Around 700,000 Scottish households – a quarter of the country – are facing £1,000 fines for failing to complete the census. Eleven years ago, the last time the census was run, it took 10 days to reach the current response rate of 74 per cent. This time it’s taken over a month. There’s not much hope in getting the rate up either. Studies from the US show census return rates don’t improve after the first few weeks.
If 700,000 households are to be fined this would be the largest prosecution in Scottish legal history, probably British too. In 2011 – where the final response rate was 94 per cent and over 90 per cent in all council areas – just five people were reported to prosecutors. By the following February there’d only been one conviction. Given this precedent not to prosecute, as well as the cost of living crisis it is perhaps unrealistic to expect the government to go after the non-returners.
But something needs to be done to boost the return rate. The census acts as the basis for a whole range of national statistics that guide policy.
A notable example of these estimates failing is Scotland’s Covid statistics. After ministers boasted about vaccination rates it emerged that, impressively, some 103 per cent of over-70s had been vaccinated. An impossible achievement: the error was caused by an underestimation of the headcount in certain age groups. That’s worrying because it means we don’t know the true vaccine coverage among vulnerable groups. A lower census return could lead to a widening of the confidence interval around future population estimates.
What’s gone so badly wrong? All we have to go on so far is anecdote but there seems to be a mixture of cock-up and anti-government sentiment. There are complaints of returning the census via the online form and then weeks later receiving angry letters reminding you to fill it out – suggesting the IT may have failed. Trust in the SNP government is undoubtedly a factor too. Some who ascribe sinister motives to a government that has increasingly been criticised for secretive and anti-democratic actions are flatly refusing to share their information. Certain women’s groups are unhappy over the inclusion of questions that allow respondents to self-identify their gender. Legal academic Scott Wortley says the design of many questions has ‘alienated’ people. Was this alienation avoidable?
The census being online for the first time has caused difficulty too. The lack of a free text approach in a paper version puts people into boxes they don’t want to be in. Options provided for white ethnic background don’t include English, for example, despite some half a million Scots being of English origin in the previous census.
It’s hard to escape the delay to the census as a major factor too. In 2020 the Scottish government took the decision to delay the census citing Covid restrictions. Behind the scenes there were issues with procuring the ‘contact centre’ that would deal with the first ever online census. Covid was perhaps a convenient cover. Either way, it makes it the first census in a 150-year history not to cover all parts of Britain. A decision that was described at the time as ‘an act of scientific vandalism’. In doing so Scotland missed out on the UK advertising campaign which would have considerably helped take-up.
Scotland faces a host of societal crises. The worst drugs deaths in the developed world, high alcohol deaths and falling life expectancies. The gap in living standards between the rich and poor in parts of Glasgow would not be out of place in the third world. To tackle these issues and to scrutinise government, statistics play a vital role. If the response rate can’t be improved it will be a failure of unprecedented scale in the history of Scottish social science.
Spectator + Scotland = what did snp do wrong?
Don’t even need to read it.
The only reason I have seen given for not filling out the census (and admittedly, this is on Reddit) it the general conspiratorial “I am not giving *them* my data to sell”.
This seems much more like another USAian import in the form of “fear a big government” rather than any credible concern.
For ethnicity there was a free entry area, so if you wanted to put “English” (or “French”, or whatever), you most certainly could.
And I’m pretty sure the more contentious questions were optional.
Even though improvements could be made to the online version [edit: and to how it was online first, paper second], it seems to me that the census has more gotten caught up in conspiratorial nonsense than anything else.
There is another blunder that the delay has caused that I don’t think the article mentions:
anyone who moved from Scotland to England/Wales/NI in between the 2021 and 2022 censuses won’t be counted in either census, which kinda goes against the entire point of a census.
Surely it would be cheaper just to have a U.K. wide census?
The various factors that seem to have led to a pathetic response rate are:
Change of date / being out sync with the rest of the UK. This delay is probably the single strongest factor. In doing so Scotland missed out on the UK advertising campaign which would have considerably helped take-up.
SNP IT cock-up where submitted returns aren’t being registered as completed
Anti government sentiment / anti Scexit sentiment
Certain women’s groups are unhappy over the inclusion of questions that allow respondents to self-identify their gender. Legal academic Scott Wortley says the design of many questions has ‘alienated’ people.
The census being online for the first time has caused difficulty too.
The lack of a free text approach in a paper version puts people into boxes they don’t want to be in. Options provided for white ethnic background don’t include English, for example, despite some half a million Scots being of English origin in the previous census.
The census isn’t due until Monday. This seems like jumping the gun a bit. Folk still have time to do it
The census isn’t due until Monday. This seems like jumping the gun a bit. Folk still have time to do it
It’s been a bit of a farce.
I got the first reminder after completing my part as you need to submit it manually after all family members have submitted their part.
The reminder had a new code to start a new census, not access my old one.
I got the second angrier reminder a couple weeks later despite submitting, again with a new code not to access my existing census but start a new one.
It took a phone call to sort out with a hold time of 15 minutes
Instead of toothless threats for non-completion why don’t they offer incentives. Like all the completed census references go into a random prize draw similar to Premium Bonds. I’m sure ERNIE can pull a double shift every 10yrs. Lol
>The worst drugs deaths in the developed world, high alcohol deaths and falling life expectancies
None of that would be picked up by the census though.
