Nov 13, 2020

Revisiting Census Records

November 13, 2020

One of the first records that new genealogists turn to are census records and for good reason. American census records are the bedrock of genealogical research in the United States. In spite of that, many of us take census records for granted. After all, they’re easily found on almost any genealogical website from paid platforms such as Ancestry, Find My Past and My Heritage to free ones like Family Search. Somehow I think that genealogists have internalized the idea that something so easily accessed, by anyone, must be of little worth. That view is shortsighted and can lead to ignoring or overlooking census records, which is a big mistake. In fact, I think that census records are so important that I’m reviewing a review of them that I wrote a year and a half ago because I still think census records are not being used enough.


Federal censuses in the U.S. have been taken every ten years since nearly the beginning of the country starting in 1790. They’ve been consistently taken in spite of wars and naturals disasters. The most recent census, taken this year in 2020, was taken in the midst of a major pandemic. Although they have been taken every ten years right up until this year, for privacy reasons, only censuses taken seventy-two or more years ago are available to the public. That means that censuses through 1940 are open to the public at present. The 1950 census will be released in April of 2022. It is possible to request a copy of your own, or  your direct ancestor’s census image on a later census (but only that person’s, not the entire page) from the census bureau for a fee.


It’s important to remember that censuses were not taken to benefit genealogists, but rather for governmental purposes such as determining states’ population in order to draw local, state and federal political districts as well as for apportioning federal funding. However, censuses have long been used by genealogists to help them trace ancestors and pinpoint family groupings and locations throughout their lives. They’re a snapshot of our families taken every ten years. They can reveal impossible to find elsewhere data such as ages of family members, years of immigration and naturalization, length of current marriage, even number of children born to a woman.


For anyone who was alive between 1790 and 1940, the censuses are a resource that should not be overlooked, particularly because so many of them survive for so many people. Some states do have scattered losses of censuses, especially from the earlier years. For example, New Jersey does not have any censuses extant before 1830, and, of course, very early censuses never existed for territories that did not belong to the United States in the early period, such as Louisiana or Texas. Unfortunately, the 1890 census was nearly destroyed in a fire 1921 so only a few scattered fragments survive in a few states such as New Jersey and New York. Although I have one friend who found one of her ancestors in the 1890 census in North Carolina, I personally have never found anyone in that census in over 25 years of research for myself and others. The loss of the 1890 census is one of the great tragedies of American genealogy, although its loss can be somewhat overcome in some cases by state censuses and other documents such as tax records.


Because so many censuses survive, most of our ancestors can be found on them, although not always easily. It’s important to remember that the censuses from 1790 through 1840 only list the name of the heads of households. The other members are delineated by tick or hash marks in various columns segregated by age and sex. Because of this many genealogists do not bother with these earlier censuses. They should be used to place a family in a time and place and sometimes follow their migration from place to place. My own fourth great grandfather is found in Weare, New Hampshire in 1790, further north in Bath, New Hampshire in 1800 and 1810 and finally much further north in Jay, Vermont in 1820. Finding these censuses gave me the information I needed to trace him in other local records such as tax lists and town birth, marriage and death records. In another case in New Jersey, I was able to determine that a family most likely had five children, rather than the three that most online trees attribute to them. I based my conclusion on the fact that the 1820 and 1830 census showed two additional children who fit neatly into two gaps between the three known children.


From 1850 forward, all members of a household are listed by name, although it is important to note that censuses often list people by a variety of names, including nicknames and sometimes middle names. This can make it challenging to figure out whether you are dealing with one person or more. All of the censuses are indexed, but the indexing was done by various individuals, including non-native English speakers in some cases. Because each of the platforms which contain the census records used different indexers, it is often helpful to try to find a “missing” ancestor on a variety of websites. Different people may have interpreted the same name in a variety of ways. Many genealogists maintain that their ancestors were not counted in a specific census. While a few people may have been overlooked, in my experience, it is more often the case that the searcher just hasn’t searched for the “right” name which is often caused by an enumerator who just couldn’t spell or misheard what the respondent said or because an indexer misinterpreted what was written.


Next week I’ll write more about finding census records and what information the various censuses might provide.


Carol Stetser

Researcher/Director at Large