Family Historian User Group

I was entering a lot of census records from 1900-1940 and noticed that some of the table columns in AS did not match the online picture of the particular census record. Did they use different census record designs in different parts of the country? Is there a process to modifying the table when someone notices it is off? I would like to be of help when I notice something different... Just let me know what I can do to help.

Chris,

What year had the mismatch? What was the column that was missing or extra in AS?

Bill

I believe it was numerous extra columns.

I think without knowing the year of the census and the fields that didn't match it is going to be hard to help you out.

Also, where were you seeing the census online? Ancestry? FamilySearch? etc.

Bill

Chrisv wrote: ↑

04 May 2022 23:39

I was entering a lot of census records from 1900-1940 and noticed that some of the table columns in AS did not match the online picture of the particular census record. Did they use different census record designs in different parts of the country? Is there a process to modifying the table when someone notices it is off? I would like to be of help when I notice something different... Just let me know what I can do to help.

Sometimes the particular formatting decisions for the original census forms doesn't really suit the data capture requirements of AS and so there are definitely differences. For example actual UK census forms will have two age columns, one for males and one for females but as the age is recorded in the same Fact field for both males and females the AS census templates use a Gender column and an age column instead. There are also decisions that I've made, such as I don't include the Road/Street column as this information gets recorded in the address field instead, above the census grid. For users who want to do things differently they can create their own modified census templates. BillH does this and has shared some of his templates on the FH Knowledge Base. You can see various AS templates here: https://fhug.org.uk/kb/download-type/ancestral-sources/

My opinion is we're just trying to capture the data in an easy to read format in a transcript and it can never be perfect. It's not written in the person's handwriting! If you want to see exactly how it looks then the image is linked to the source.

I seem to recall that a subset of some US Census returns do have a lot of extra columns.
It was discussed some time ago and I think the decision was to omit those columns from the AS Grid.

Mike, you may be thinking of this topic US Census 1940 (18603) where the problem was that some of the extra columns at the bottom of the page in the US 1940 census were for supplementary questions only required for a sample of people. There was some discussion about whether they should be omitted from AS to reduce the width of the grid and the Text from Source.

LornaCraig wrote: ↑

05 May 2022 10:15

Mike, you may be thinking of this topic US Census 1940 (18603) where the problem was that some of the extra columns at the bottom of the page in the US 1940 census were for supplementary questions only required for a sample of people. There was some discussion about whether they should be omitted from AS to reduce the width of the grid and the Text from Source.

BTW, the US Census 1950, has the same layout. There are supplementary questions for only a sample of people at the bottom of the page.

It is the 1940 USA Census. There are about 20 additional columns in the AS template, compared to the familysearch.com census records I have seen for 1940. Here is a sample of one of them. Not a big deal - just thought it was odd there were that many.

Do any AS users reading this thread think it would be useful to have an option in AS to only include columns with data in them in the source text produced by the auto-text? It would reduce the size of the tables down if any blank columns were automatically not included. If so then I'll look to do that for the next version.

I'm probably not allowed to have an opinion but I think the absence of information is important to record...

I'm not sure if what I am saying is understood. It's not the unfilled columns, rather the extra columns not included in the actual census. I had to break it into 3 snips to be legible.

Chrisv wrote: ↑

05 May 2022 15:31

I'm not sure if what I am saying is understood. It's not the unfilled columns, rather the extra columns not included in the actual census....

Chris, in my earlier post in this topic I referred to another topic, US Census 1940 (18603). Did you read that topic? The point is that the extra columns are included in the actual census, but you might not have noticed them because they are at the bottom of the page and are for supplementary questions which were asked of a sample of people.

Chris,

This is the full page for your Elizabeth Plumb example. Notice at the bottom there are two lines with extra questions for two people on that page (shown in the red box). These are included in the template for AS so you can input the information if your person appears in these two lines. If they do not, you can just ignore those columns in the template.


Bill

ColeValleyGirl wrote: ↑

05 May 2022 14:52

I'm probably not allowed to have an opinion but I think the absence of information is important to record...

Of course you are! But I was thinking that perhaps it could be an option in a template to specify that certain columns would be not included if empty. e.g.those extra columns on the us census we've been discussing. Other columns could be marked as 'include even if empty'

Nick,

Like Helen I like to include everything, empty or not. I think your idea would be fine as long as there is an option to "include all columns even if empty". I wouldn't want to have to go through and mark them one by one.

Bill

NickWalker wrote: ↑

05 May 2022 17:31

...was thinking that perhaps it could be an option in a template to specify that certain columns would be not included if empty. e.g.those extra columns on the us census we've been discussing. Other columns could be marked as 'include even if empty'

Yes please! I export my data, including the source text, to TNG. Since Rich Text is not supported it gets converted to simple HTML tables. Generally the results are acceptable, but things can get somewhat messy in extra wide tables. The ability to eliminate empty columns could be very useful in this case. Perhaps optionally adding a short note below the table to explain missing columns could be helpful.

H.

Well, I'll be chagrinned!
I did not see your post, Lorna. Thanks. It was interesting reading, and I learned a lot about the history of the US Census! That is the first I have ever seen that version - or the first I have noticed it!?!

NickWalker wrote: ↑

05 May 2022 17:31

ColeValleyGirl wrote: ↑

05 May 2022 14:52

I'm probably not allowed to have an opinion but I think the absence of information is important to record...

Of course you are! But I was thinking that perhaps it could be an option in a template to specify that certain columns would be not included if empty. e.g.those extra columns on the us census we've been discussing. Other columns could be marked as 'include even if empty'

Nick,

From my perspective that would be helpful, i.e. having the ability to select certain columns that are always present and others that are removed if empty.
I understand what others are saying about recording the absence of data but to try to reduce the table size I'm removing columns in certain cases.
E.g:
1939 Register - Status column if not populated
1911 - all 3 children columns if single and not populated or 'children died' column if 'children born' = 'children alive'.
Yet I always leave in the infirmity column even if blank.

Might just be my weird logic, but that way I think I'm getting the most data for the most efficient table.