In 2014, voters aged 16-41 were more likely than not to vote Yes whereas voters above 41 were more likely to vote No. More than ten years later, half a million predominantly No voters have passed away leaving the electoral roll and a similar amount of predominantly Yes have joined it.
Despite this many pollsters for Scottish elections tend to adjust, ‘weight’, their sample to match the 55/45 No/Yes ratio in 2014. Critics suggest this skews the results and point to the Yes leads and better SNP figures in those polls which do not do so.
Norstat for the Sunday Times in their poll today giving, for the Constituency vote:
- Con 14%
- Lab 21%
- LD 10%
- SNP 37%
- Reform 12%
- Greens 5%
weighted the sample for this, from their total sample of 1 013, to replicate the 2014 referendum result more than 10 years before, to give 292 Yes voters and 349 No voters. This surely predisposed the result to favour the Unionist parties.
In their poll today giving, for the Regional vote:
- Con 13%
- Lab 21%
- LD 8%
- SNP 27%
- Reform 10%
- Greens 8%
- Alba 3%
also weighted the sample for this, out of their total sample of 1 013, to replicate the 2014 referendum result more than 10 years before, to give 292 Yes voters and 349 No voters. This surely predisposed the result to favour the Unionist parties.
I have limited expertise in this so cannot say whether and to what extent, had they used equal numbers of 2014 Yes an No voters, the above results might have changed but it’s hard to imagine them giving the SNP other than a greater lead over Labour.
Here’s an explanation of how the process of weighting works:
Another issue is how to ensure the sample is representative of the general population. To achieve this, polling companies “weight” their data to match the demographic profile of Scotland.
At its most basic level, this means that if a poll of 1,000 people is made up of 550 men and 450 women, it is unrepresentative because it does not reflect the profile of Scotland (48% male, 52% female)
So the answers of female respondents will be given slightly more weight (in this case they will each count as 1.133 people) to give them a representative impact on the final findings.
Conversely, the men will be weighted to each count as 0.891 people.
The same procedure is routinely carried out for age group, social class and region.
https://www.bbc.co.uk/news/uk-scotland-scotland-politics-27585550
I welcome comments.
Talking-up Scotland 
Been saying that for a while, now, it’s impossible for the polls to be so close, with more younger voters coming of age & those of my generation & older are dying off.
The vast, vast majority of people that have a vote now that didn’t back then that I know of support Independence, & I have only met one person that voted YES that would now vote No.
We’re being lied to, by the polluters, the media & every single tiny bit of the british state.
LikeLiked by 6 people
There is only one poll that counts.
The POLLing station on the day of an election.
Too much is read into all these surveys.
LikeLiked by 2 people
Weighting by a DECADE OLD vote is absurd and frankly inexcusable.
LikeLiked by 5 people
more valid poll would be last Holyrood election. It was 16+.
dottieb
LikeLiked by 4 people
Ahem. You believe the Brit Nat mejah, politicians and commentariat would be fair, fair-minded and honest? Then, good sir, I have a bridge to sell to you!
Shiny, well built and goes from here to there. It is only available once and has to be seen to be believed.
Now it may be you are too stupid, too poor and to Scottish to know when you are well orf, but if you send me all your money and let me decide, then you won’t regret it……….signed……David, Gordon, Tony, the Dumb guy, BBC, Hootsmon, Times, Torygraf, presidents of Spain and All the Russias, Obama , Herod and all of us Saff ov ra bordah. We always want you there, right where we can see you. Luv. Haha!
LikeLiked by 3 people
Oh. And Charles, Camilla and Wills. We really, really, like our wee holiday castle——very small, but well formed. Do keep sending your dosh to keep us Saxon-Coburg- Whozits in luxuries, won’t you? Good little Jockos. And Eddie likes his Embra Dukedom—the only foreign one we have…. Oh other than being Prince of the Welshers. They don’t like us Anglos bossing over them.. little Taffies! Har de har.
LikeLiked by 2 people
Remember that most polls are intended to shape public opinion, not measure it.
LikeLiked by 2 people
Certainly true now, it didn’t used to be so obvious….
LikeLiked by 1 person
I must be thick – if polling companies must ‘weight’, why do so using data from an Independence referendum and apply it to that from a Holyrood election?
They’re votes on two different issues – not to mention the former being a binary choice and the latter a multiple choice. How could they accurately decide which political parties benefit from the No factor? They can’t, so presumably the object of the exercise is simply to make the SNP look less popular than they actually are.
LikeLike