UK's 2021 census could be the last, statistics chief reveals
bbc.co.uk
bbc.co.uk
Data from U.S. census from centuries ago is still being used to teach econometrics. But hey, if they were stupid enough to leave the EU then they're capable of doing anything.
This is quite offensive. Please don't tar the entire country with the same brush.
This was - and still is - a highly controversial & divisive subject. It was only voted in by 51.89% vs 48.11% of the population ...at the time - more recent polling suggests it has flipped (1)
1 - https://www.independent.co.uk/news/uk/politics/brexit-second...
It was voted for by 51.89% of the people who voted, which was 72.2% of the electorate.
It was only voted for by 37.4% of the electorate (46.5m) and only 26.2% of the overall population (66.4m).
If you cast the "I don't give a shit" vote, I'm not going to take you seriously if you cry about the outcome afterwards.
[1]https://en.wikipedia.org/wiki/List_of_United_States_presiden...
But to cheer you up: it's still better than in the US, where it's entirely possible to screw the country with slightly less than a majority (or even plurality).
Ok, that attitude is fine. If I thought the EU was doing that, I would‘ve moved to Canada.
Trouble if, the only thing the EU has the power to do is to remove U.K. access to EU membership benefits. That’s not “vindictive”, it’s what the UK voted for after being repeatedly told this would happen.
I'm afraid that also applies to the UK's antiquated electoral system. Conservatives got 43.6% of the vote share, and they are screwing the country (if you think that Brexit, and particularly the hard Brexit they are heading for, counts).
The fact that there is -again- a conservative majority is quite telling with regards to /either/ how much of the country can be swayed by conservative media or by how much of the population seriously wants to leave the EU. (possibly a combination).
[0]: https://d25d2506sfb94s.cloudfront.net/cumulus_uploads/docume...
Disagree: a belief can be true and stupid at the same time, for example any national lottery winner who bought the ticket because they thought it would be a good investment.
(I don’t expect success, but that’s both a separate argument and an entirely pointless argument given the events are now water under the bridge).
True. As a matter of fact it's good to use the census as a sort of periodic anchoring point for your surveys (which can happen more often).
To completely give up on the closest thing to ground truth that can be had is foolish.
It's quite possible that the answer will be "no there isn't". Worth taking a look at though.
Should we judge the entire United States by the decision to elect Donald Trump then?
Why do you believe they are stupid to leave the EU?
https://www.prospectmagazine.co.uk/magazine/thelastcensus "the enduring problems with the 2001 count may mean that after 200 years it could be the last one in its present form. It may have demonstrated that measuring population in our more mobile society is just too difficult."
Could it be the the UK govt has yet again had an IT project spiral into runaway costs?
I’m somewhat amused sibling comments have been downvoted for pointing out the demographic shift these censuses illustrate, reason enough imo to think the decision to end them could be more political than from a concern for what they might cost.
Peter Thiel after all predicts the future for European countries is to end up as either surveillance or Islamic states, something I’m guessing the natives would neither take kindly to nor be too comfortable knowing the likely consequence of this for their offspring. This is the elephant in the room & the deciding factor that led to the Brexit vote, & you can pick any of Angela Merkels recklessness, Daves dodgy timing or the Saudi / Turkish / UAE funding of mosques that encourage next to no mixing or liberalisation of this demographic to thank for that.
An actual census seems like a quaint atavism. The UK can just ask the NSA to count its citizens and provide whatever demographic data is required, and the US can ask the same of the UK. Problem solved.
On the other hand, stopping a census for dumb reasons is one way to do it without publicly admitting they already have the data.
Since we don't have any choice about whether or not to participate in the panopticon, and it's clear no one is going to do anything to stop it, then it should at least serve some public good. Offer a public API and let Western countries build services directly on top of it.
I'd really like any of the downvoters to explain their objections. How can any one working in tech or policy deny the existence of ubiquitous demographic databases? How is this even controversial?
Source: Worked very hard for a decade to protect both patient and voter privacy.
On the other hand relying on indirect data or even surveys can have serious limitations, especially if you base administrative/budgetary decisions on the data collected this way.
https://www.ons.gov.uk/peoplepopulationandcommunity/househol...
We now have multiple demographic databases accounting for all peoples, both living and dead, updated in near real time.
Facebook & Google, telcos, credit rating agencies, data aggregators, NSA, law enforcement, credit card companies, voter registrations, medical records, etc.
Instead of sending people around with clipboards and surveys, why don't we just run some queries on the databases we already have?
But the UK doesn't necessarily have what those countries do. Sweden doesn't need a census because it forces everyone to register their address and family relationships with the central government within 7 days of those changing. The UK by contrast has many disparate single-purpose databases which do not form a coherent whole.
Please fill in this online census questionnaire https://[unique_link_here]
Some ads/incentives here and there to remind people to fill it and one should have 90%+ completion rate.
I'm fairly sure we're having one next year over here in Eastern Europe and no way is our government spending that much on a census, we got some jet fighters for about that much...
Romania does it for 1/3 of that, 84 million Euro for 20 million people = 4.2 Euro/capita.
There's a breakdown of how it's done and the spending, too:
https://www.libertatea.ro/stiri/costuri-uriase-recensamant-b... (translate with whatever, Google and Bing are useless for me)
OK, so wages are a factor, but the UK's cost seems like it should be lower.
I wonder how exactly it costs this much? What is the money spent on?
Lets imagine the 'MVP' census. A printed 10 page document (£0.45), mailed to each household (£0.61), in an envelope (£0.02), times 27.6 million houses. Return mailing costs another £0.61+0.02. Scan every sheet. OCR it (free). Hand-enter 1% of responses where OCR errors are excessive. 276,000 documents hand entered at 5 minutes per document, £10/hr. £0.0083/household. Total costs: £47M
Cost of running a few trial runs of one 10,000 house town or city to perfect the process: £100k
Pay for a year 1 statistician to make graphs, one social scientist to design questions, one manager for the team of printing/scanning/hand-entering people, one IT person to keep/archive the data and get it on the web, and a head of department. Also a little office. Total fixed costs £800k.
The whole thing can seemingly be done for 5% of the quoted cost. And that excludes any potential efficiencies gained by having an app to collect responses, or bulk discounts from postage companies. Also, while a redacted version of the full dataset is published, I would expect the government department to also offer commercial statistics services on the non-redacted dataset where such analysis doesn't impact customer privacy. That should earn them tens of millions at least.
Any real-world project, especially at that scale has much more external costs. There is also enforcement, publicising, security handling, approvals, way more than 1 person to come up with the questions (1 random employee designing the census would be such a waste if you get an average social scientist, that it's ridiculous to propose), compliance, double-checking, etc. etc.
Going from 47m to 906m - when you account for the costs that exist in the real world for something as big and comprehensive like this - is actually fairly reasonable.
Nothing at this scale is ever free, and handwriting OCR is terrible.
The current census process includes a quantity of in-person question-asking; I don't know how much of the cost that makes up but it can only be considerable.
Data protection and information security issues are probably considerable too.
> Cost of running a few trial runs of one 10,000 house town or city to perfect the process: £100k
The census trial ran in four local authority areas each with populations in the hundreds of thousands.
There is no way this is anywhere near approaching free.