My youngest sister born in 1937 is not listed in family household on sheet 7B, ward 1, area E, ED 30-23, resident code 8414, enumerated 19 April 1940. She definitely would have been too young to be out of the household, and cannot be found elsewhere.
How do I correct error by census taker?
Hopefully you've already searched the census for the missing individual outside of the expected household. She could have been recorded:
If truly omitted, it could be due to error by the informant (eg. forgetting a new baby in the household) or a copy error by the enumerator (similar to the example noted above).
"Correcting the error" could interpreted in 2 ways. However, an official correction (informing the U.S. Census Bureau that someone was omitted and emending the original pages and images) is not possible, especially after almost 75 years. Also, a certain error rate during the enumeration process was expected.
Corrections for the benefit of other researchers, present and future, depend on the image provider:
Ancestry allows corrections to some indexed fields on census images. But adding a new person is not possible, even if they are on the image but missed during indexing. My recourse is to add a comment to the indexed record of the head of household noting that the household contains (or should contain) additional persons, with the pertinent details.
FamilySearch does not allow corrections at this time. If the record is attached to someone in the FamilySearch Family Tree, source notes can be added. However, these are not very visible when searching the records side of the site.
There are several other providers of U.S. census images, including (images only) archive.org. I'm not aware of any mechanism for adding comments to these offerings.
It's annoying when we are looking for a family member, and the record we expected to find them in didn't record their name, or we can't find the record at all.
In my own research, I found my husband's maternal grandmother attached to a different family on Ancestry.com because she was enumerated on one of these supplemental pages. Margin notes on both her page and the page for the rest of her family indicate to the reader which household she actually belongs to.
If your sister was recorded on a different page, an indexer could have accidentally put her in the index with a different surname. I have seen many cases where the indexes don't match what is on the digital image; in one spectacular case, one of my husband's relatives was indexed with his surname as his first name, and with the surname of the family above him. His wife was also recorded with the surname of the previous family.
If you haven't already done so, I encourage you not to rely on the every-name index, but instead to browse all the census images (sometimes referred to as "walking through the district") to make sure your sister is not somewhere else in the ED on a supplemental page.
I have also seen many censuses where young children were enumerated in the households of their grandparents. In the Census of England and Wales, since the instructions are different than in the United States, these children are sometimes marked as "visitor" even when they are related to the head of household. The instructions for the 1940 Census say:
section 441. Column 7. Name of Each Person Enumerated.-Enter in col. 7 the name of each person whose usual place of residence is with the household. Be sure to include persons temporarily absent and all children, even the very youngest. Do not include persons visiting the family, whose usual place of residence is elsewhere, unless they will not be reported in another enumeration district. For a new-born infant who does not have a given name, write "Infant." Write "Ab" after the name of a person temporarily absent at the time of enumeration, such as a traveling salesman, a student, etc., who has sleeping quarters elsewhere, but whom you enumerate as a resident of hour district in accordance with the instructions in paragraphs 305 to 307.
So while the enumerator should have included your sister's information even if she were visiting a relative, it could be that if your sister was absent from the household on that single day, she might not have been recorded.
The National Archives and Records Administration's page About the 1940 Census says:
Unlike more recent censuses, the 1940 census was taken entirely by census enumerators going door to door and collecting information. If a person wasn't home when the census taker came, the census taker would make a return visit. People who were counted on return visits are listed at the end of the regular pages for the enumeration district on pages that begin with number 61.
It's possible that if someone in the office made a copying error and caught it in time, your sister could be on page 61 or the following pages along with the other people who were counted on return visits. But it also could be that your sister was missed and doesn't appear in the census at all.
On some sites like Ancestry.com, if the index is in error, it is possible to submit a correction for the index, but if a person who appears in the image gets skipped in the index, there is not usually a mechanism in place to insert new lines in the index. For a case like this, where the person's name does not appear on the record at all, there is no way to submit a 'correction' because that would be changing the historical record. Unfortunately, records are made by people, and people make mistakes. We can sometimes ask for corrections to certain records for our immediate family, like vital records (birth, marriage, and death), but for other records like the census, which are intended to collect data about a broad population, there may not be a way to submit a correction.
Some sites like Ancestry allow the user to leave comments on a record, so theoretically you could make a note on the entry for the household that your sister's information must have been dropped in the copying. However, many companies have policies about not posting details about living people online.
I think it is important to remember that when we study our families' history, we are looking at records, not people. The 1940 Census was supposed to be a record of the people in your household, but that's not what it actually is. It is the record of what the enumerator for your district heard your mother say about who was in the household -- and while it was supposed to be a complete record, there's no guarantee that it is. Unless you were in the room and witnessed the interview with your mom and the census taker, you don't know what your mom said -- you can only vouch for what she should have said. This may seem picky, but these are the kind of things that need to be considered when we do evidence analysis.
So even if we could write the US Census Bureau and say "Hey, you missed my sister, you bums!", and they recorded a note that would become part of the record, then the generations of genealogists after us would have to consider that your correction was made more than 72 years after the fact, and that you were a child at the time of the event.
Further reading:
