I was rung by an opinion poller yesterday. I was curious. So I took part. And I asked questions.
The person calling me would not say who had paid for the poll.
The poll was very definitely of marginal seats. My seat is marginal.
I asked how I had been selected. Purely on the basis of my telephone number, I was told.
I commented that the call had been one of very few now received on my landline, that my children never use, or answer.
I asked if all calls were to landlines. I was told they were. That is how they could determine they were reaching the right geographic area.
I asked if they realised that this also meant that their polling was very heavily biased towards those who are older, and so inherently more conservative. They are the people who have landlines now.
I got no response. But if this is normal, and I do not know that, no wonder polling is so unreliable. This poll was pre-selected to deliver a Tory bias.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Landline polling looks highly dubious to me and is becoming more so, for the reasons you cite, Richard. If there was a database of mobile numbers by residential postcode that polling companies could use, that would probably be better. Then there are the firms that do online polling (e.g. YouGov) but they have a completely different set of biases to consider…
Howard, could you mention a couple of them for me? Thanks for this anecdote, Richard. I have been thinking something similar for a while now. In the US, mobile numbers look like landline numbers.
Sure Larry – the main biases with online polling are:
1) it’s a selected sample – in the sense that people actively sign up for YouGov and then they send you links to YouGov polls that they think you might be interested in completing. They have data on your age, gender, place of residence, and various other bits of information about you so presumably they try to send requests for polling completion across a varied sample of people – but they’re all people who signed up to be in the YouGov panel in the first place. And it may be that it’s particular types of people who sign up, and they are not representative of the general population. For example they may well be more politically engaged than people on average.
2) the subsample of people who use the internet in the UK is not random. Young people are more likely to use it, for example (on average – there are of course some very high-usage ‘silver surfers’ in the pensioner age group).
So there is a bias because the sample of people who are online is not random, and then within that online group, the sample of people choosing to fill in YouGov polls is not random.
Thanks very much, Howard. I learned something about this polling organization I didn’t know.
DunGroanin https://www.taxresearch.org.uk/Blog/2019/11/24/why-the-observers-opinion-polling-is-wrong/
i was reading his comments on why the observer polling was way out by 19%. I found it interesting but a confusing. Does he predict a landslide for the tories or labour. I got the impression it was for labour, but he did not clear that up. He did give many pointers to a labour victory but… erm the polls.
There has been a lot of talk about the accuracy of the polls especially by you gov and co. They have just released another poll and its supposed to have said
“YouGov’s ‘MRP’ poll was the first to accurately predict a hung Parliament in the 2017 election.”
“A seat-by-seat analysis by the company showed the Tories would win 359 seats, Labour 211, the SNP 43 and the Lib Dems 17 if the election were held today.”
https://www.mirror.co.uk/news/politics/breaking-general-election-major-poll-20976740
What do u think of this? I read another polls where which did a seat by seat anayalistic on the same lines but i forgot where i saw it. It has labour on 34% and the tories on 41% or something like that.
Labour members who have seen the poll by yougov are all in despair they think its over and completely demoralised. I told them to ignore it as it is rubbish, ie its online, small panel of members repeatedly being used, payment for filing in polls etc.
Take a look at this thread on twitter about yougov methodology via the latest poll et al https://twitter.com/prospect_clark/status/1199826967375351808
So what do u think? Does DunGroanin think labour will have a landslide now or the tories getting a majority as the yougov suggests
Best Wishes
If the LD vote has collapsed then there is but one choice to oppose the Tories south if the Border, and that is Labour
North it is SNP
The clarity adds value
Complete stitch up by the neoliberal elite..the whole thing is
There doesn’t appear to be a notion of call forwarding either.
Well, I think that you are the only person I have ‘associated’ with anywhere (online or in person) who has been subject to one of these pollsters.
I was beginning to see such folk as mythical but now it seems to be that rather than being non-existent, such folk are rather rare which is still saying something about modern polling.
