England Census of 1841
Bill
family lines, particularly using things like their Census data sets---
not so much for "person specific" data, but for more global information.
While exploring the 1841 Census for England for the Maxwell family
line I noticed a number of peculiar things.
Initially I was trying to see where in England the Maxwells were
focused. An initial run showed a total of 1328 Maxwells in England.
I then ran individual runs for each county. This gave me a sense of
the distribution of the surname---there were two focus areas, one
near the border of Scotland, and one surrounding London. A
preliminary mapping of the results is given at
http://www.werelate.org/wiki/Maxwell_Distribution_in_England%2C_1841
What I found as peculiar was a discrepancy between the total number
of Maxwells in England (generated by a simple search for "Maxwell",
"England"
and the total built up from separate searches for the individual
counties. Had the later total been the lesser, I would probably have
passed this off as "Okay, they must be recording some individuals in
some other units besides the "county" (like maybe the "East Riding of
York", or perhaps under a city name, but not the county.
Unfortunately, the population built up from the separate counties was
actually larger (1345) than that given by the total for England as a
whole(1328). This is a fairly tedious extraction, and its easy to
make errors, but redoing the extraction yields the same result---more
people in the build up total, than in the direct total for England as
a whole.
Even so, its possible I'm making some sort of systematic error.
However, in exploring this a bit more, I noticed other discrepancies.
For example:
Ancestry allows you to search by sex. Their totals are
Male 387
Female 438
(Sum 825)
Sex not specified* 820
*results of a direct search for Male, female and Total
The fact that the separate extractions for male and female add up to
a different value (820) than that obained when sex is not specified
(820) suggests that something is wrong in their data extraction routine.
I'm wondering if I'm missing something obvious? Has anyone else
encountered a similar problem with Ancestry data sets like this?
Bill