1921 census

[webwiz]

I was a little alarmed by one paragraph in Stephen Rigden's interesting description of the digitisation:

After the collection was digitised, it was uplifted from its temporary location and conveyed under conditions of similarly strict security to its place of permanent long-term storage. From that point onwards, the physical census is effectively closed again in practice. For the foreseeable future, then, access to the 1921 Census will only be through the digital surrogate that Findmypast has created.

If this is literally true then how will corrections to transcription errors be made?

(I assume that posting in this forum is not a breach of the instruction not to post parts of the newsletter elsewhere. Apologies if I'm wrong)

 

[Tim]

Corrections/updates will be made to the transcribed data, the same as is done for every other census.

 

[webwiz]

I hope so (in which case the article is misleading) but I also notice that there is no mechanism for reporting errors.

 

[peter]

I hope so (in which case the article is misleading)

I'm not sure that anyone else would have interpreted it in the way you have.

but I also notice that there is no mechanism for reporting errors.

There is, but only from the transcript; this was mentioned in the 6th January newsletter.

However many of the apparent transcription errors are not, in fact, errors by the transcriber but the result of misguesses by the data processing algorithms - hence the need to clean up the data that I referred to in the latest issue.

 

[Pauline]

You can report transcription errors by email without buying the transcript - see this FMP article. (3rd & 4th paragraphs)

 

[webwiz]

Thanks Pauline. That is a helpful reply.

 

[canadianbeth]

ut I also notice that there is no mechanism for reporting errors.

I reported an error by using the 'support at findmypast.com' from the payment e-mail I received. I doubt it will help but I tried. My aunt was listed twice; I thought I had paid for the wrong one so paid for the second one as well and it turned out to be exactly the same as the first but with big 'x's throughout. One listed her as born in 1912, the second 1913.

 

[Stephen L]

From What you really need to know about the 1921 Census Census references: the piece number is handwritten on the 'Cover' (it's preceded by the reference RG15) but is also part of the filename when you download the image,

There is another place where we can see the piece number and that is in the email received immediately after purchase.

 

[Stephen L]

You can report transcription errors by email without buying the transcript - see this FMP article. (3rd & 4th paragraphs)

Yesterday I reported 6 errors from 3 census pages getting a same day reply to each email (one per page). "We have now reported the errors within the entries you have provided and these will be updated in due course. " I do not know if they will report back when the transcription has been updated.
One error I reported was without purchasing the image. A 2 x great aunt was obviously not Reginald Elizabeth. I know her first name was Regina.

 

[peter]

A 2 x great aunt was obviously not Reginald Elizabeth.

You can never be sure these days....

 

[peter]

I'll allow someone else to have the pleasure of reporting these transcription errors to Findmypast:

 

[Stephen L]

I'll allow someone else to have the pleasure of reporting these transcription errors to Findmypast:

Done!
The Pratt one was Rhoda Elizabeth. There was a Reginald Pratt in the row above.

 

[Stephen L]

You can never be sure these days....

True for these days but not likely for a birth in 1882.
Without receiving any further email notification I can report today that her name has been corrected along with 4 out of the other 5 suggestions I made. on Wednesday. So not more than 3 days.
The one change that appears to have not been accepted is WHITBREAD was incorrectly transcribed as WHITTERS. At least that could be found with a search for WHIT*.

 

[PhilGee]

The problem I recently found is that there doesn't appear to be an easy way to indicate "the last two people in the household are actually in a different household and there is no transcription of the third member of that household". They did actually carry out the correction of changing the house number from 8 to 9

 

[Britjan]

Interesting analysis of the 1921 census just posted at https://t.co/hfmnVf2EyT . "From a lack of living space to an increase in the number of widows and orphans, newly digitised county and district level statistics from the 1921 Census show changes after World War I and the Spanish Flu pandemic." from O.N.S.

 

[Tim]

Find your 20sPeople in the 1921 Census with 50% off this weekend

Make the most of this offer and crack on with some of your research.

The offer begins Friday 22 April at 10:00 BST, and ends Monday 25 April at 23:59 BST.

The 50% off discount is automatically applied at checkout, with no code required. No subscription is needed to take advantage of the offer. 12-month Pro subscribers who already benefit from 10% off 1921 Census purchases will automatically have their discount increased to 50% for the offer period. The offer applies only to 1921 Census image and transcript purchases.


I just received this email

 

[peter]

I just received this email

That's interesting - the press release gives an embargo time of 1 minute past midnight tonight, so I haven't been able to mention it. Was that email from TNA or FMP?

EDIT: I've just noticed that I have a similar email from TNA in my inbox.

 

[Bob Spiers]

Would you believe (yes you would because it is Sod's law) I decided -after a good month or more after my previous sampling of the 1921 Census - to buy a single transcript for a household and to dot the i's and cross the t's on its members to confirm what I had previously conjectured. Then I too received the email about the 50% reduction.

But that is not the main reason for posting because as my FMP membership is up mid May, they emailed to advise of a subscription rise to £179.99 "an increase of £1.67 per month"...they add, of course: "as a 12-month subscriber, you will continue to receive your 15% loyalty discount if you renew for another 12 months".

I know price hikes are commonplace (save perhaps LC of course) and up to the subscriber to consider taking or rejecting. In my case I intend to renew but could not help thinking that the increase has more than a little to do with recuperating costs on transcribing the 1921 Census. I am sure there must have been quite a fall back in revenue after the first mad rush when everyone bought Census transcripts, with few I would think, paying for an original.

It crossed my mind not to be surprised if a discounted 1921 offer would be in the wings quite soon, and lo and behold it came. But how soon before it is all rounded up in the subscription and a little more than a £1.67 a month increase I am sure?

 

[peter]

subscription rise to £179.99

They might have warned me - but they didn't. I see that the Plus subscription is also going up, by £10 a year to £129.99

There could well be matching increases at other FMP sites - I haven't checked (yet).

 

[Britjan]

True for these days but not likely for a birth in 1882.
Without receiving any further email notification I can report today that her name has been corrected along with 4 out of the other 5 suggestions I made. on Wednesday. So not more than 3 days.
The one change that appears to have not been accepted is WHITBREAD was incorrectly transcribed as WHITTERS. At least that could be found with a search for WHIT*.

I've just been taking a look at my Pratt relatives again so any corrections are appreciated.