When I was a young child (so long ago there was no TV in the house) political opinion was shaped around the kitchen table. My grandparents and parents would discuss topics picked from the Daily Herald and contrast those views with their own and with reports on the BBC Home Service. At college in the late 60’s we had a lesson called ‘Liberal Studies’. This introduced students to the mechanics of our legislature and the agencies of state. As a result I have always sought to understand and form an opinion on policies not personalities or popularity. I fear the British electorate have become too dependent on prefabricated opinion often masquerading as fact. The modern electorate fear being left behind by ‘the tribe’ and they add their support to the tribe that looks most likely to ‘win’. Opinion polls reinforce these attitudes and wilful ignorance is the order of the day. If YouGov say party X is winning then they must have the best policies. It never seems to occur to people to question that the methodology of a polling organisation set up by two tories might shade results in a particular way. That YouGov hold no face to face surveys appears to be of no consequence to our naive electorate. And if such an independently minded commentator as John Humphrys offers support – well it must be kosher.
Bill
I met a 24-year-old recently who told me they really knew nothing about politics
She seemed to not much care
But as a mother of 4-year-old I suggested to her that what politicians did impacted her greatly
She still seemed indifferent. Politics was in another world and she had no relationship to it
I was tempted to make a merry quip about Vicky Pollard but that is so worrying it would be crass.
On Saturday morning, I caught a brief glimpse of BBC Breakfast news and a single mother was interviewed at food bank in Bristol, I think it was (might have been Avonmouth or Portishead). She indicated she would probably vote for Johnson because she thought he’d “fight for a better Brexit deal” or some such nonsense.
Yeesh, as I thought at the time.
most curious!
Richard, how guarded is your landline number from telephone marketeers, the bane of society over the last decade or so?
are you in the published telephone directory?
are you registered with the Telephone Preference Service ( TPS )?
if a BT customer are you using their BT Call Protect service?
are your details included in the publically available version of the Electoral Roll?
I’m pretty obsessive about limiting my exposure to nusance callers these days and have never been approached by a pollster by landline telephone.
in fact, my telephone ringing is a rare and unusual event, my landline connection is primarily to facilitate a broadband internet connection,
the recent polls seem to be prompting a conversation about the nature of polling, not the results the polls claim to represent.
I make the information widely available
People need to contact me!
I appreciate that and rather suspected it to be the case,
I was prompted by PSR’s comment that you were the only person he had anecdotally heard of being contacted by a pollster,
I find your experience equally unique!
Thanks for this. I have never been asked my opinion in a poll. Now I know a possible explanation!
Our landline has been taken over by windows/doors/kitchen etc sales to the extent that we have the answerphone on permanently. Friends know this and have our mob no.s anyway. Any no. That rings, listens to our message, just clicks off so we have no idea if any are pollsters! Our ” highly valued” vote may be lost in their counts so I have absolutely no confidence in polls & get a little annoyed – even with Curtice! – when pollsters pontificate on Radio or TV!
Lets not forget the call centres operated out of Arrin Banks’ insurance company offices – illegally – in the referendum AND the 2017 elections (even while campaigning was suspended for a week because of the terror murders!
I have a landline number and phone, but nobody I know ever calls me on it.
Mostly, the callers are phone scams, recently one telling me that my amazon prime account needed urgent renewal (press 1 to connect to suspicious operative with atrocious accent). I don’t have an amazon prime account.
All I, and family members use, is a mobile.
I am a little surprised I actually answered….
We had a new landline installed just over a year ago – ex-directory and only about two people ever ring it – family members who know we will be sitting nearby and having a cuppa.
In the past couple of months I have had two survey calls – the first was about the NHS but I cannot recall much about it. For the most recent one, I asked how they had my number “random dialing equipment” was the reply. It was election related and became more obvious as the questions continued. I am not sure what they made of my responses some of which consisted of laughter.
My brother in law used to run surveys for local authorities – he would always start by asking the authority what response or outcome they required from the survey. In horror they would respond that they genuinely wanted to know what people thought. He would then explain to them what the outcome was likely to be from any given subject – at that point they became uncomfortable with such responses and asked if there was any way to get a better outcome. He could always oblige.
