Family Historian User Group

Hello - I’m trying to think of a way to enable data from different censuses ( censa?) to individuals and need some help please.

For instance I am looking to upload about 1500 individuals with census data from 1841 to 1911 and I wondered if there’s a smarter way to do it than searching for each one on FH, once added of course, and then merging. This would be long winded for so many I think

For instance below for family Smith

1841 Mr Smith, Mrs Smith
1851 Mr Smith, Mrs Smith, smith1, smith2, smith3
1861 Mr Smith, Mrs Smith, smith1, smith3, smith4, smith5
1871 Mr Smith, Mrs Smith, smith4, smith5
1881 Mr Smith,
1891 Mr Smith

What is the most efficient way to reference all Mr Smith’s census data for 6 censuses against Mr Smith’s record in FH; and similarly for the other Smiths etc

All advice welcome, thank you

ancestralsources:index|> Ancestral Sources will let you enter each Family as a group for each census with all the details and source citations. I recommend reading the introduction on the linked page above to find out more. You can open Ancestral Sources as standalone program and keep it open as you do each family. You don't need to find each person in FH before loading AS and you don't need to close if after each one.

Mike, I don't think this question is related to the Ancestral Sources program so moved it to the FH General Usage forum.
Is it related to your earlier postings in Work in progress - Import CSV (12536)?

Have you seen how_to:recording_census_records|> Recording from a Census Record?

I'm not sure I understand the problem, but here is how Census records are often handled.

• Source Records
  The 1841 Census record for Mr & Mrs Smith is captured in a Source record.
  Likewise, the subsequent 1851 Census to 1891 Census record for the Smith family household will each be captured in a separate Source record.
  So there will be six Source records holding the transcript and image for the Smith family for each year.
• Census Events
  Both Mr Smith and Mrs Smith will have an Individual Census Event dated 6 June 1841.
  Both will record the Place and their Address and their Age taken from the record.
  Both will also have a Citation that links it to the 1841 Census Source record described above.
  
  For 1851 all five members of the family will have an Individual Census Event dated 30 March 1851.
  The details will be much the same as above and cite the 1851 Census Source record.
  
  So each household member will have a series of Census Events, one per year.

There are several techniques for finding all the Census Events that are linked to the Census Source records.

If you are capturing these records from scratch then the Ancestral Sources program will simplify the process.

Thank you both - I think I’m just getting frustrated at my progress and was looking for speedier data capture/entry

Not the speediest solution, I know, but once you’ve entered census data for a year for one member of the household, you can use the copy and paste icons in the property box, to copy this fact, with its citation, then paste it to all other household members.

I prefer not to use Ancestral Sources, so this is a reasonably speedy way to enter censuses for a family household, and a good work round.

If you have all the information in a spreadsheet you could use a plugin to get it all in if you wanted to do so and don't want to attach all the images etc.

Mike/Mike

Is this an FH question of more one of "family reconstruction"?

FH is great for recording, but I do find it sometimes difficult for analysis and synthesis. A spreadsheet is often flexible enough to allow you to use it for family reconstruction, but it is then a pity that you cannot quickly load your conclusions into FH.

So what do I mean by "family reconstruction"?

In my late mother's genetic line (she has a - publicly known adoptive line) there is a mystery about her maternal grandfather. Sometime in the early 1900's (1907?) he left home "for a better life in Canada" saying that he would "send for his wife and family" once he was settled. He was never heard from again.

The last record I have for him is the 1901 Census in the family home.
No application found for a passport in his name 1901 to 1911
No embarkation records (or disembarkation/immigration records in Canada or USA) found in his name 1901-1911

So I have a suspicion that he "did a runner" and (under an assumed name)
- left for Canada, or more feasibly
- caught a train for say Coventry or Nottingham (where as a cycle fitter he might have found new work)

But before trying to get my mind around how to trace someone living under an assumed name, I need to do an exhaustive search for records that might be him under his own name in England & Wales (and also as an extended exercise for Ireland & Scotland)

His surname is relatively rare and rarely if at all miss-spelt (thank goodness).

