Censuses are the bedrock of genealogy. More than any other record they track our ancestors through time as they grow up, get married, have children, suffer losses, move addresses, and change professions. But for as useful as they are, they are frustratingly unreliable. Why don’t people admit they’ve aged ten years from one census to the next? Why do their names sometime differ dramatically from what they’re supposed to be? Why aren’t countries of origin and dates of immigration consistent over time? Despite being government records, we can’t take censuses at face-value without corroborating evidence, and so we spend an awful lot of time debating how the information they’ve recorded got so bolluxed up.
A trove of articles I’ve uncovered from The News-Messenger, the local paper of Homestead, Pennsylvania, sheds some light on these discrepancies from the perspective of the census takers in 1900. It’s clear that everyone understood the odds were against the enumerators to get accurate information, but the paper went to great lengths to ensure that “Homestead will not be found in the [class represented] as obstructing the endeavors of the enumerators to get the exact truth in the information they seek” (5/2/1900). It started by printing the census instructions several times to prepare readers with the list of questions they’d be asked and how to be sure of answering them accurately. In case the instructions weren’t enough, an often hilarious article from May 17, 1900 resorted to dire warnings:
YOU MUST ANSWER HIM
Further Precautionary Advice Given Those Whom the Census Enumerators Will Visit.
Don’t lie. When the census enumerator comes around June 1 tell him the truth. If you don’t you will go to the bad place and if he finds out you may go to a worse place. He may sue you and take you before an alderman…
Some of the questions the enumerators are expected to ask may seem a little obnoxious, but that is not the fault of the enumerator. He is there to ask all the questions he as printed, and he is expected to get true and correct replies. If any person refuses to answer them, he is liable to arrest, fine and imprisonment. It will do no good to kick…If you do not like [the questions] it cannot be helped. Your Uncle Samuel had a great big family, and it is impossible for him to get around among you all personally, so he sends his enumerator…So if you do not like the enumerator and the questions he asks you, just remember he doesn’t like you or the questions, either…
The paper took special care to address the problem the vexes us genealogists most: age! “The most difficult task of a census enumerator,” the editor wrote on May 2, “is to obtain information about the ages of people. Young people usually want to be considered old, middle aged people report themselves younger than they are and very old people will add a few years to their actual age.” An article from April 23 takes a more pointed tone towards one particular subset of the population:
The United States census department has issued a request to the women of uncertain ages to tell the truth as to the length of time they have been inhabitants of this terrestrial globe. The manifest inaccuracies in the the census statistics is the cause of this appeal to the honesty of the aforesaid ladies…The habit among the dear women of adding to and subtracting from their ages is an old one — much [older] than the United States census bureau –and, as has been said before, it is hard to learn old birds new tricks.
We genealogists joke a lot about this sort of vanity in our ancestors; it tickles me to see it addressed head-on a month before the enumerators would collect the very answers that flummox us now.
Like many cities and factory towns in the U.S. at that time, Homestead had another challenge: its non-English-speaking residents. As the center of the American steel industry, Homestead was then experiencing a massive influx of Eastern-European immigrants, who came largely to work in the town’s world-famous mill. The paper noted early on that interpreters would be needed to assist the enumerators sent to question this population, and the jocular article I quoted from at the start turned much more serious when addressing the preparations for non-English speakers:
It is always expected that trouble will be found in the foreign quarters. To relieve this as much as possible this year the census bureau is sending out circular letters to the ministers of all churches attended by those of foreign tongue, explaining the object of the census. These circulars will be read from the pulpit and the priests will give the people instructions to answer all the questions cheerfully and promptly. It is thought this plan will enable the enumerators to get more satisfactory results from the foreigners than ever before.
Towards the end of July, almost two months after the enumerators passed through, preliminary numbers for the town were published, and at 12.3K they were 3K lower than the police census the town had itself conducted earlier that year. Three days later the issue of undercounting the population of the second ward, the immigrant district, came to the fore:
A great deal of complaint is being made about the manner in which the census of the second ward was made. It is claimed the census is incomplete and will not show the true population of the ward by many hundred (sic), and that the showing of the town will suffer in consequence. It is said that in many cases family after family was missed where more than one family occupied a single house, the numerator not being familiar with the neighborhood. But this is not all. The claim is also made that blocks of houses were missed and one case has been brought to our notice where a boarding house containing 28 people was not visited at all, and the inmates are all English speaking people at that. If there are many cases like this, no wonder the figures of the official census do not fit up the total shown by the police census, which was most complete in this ward. (7/23/1900)
A month later a most unexpected decision was made:
The Second Ward to be Re-Canvassed.
James Esler, supervisor of the census of Allegheny county, was in Homestead yesterday and ordered a reise to be made of the census of the Second ward. Mr. Esler is not satisfied with the returns from this ward…he has been making an investigation for several weeks past and has found sufficient irregularities in the returns made by the census enumerator in this district to order the word done over again…Name after name appeared on the police census that were omitted from those taken by the census enumerator…
He has ordered a revise to be taken however in order to set at rest all doubts about the matter, the same man being employed as made the first canvass. Only that district below Fifth avenue will be re-canvassed at present but if anything radically wrong is discovered here, the entire ward will be gone over again. (8/23/1900)
The paper does not report on the results of this do-over or its ultimate extent, but I have personal knowledge that people were still missed, as my own immigrant ancestors who lived in the ward do not appear in the 1900 census. (And yes, before you ask, I have numerous other sources that document them there…)
At the end of the day, the census data we’ve received is what it is, and this fascinating context for some of the inaccuracies doesn’t change it. To paraphrase the Homestead paper, if you still do not like your town’s enumerator and the answers he recorded, just remember that apparently he didn’t like your ancestors, either!