>question 5: have you ever died of a drug overdose? yes/no
11 comments
Around 700,000 Scottish households – a quarter of the country – are facing £1,000 fines for failing to complete the census. Eleven years ago, the last time the census was run, it took 10 days to reach the current response rate of 74 per cent. This time it’s taken over a month. There’s not much hope in getting the rate up either. Studies from the US show census return rates don’t improve after the first few weeks.
If 700,000 households are to be fined this would be the largest prosecution in Scottish legal history, probably British too. In 2011 – where the final response rate was 94 per cent and over 90 per cent in all council areas – just five people were reported to prosecutors. By the following February there’d only been one conviction. Given this precedent not to prosecute, as well as the cost of living crisis it is perhaps unrealistic to expect the government to go after the non-returners.
But something needs to be done to boost the return rate. The census acts as the basis for a whole range of national statistics that guide policy.
A notable example of these estimates failing is Scotland’s Covid statistics. After ministers boasted about vaccination rates it emerged that, impressively, some 103 per cent of over-70s had been vaccinated. An impossible achievement: the error was caused by an underestimation of the headcount in certain age groups. That’s worrying because it means we don’t know the true vaccine coverage among vulnerable groups. A lower census return could lead to a widening of the confidence interval around future population estimates.
What’s gone so badly wrong? All we have to go on so far is anecdote but there seems to be a mixture of cock-up and anti-government sentiment. There are complaints of returning the census via the online form and then weeks later receiving angry letters reminding you to fill it out – suggesting the IT may have failed. Trust in the SNP government is undoubtedly a factor too. Some who ascribe sinister motives to a government that has increasingly been criticised for secretive and anti-democratic actions are flatly refusing to share their information. Certain women’s groups are unhappy over the inclusion of questions that allow respondents to self-identify their gender. Legal academic Scott Wortley says the design of many questions has ‘alienated’ people. Was this alienation avoidable?
The census being online for the first time has caused difficulty too. The lack of a free text approach in a paper version puts people into boxes they don’t want to be in. Options provided for white ethnic background don’t include English, for example, despite some half a million Scots being of English origin in the previous census.
It’s hard to escape the delay to the census as a major factor too. In 2020 the Scottish government took the decision to delay the census citing Covid restrictions. Behind the scenes there were issues with procuring the ‘contact centre’ that would deal with the first ever online census. Covid was perhaps a convenient cover. Either way, it makes it the first census in a 150-year history not to cover all parts of Britain. A decision that was described at the time as ‘an act of scientific vandalism’. In doing so Scotland missed out on the UK advertising campaign which would have considerably helped take-up.
Scotland faces a host of societal crises. The worst drugs deaths in the developed world, high alcohol deaths and falling life expectancies. The gap in living standards between the rich and poor in parts of Glasgow would not be out of place in the third world. To tackle these issues and to scrutinise government, statistics play a vital role. If the response rate can’t be improved it will be a failure of unprecedented scale in the history of Scottish social science.
Spectator + Scotland = what did snp do wrong?
Don’t even need to read it.
The only reason I have seen given for not filling out the census (and admittedly, this is on Reddit) it the general conspiratorial “I am not giving *them* my data to sell”.
This seems much more like another USAian import in the form of “fear a big government” rather than any credible concern.
For ethnicity there was a free entry area, so if you wanted to put “English” (or “French”, or whatever), you most certainly could.
And I’m pretty sure the more contentious questions were optional.
Even though improvements could be made to the online version [edit: and to how it was online first, paper second], it seems to me that the census has more gotten caught up in conspiratorial nonsense than anything else.
There is another blunder that the delay has caused that I don’t think the article mentions:
anyone who moved from Scotland to England/Wales/NI in between the 2021 and 2022 censuses won’t be counted in either census, which kinda goes against the entire point of a census.
Surely it would be cheaper just to have a U.K. wide census?
The various factors that seem to have led to a pathetic response rate are:
Change of date / being out sync with the rest of the UK. This delay is probably the single strongest factor. In doing so Scotland missed out on the UK advertising campaign which would have considerably helped take-up.
SNP IT cock-up where submitted returns aren’t being registered as completed
Anti government sentiment / anti Scexit sentiment
Certain women’s groups are unhappy over the inclusion of questions that allow respondents to self-identify their gender. Legal academic Scott Wortley says the design of many questions has ‘alienated’ people.
The census being online for the first time has caused difficulty too.
The lack of a free text approach in a paper version puts people into boxes they don’t want to be in. Options provided for white ethnic background don’t include English, for example, despite some half a million Scots being of English origin in the previous census.
The census isn’t due until Monday. This seems like jumping the gun a bit. Folk still have time to do it
The census isn’t due until Monday. This seems like jumping the gun a bit. Folk still have time to do it
It’s been a bit of a farce.
I got the first reminder after completing my part as you need to submit it manually after all family members have submitted their part.
The reminder had a new code to start a new census, not access my old one.
I got the second angrier reminder a couple weeks later despite submitting, again with a new code not to access my existing census but start a new one.
It took a phone call to sort out with a hold time of 15 minutes
Instead of toothless threats for non-completion why don’t they offer incentives. Like all the completed census references go into a random prize draw similar to Premium Bonds. I’m sure ERNIE can pull a double shift every 10yrs. Lol
>The worst drugs deaths in the developed world, high alcohol deaths and falling life expectancies
None of that would be picked up by the census though.
>question 5: have you ever died of a drug overdose? yes/no