TMG-L Archives

Archiver > TMG > 2007-04 > 1176598828

From: "GRSwede" <>
Subject: Re: [TMG] Poll - Census: An event or a source or both?
Date: Sat, 14 Apr 2007 19:00:28 -0600
In-Reply-To: <000301c77ef4$d651bef0$6400a8c0@THINKPAD>

My aunt thought that she was two years younger than she was and the census
enumerations perpetuated the lie. She was illegitimate because my
grandparents did not get married until two years after she was born. My
grandparents didn't want the world to know so they lied. So much for the
census being a source of birth data.


-----Original Message-----
From: [mailto:]On Behalf Of
David Ball
Sent: Saturday, April 14, 2007 6:27 PM
To: 'Darrell Martin';
Subject: Re: [TMG] Poll - Census: An event or a source or both?

> -----Original Message-----
> From: [mailto:] On Behalf
> Of Darrell Martin

> 1. Although the census RECORD provides good evidence that a census
> ENUMERATION occurred, it does *not* prove that the enumerator visited the
> household in question (in many cases he or she did not).

I certainly have examples of wild guesses by someone that clearly was not
privy to the real details. That said, when one looks at three or four or
five censuses for the same family, one can have more faith in the consistent

> 2. There is a tendency to treat the RECORD as an event; that is, as if
> what it includes had an objective existence of its own. Unfortunately,
> anyone who has cranked the old microfilm reader (or clicked a mouse on an
> online image) has discovered that the correlation between census data as
> recorded, and reality, is something significantly less than 100% -- in my
> experience, they are much less reliable than birth records.

Agreed, but when nothing better is available, I'll go with the census data
that I can find. Again, two censuses agreeing are better than one and three
better than two, etc.

> 3. The mechanics of census enumeration frequently lead to errors in key
> information. For genealogists, this includes age, spelling of name,
> whether the name is a nickname, and whether a person (especially a child)
> really resides at that place at that time. Yet a number of TMG users enter
> every variation of that sort of thing as if it were a credible reflection
> of some truth. Instead, **in my opinion** the census RECORD should be
> evaluated and categorized as any other document. Where it is in error, the
> mistake should be duly noted. But Tags should not be created to record
> silly errors just because "it's what the census says".

I do not create name variation tags based on census, mostly because I record
each person separately (although the HOH gets the full list) and put the way
their name was recorded in each census in quotes in the memo of the tag
(along with their age, place of birth and occupation; that way the
census-sourced data are kept together). Lots of people switched from first
to middle names, from children versions (Annie) to formal spelling (Anna),
from given name to nickname, etc. The census naming is more likely, in my
view, to record common usage, rather than formal "official document signing"
names, and I find that interesting.

I treat the census as reflecting the "80/20 rule", 80% of the time more or
less the data is pretty close to correct; 20% of the time more or less it is
off, sometimes way off (particularly with some women's ages <g>). That is
why I really try to get all of the available censuses for a person, to get a
feel for the accuracy of the data.

I also try to help out the other guy by submitting comments to
on both indexing errors and on census enumerator errors, so that others
might not have so much trouble finding their relatives. If you haven't done
that, just click on the "see record" to the left of the name that is in the
search results and click on the invitation to add comments.

Dave Ball

For searching the list archives and other list information, please go to
To contact the TMG List Administrator, send a message to
To unsubscribe from the list, please send an email to
with the word 'unsubscribe' without the quotes in
the subject and the body of the message

__________ NOD32 2189 (20070415) Information __________

This message was checked by NOD32 antivirus system.

This thread: