked for parents’ birthplaces; mother tongue; whether a veteran or veteran’s spouse, widow or child; whether in possession of a Social Security number; whether paid into federal retirement programs; usual occupation; and marital and childbearing history.


Search all censuses

Original and microfilmed population schedules are at the National Archives and Records Administration (NARA), but most originals are closed to researchers. It’s easier to find relatives by searching digitized, indexed versions of decennial censuses at major genealogy websites. Online versions of the 1885 population censuses for most or all of Colorado, Florida and Nebraska, and the territories of Dakota and New Mexico, are at

Try to locate each relative in every census taken during that person’s life. Begin with the most recent census in which someone should appear and work back through time (then forward, to help verify your death date for the person). You’ll be able to track migrations and note slightly (or widely) different answers to similar questions over the years—even for categories you’d expect to be consistent, such as age. You might discover surprises and detours in relatives’ lives.

Search first by the person’s name. If needed, use the site’s filters to narrow your search to a specific census and location (state, county or town). Restrict your results to see just census records (there’s usually a filter called something like “restrict by record type”). Then broaden the location if you can’t find the person.

Census takers didn’t ask people how to spell their names, so misspellings are common. In addition, indexers may have made errors when transcribing the names to create searchable indexes. To overcome these problems, try spelling variations, nicknames, different combinations of first and middle names and initials, and both maiden and married names for women. If the site allows it, you can leave out the name and search on other variables. Keep track of your searches in the Online Database Search Tracker.

Still can’t find someone? Browse the census in a probable location. You can do this on several genealogy data websites. For example, at, select the right census from the card catalog and choose Browse This Collection. On, go to Search, then scroll to and select Browse All Published Collections, then scroll to United States Census and select the right year. Scroll to the Browse option. In, you must find someone in the census first (such as a neighbor from an earlier or later census), then page forward or backward from there.

Beginning in 1880, you may need to know the enumeration district (ED) as well as the town or county to browse by location (the city of Chicago alone has nearly 200 EDs in 1880). One-Step Webpages by Stephen P. Morse offers several tools for finding EDs; on the home page, scroll to the US census section.

At times the census missed people. Almost all of 1890 census was destroyed in a fire, and significant portions of the 1790 and 1810 censuses are lost. In such cases, look to census substitutes—records created around the same time with similar information. These include state censuses, tax records, school censuses, voter lists and other local enumerations. Consult state and local genealogical guides for availability.

Compile the data

As you move across time through censuses, a household picture emerges. A single person marries, children are born and leave home, relatives move in or out, spouses die, a widower remarries. Keep track of all this data properly and you’ll make a lot better sense of it.
Every time you find a family member in a census, attach the source information to the individual’s profile in your online family tree and/or cite it in your family tree software. Save the record image, too (you could use a photo editor to add source information right on the document). If you’re paper-based, print the census image, jot the source information on it, and keep it in that family’s file.
Next, extract every piece of data you find, even if you think you’ll remember it or it doesn’t seem valuable. Census worksheets give you handy blank census forms to copy in your ancestors’ stats under preprinted column headers.
Pay special attention to the “relationship to head of household” column that first appears in 1880. The head of household (usually a man) is designated on the census. Keep in mind that person’s child may not also be the spouse’s child.
You also can use the Census Record Extraction form to create an at-a-glance summary of several types of census data (most useful beginning in 1850). Remember: The same information may be requested in different ways in each census, and even presumably “unchanging” information like birth year may be reported differently over time. The two reports of the same data are valuable, so log each year’s statement for each category.
It’s not uncommon to find someone who’s 22 in one census, then 30 ten years later, then 39 ten years after that. Before birth certificates and qualifying for Social Security, many people didn’t know or care exactly when they were born. Others misrepresented ages to enroll in the military or marry, and may have kept up the lie. Officials didn’t have a way to double-check for accuracy.
Discrepancies may show up for other reasons. The census taker’s source of information may have been a child or neighbor. Language barriers and cultural prejudices could result in miscommunication and inaccurate writeups. Mistrust of the government may have led to lies, particularly about income or naturalization status. Someone wishing to hide a former marriage or childbirth might not report these events. Consider which of these reasons might best apply in your ancestor’s case and look for other records to confirm your suspicions.
Censuses often contain clues to extended family. The easiest way to jump back a generation is to find an aging parent living with an adult child. Use the head-of-household column to confirm the relationship. Remember that a “mother-in-law” is the head-of-household’s mother-in-law, not the spouse’s. Other strategies include:
• For 1850 or later, look for your relative as a young person in a parent’s household. You’ll be more confident you’ve got a match if your ancestor had an unusual name, if the family lived in the same neighborhood, the birth years match, and/or there’s only one reasonable search result. But treat these findings as circumstantial evidence and seek other sources for the parents’ identities.
• Use noncensus records to identify parents: birth, marriage and death records; Social Security applications (SS-5); draft registrations, military service or pension paperwork and more. In city directories, an adult child might have the same residential address as the suspected parents. Once you identify the parents, return to the census and look for them.
• Browse census pages before and after your family’s listing for relatives who live nearby. Do additional research to reveal how these folks might be related.
Tip: See what special census schedules may still exist for your ancestor in each census year.

