I wouldn’t have a decent genealogy blog if I didn’t talk about the upcoming release of the 1950 Census, which the National Archives and Records Association (NARA) will release on April 1st of this year - less than two months away. I’ve summarized the important things you need to know about this census below, including why you’ll be able to search it right away.
Will be found on a free website from the NARA
You should be able to access these records for free, an Amazon Web Services open data set picked up by FamilySearch and others but also directly from a census dedicated website built by the NARA: ““Employees from across the agency have worked on digitizing and indexing the records and developing and testing a new, dedicated 1950 Census website,” said Project Manager Carol Lagundo, who leads the 1950 Census project at National Archives” (Victoria Macchi, National Archive News). You will also be able to download the whole data set from Amazon Web Services.
An index created by Artificial Intelligence, checked by human volunteers
According to FamilySearch.org, “Ancestry's sophisticated artificial intelligence and handwriting recognition technology will save time by creating an initial index from digital census images. While this index won’t be perfect, it will expedite the review and publication process to let you search for your family’s names in the 1950 census sooner after its release” after which “volunteers helping with the 1950 census indexing will be invited to review the automated index to ensure that every name is included and indexed correctly” (FamilySearch). This is different from past census releases where people basically created the index from scratch and took several months to complete.
We CAN search it!
We’ve been told previously that at the time of the release, the census will not be indexed and to expect limited search functions - but this may not be 100% true. Before you go through the efforts of researching each ancestor’s applicable enumeration district (utilize the Stephen Morse Unified ED finder), read on! Just recently NARA told us we will be able to search this initial data set, though it’s unclear how reliable it will be: “the new website will include a name search function powered by an Artificial Intelligence/Machine Learning (AI/ML) and Optical Character Recognition (OCR) technology tool” (Victoria Macchi, National Archive News). This is very exciting!
In another article, they explain “You will be able to search the 1950 Census website by name and location. You will also be able to search by Indian Reservation for form P8 Indian Reservation Schedules. To develop the initial name index, we are using Amazon Web Services’ artificial intelligence / optical character recognition (AI/OCR) Textract tool to extract the handwritten names from the digitized 1950 Census population schedules. Because the initial name index is built on optical character recognition (OCR) technology, it will not be 100-percent accurate” (“1950 Census Records” Archive.gov)
A Community Effort
This group effort by FamilySearch and Ancestry, along with many genealogical societies and organizations as well as individual volunteers, is called the “1950 US Census Community Project”. My company, Start Researching, has joined the project and is excited to help index these records as well! You can sign up for updates and request to volunteer here: FamilySearch.org/1950-Census
What Questions will be on the Census?
Two great resources for this are “Questions Asked on the 1950 Census” on The National Archives site, and for a more brief synopsis: “What Questions Appeared in the 1950 Census?” by Jason Wright, published 12 Dec 2021 on FamilySearch.
View a blank form P1: Census of Population and Housing (pdf file)
Enumeration Date: April 1st, 1950
Keep in mind the enumeration date of the census was April 1, 1950. That means all the census takers were told to record information for the household as it was on that date, even if they had arrived to the house a few weeks or months later. So if a relative died in March 1950 they will likely not appear in the census, but someone who died in May should be included. Within the same logic, only an infant born in January, February, or March of 1950 or earlier (or also on April 1st) will be included.
Read the Enumerator Instructions
Full census enumerator instructions are available here on Census.gov, on the page click the link to open the pdf download. Understanding these instructions given to census takers can help you figure out how to find certain ancestors, from those in the military to those in boarding houses. For example, in 1950 students in college were to be counted at their colleges, instead of at their homes like in past censuses.
More information available on “Instructions for Enumerators and the Public” on Archive.org.
Also helpful is the “1950 Overview” on Census.gov details how the census was conducted in brief, with further details available as a downloadable zip file of 17 chapters. Lots of detailed information to comb through in those files.
A Variety of Forms
Most of the census data was recorded on the Form P1, Population and Housing (pictured above) with some exceptions, like Form P8, Indian Reservation Schedule. Different forms were also used in some areas like Alaska, Hawaii, American Samoa, Guam, Panama Canal Zone, Puerto Rico, and the U.S. Virgin Islands, as well as “selected areas in Michigan and Ohio” (Archive.gov), and various forms for military, crew members, and others.
View more detailed information about each of the forms at “Census Forms in the 1950 Census Dataset” at Archives.gov
Counted Later - ED 71+
If there was no one home when the census taker came by, the census taker (also called enumerator) had a few different options. When the enumerator chose to come back later, and therefore had information out of order, they would start on sheet 71 or later. “The enumerator was to number the sheets (pages) of the census form in order beginning with “1” (one). Nearly all Enumeration Districts could be completed on fewer than 70 pages. Persons enumerated out of order were counted on sheets beginning with the number “71.” (Archive.gov)
So if you can’t find your ancestor’s family in their expected district with their neighbors, try the later sheets of the same area.
Read “1950 Census: Enumerated Out-of-Order: “Callbacks” and Others on Page 71 and Up” by Claire Kluskens on HistoryHub
Still Missed Some People
Despite best efforts, which included a post-enumeration survey of sample groups and “Missed Person” notices published in newspapers, “The Bureau of the Census later estimated that the census undercounted (missed) approximately 4.1 percent of the population” (Archive.gov).
The full U.S. population that was counted, and determined, by the 1950 census was 150,697,361 people (Wikipedia), with the highest populations in New York, California, and Pennsylvania. If we think about 4.1% of that number, we’d have about 6,178,591 people missing from the 1950 census, or just over 6 million individuals not counted.
Bonus - How to Prepare?
If you haven’t done so yet, now is a great time to look through your ancestors and make a list of those alive in 1950 (specifically April 1st, 1950). Determine the following:
Your ancestor’s full name, along with maiden name, alternate spellings, and nicknames
Names of spouse and children
Ages of all (to help confirm you have the right family)
Birth places of all above (again, to help you confirm you have the right family)
Known addresses. Look at other records if you need help determining, starting with where they were in the 1940 Census, then other records like draft cards and city directories.
Enumeration districts of those addresses (may not be needed, but helpful to have)
Have this reference list with you so you’re ready to search come April 1st.
Best of luck!
Article by: Sherri Mastrangelo, published 6 February, 2022.