So I download every 1901 Census record for "households" containing someone of his surname. I do the same for the 1911 Census, for 1901-1911 Births (add surname), 1901-1911 Deaths (subtract surname), 1901-1911 Marriages (add & subtract female surnames).

I then need (exhaustively) to go through this pot and match up all records (hoping to find a spare male of about the right age with a plausible occupation in the 1911 Census - who could only be mapped onto my mother's missing grandfather in the 1901 Census)! I don't think FH is the primary tool for doing this.

I used a spreadsheet and went though everyone in the 1901 Census and - going forward - either found their death record (m&f), marriage record (f) or their 1911 Census record, tagging each record as I found it. Then for all untagged records in the 1911 Census - going back - and either found a birth record (m&f), marriage record (f) or their 1901 Census record, and tagged them. Then I went through all untagged BMD records and sought to reconcile them (e.g. B & D in 1901-1911 period). Then for the unreconciled looked for migration records.

To distinguish some families I found I had to dig back into the 1891 census!

Once I have done this there may be benefit (as an exercise separate from searching for the "run away") in loading the whole lot into FH and then family by family work back through the various censuses. The key thing at this stage is that I am no longer needing to be as exhaustive in finding every individual.

Given that this exercise is unlikely to find him (say under a nick-name + known surname), I may then be faced with doing a similar exercise for everyone in the 1911 Census within a certain number of streets distance of say the Raleigh works in Nottingham etc. and then try and match everyone back to their 1901 Census entry - hoping to find someone "new" in the 1911 census of about the right age with a plausible occupation etc.

Both these exercises require bulk matching of data from different censuses. I don't think even a plug in would really help! A spreadsheet with copious use of INDEX() and MATCH() functions is probably better.

All - thank you for the advice - all very helpful - I think I’m nearer to Davidf in what I’m trying to do.

My surname is common in one town but throughout the 19th century censuses was rare elsewhere. 350 instances in 1891 census so I’m attempting to see if there was a small number of families and how they are linked as there are parish register references to the name back to 1580.

For this sort of thing, I use Outwit to scrape census data a page at a time into a spreadsheet (very careful not to trigger the checks for high-volume scraping that many of the sites monitor for -- doing it a page at a time is much that same as copying and pasting the browser window contents, but Outwit does the searching for tables in the html and saving them to a spreadsheet. Then it's sorting, indexing, manually matching... I use excel so the Fuzzy Match add-in is useful as well.

Thank you I will look at Outwit.

Re csv import; for an individual is it best to put all census data for multiple years onto one row for import (with all columns appropriately headed) and then import to FH Where all should be referenced for that individual

Or

Separate rows for each census for each individual and use FH to bring together

Conceptually the first option appears better as less keyboard work and Excel can do the matching and sorting. But, can the import plug in cope with rows of different lengths ie different individuals being recorded in different numbers of censuses?

Sorry to drag this out but I want to try to avoid blind alleys.

My instinct is to not put multiple individuals on a single row, but split them across multiple rows as this permits more useful sorting and filtering (e.g by individual, by year, by address, by Head of Household). It also allows you to add BMD records for the individual who "owns" the row.

So your headings (in one row) might be:

Name (standardised), Address (standardised), Head of Household (standardised),
Census Year, actual name, address, HoH, Relation to HoH, Age, PoB, Occupation etc.

You may find that you can do without the standardised names etc, but I usually find myself adding them so that the table can be indexed and sorted with some efficiency. I often do this as I identify households across census years.

Thank you davidf - sorry, I didn’t explain clearly - I wondered whether to put all census info for a single individual on one row -

So
Row 1 Jo Blogs 1851 data, 1861 data, 1871 data, 1881 data, 1891 data
Row 2 Mrs Boggs 1851 data, 1861 data, 1871 data

Etc - so over my dataset there will be rows of different lengths and I wondered whether the plugin could cope with that (assuming that I put space into blank years to get all rows of similar length

I can do this readily in Excel