Check out special censuses

Several decennial population censuses included schedules for segments of the population, such as military veterans, slaves, the disabled, owners of farms or manufacturing businesses of a certain size, and the recently dead. These special “nonpopulation” censuses are available on microfilm and often, on the same genealogy websites as population censuses, but in different databases:

Agricultural schedules (1840-1940): These list farm owners or managers, acreage, cash value, crops and other products, livestock number and value, and value of homemade goods. Availability is spotty, though: The 1840 data remain only as statistics, schedules for 1890 to 1910 were destroyed, and individual schedules after 1910 are almost entirely unavailable (ask at state archives). But at least one farm schedule for 1850 through 1880 survives for at least 21 states. State repositories and Duke University have many originals. NARA has most on microfilm and has a good collection in a database called Selected US Federal Census Non-Population Schedules, 1850-1880.

DDD Supplemental Schedule (1880): If the 1880 population census designated any relatives as sick, disabled, blind, deaf and dumb, idiotic, insane, impoverished, homeless or a prisoner, look for them in the Schedules of Defective, Dependent and Delinquent Classes. You can search these “DDD” schedules for 21 states at For other states and original or microfilm versions, see Family Tree Magazine’s state-by-state listing of DDD schedules’ whereabouts.
Manufacturer/Industry Schedules (1810-1820, 1850-1880): Records from 1810 are largely lost. Reports for 1820 and 1850 through 1880 exist on manufacturers producing at least $500 annually in goods. These contain the name, business/product type, investment amount, values of raw materials and machinery, annual production, number of employees and labor costs. Originals or microfilmed copies are at the National Archives and its regional libraries. An index to 1820s data is listed in the toolkit. Many 1850-1880 schedules are on
Mortality Schedules (1850-1885): During each census, deaths from the previous June 1-May 31 were logged with the deceased’s gender, age, color, widow(er) status, birthplace, death month, occupation, cause of death and length of illness. The 1870 and 1880 enumerations add more detail. Original schedules are scattered across state repositories and the DAR Library, with most on microfilm from FamilySearch. Some transcriptions are searchable free at has an incomplete collection; and have 1850 schedules.
Social Statistics (1850-1880): These list no names, but they do list churches, cemeteries, fraternal organizations, clubs, schools, libraries, newspapers and other organizations in your ancestors’ communities. Originals are at NARA; major genealogical libraries have published lists in 1880. You can browse several schedules at in the aforementioned collection of nonpopulation censuses. Under Browse this Collection, choose the state and, if available, “social statistics” as schedule type.
Slave Schedules (1850-1860): These censuses name slaveowners, but generally identify their slaves only by age, sex, color (black or mulatto) and whether a fugitive or manumitted. Use other sources to confirm a possible match for an enslaved ancestor. has 1850 and 1860 slave schedules; has 1850.
Veterans Schedules (1840, 1890): Revolutionary War pensioners identified in the 1840 census appear in A Census of Pensioners for Revolutionary or Military Services, an e-book that’s free at Internet Archive and searchable on The 1890 census listed Civil War Union veterans and their widows, and some Confederate veterans. Schedules survive for half of Kentucky and the states alphabetically following it, plus Indian Territories and US ships and navy yards. You can search these schedules at and

Fast Facts

• First U.S. federal census: 1790
• Frequency: every 10 years, plus 1885 for some areas
• Latest census available: 1940
• Next census available: 1950, scheduled for release April 1, 2022
• Significant missing data in: 1790-1810, 1890
• Location of original records: National Archives and records Administration
• Research online at:,,,,
• Search terms: US census , federal census records [year]
• Alternate or substitute records: State and local censuses, tax records, voter registrations, city directories, Sanford maps (for neighborhood layout). has census substitute databases for 1890 and 1950. 


US Census Quiz

1. For what years are US censuses available to genealogical researchers?
2. What was the first census to name more than just the head of household?
3. What special schedule would you consult to learn more about a farm-owning ancestor in 1870? Which website might have this schedule online?
Exercise A: Go to’s main Search page. Sign in (an account is free). Look for Thomas Selby’s household in the 1870 census. He was born about 1815 in Pennsylvania. Restrict results to censuses only.
1. What is Thomas’ wife’s name and what is his occupation?
2. How old is Chester? What would you guess is his relationship to Thomas?
3. Who on that census page would you guess works with Thomas?
4. Write a citation for this record.
Exercise B: Choose an ancestor you want to learn more about who should appear in at least two to three US censuses. Search for him or her in every census during his lifetime. Document your findings in at least one way recommended in the article (attach to individual profile in online family tree, add as source in family tree or citation manager software, or fill out Census Record Extraction Form).
Quiz answers: 1. 1790-1940 and 1885 2. 1850 3. Agricultural, Exercise A: 1. His wife is Huldy and he is a “circuit and county clerk.” 2. Chester is 20. He may be Thomas’ son judging from their ages; finding the family in an earlier census and in other records would help confirm this. 3. David Simons (on the next line), a “judge of court” 4 1870 US Census, Osage Township, Camden County, Missouri, population schedule, p. 1 (penned), p. 512 (stamped), dwelling 2, family 2, Thomas Selby; digital images, ( : accessed 07 Apr 2014); from National Archives microfilm publication M593.

More Online

From the July/August 2014 Family Tree Magazine

Related Products

No Comments

Leave a Reply