Like Richard, I have to wonder why this sort of polling still continues but I suppose it keeps the service economy moving.
Prof.
I have come across useful studies last night that explains the why? & the how? Of the fake polls.
1. Why? – it seems close polls INCREASE turnout.
https://www.nber.org/papers/w23490
So the purpose of the Opinium type gross distortion is to LOWER the turnout.
It is good to get that proof of what I decided ages ago.
2. How? – well your experience pretty much covers it but as i noted yesterday it is the self selected group of recipients and their repeat inclusion, and ‘secret’ weighting formulae by the pollsters that delivers that result.
https://blogs.lse.ac.uk/brexit/2019/10/29/long-read-are-opinion-polls-pro-leave-biased/
‘A extremely essential and timely piece of research by Thiemo Fetzer (University of Warwick) illustrates some of the technical issues that arise using the example of the British Election Study, which is an important point of reference for much UK political science research. The analysis suggests that repeat participation in the BES panel may systematically skew the implied Leave/ Remain split in favour of Leave. ‘
———
I ask one further question that has been nagging at me for months – why try and lower the turnout?
The only reason I can think of is that it makes it easier to ‘fix’ the result.
Ballot stuffing – the lower the turnout the fewer fake ballots needed to skew the outcome; and so the lower the chance of being caught out by a suspiciously higher turnout!
It’s my theory, I have no proof of the illegality, just circumstancial evidence, that the vector for the ballot stuffing is postal votes. But that makes me a conspiracy theorist according to commentators elsewhere when i have posited my suspicion!
Until the detailed data is handed over by the Electoral Commission and examined by scientific methods and experts – you may want to consider me a conspiracy theorist too and not support that – when I say that was how the brexit referendum was ‘fixed’ (and i think the Scottish one too), going by the suspiciously high turnout in certain counting areas – in excess of 80% even in some.
Can we rely on the supposed integrity of the privatised Electoral Commission at such a crucial election without independent oversight?
In a word, no….
Strangely I had one of these election intention voting calls on my landline and the person would not tell me until the end of the call who had commissioned it – and yes it was the Conservative party
Will you be commenting on the announcement that SSE are moving their UK operations to a non-UK holding company for fear of Labour! Is this the sort of behaviour you would expect from a company with the Fair Tax Mark?
https://www.bbc.co.uk/news/business-50536205
That’s an issue for the FTM to consider. I am no longer a director of the FTM. I am sure it will come up in their appraisal
Interesting story. A polling company using landlines could never include me because my landline is used only as a means to connect to the internet. They might get me on my mobile, but even that is not a sure thing (I hate telephones of all types, and people who know me know that email is the best way to contact me). Relying in landline phones in 2019 is for the reasons you attempted to explain to your pollster guaranteed to produce a biased result.
You need to have a word with the ftm people, because you are still listed on their website as being on the board of directors
https://fairtaxmark.net/who-we-are/team-2/
And also on their technical advisory team
https://fairtaxmark.net/who-we-are/advisors/
I would imagine in the latter category you will be advising them to remove SSE’s accreditation ?
I will ask FTM to update the website
I am advising FTM still
Of course SSE will be reviewed on next renewal
Talking of SSE, I see they won their £200m capital allowance claim against HMRC. I am surprised that a company with large scale litigation against HMRC was awarded a Fair Tax Mark. How was this taken into account in the process?
Maybe you would like to ask the FTM
I am not involved in the SSE appraisal
Attempting to dodge the issue in SSE is going you no favors I’m afraid.
A company who have challenged HMRC on a controversial tax ruling, and who have chosen to locate to a country which consistently features on the TJN list of harmful regimes, can only lead observers to doubt the veracity of the FTM process, and to believe that its really only about squeezing money out of companies wishing to paint one public face to disguise their real behavior.
This has the potential to kill the FTM unless they are seen to take immediate action, not to invent excuses about ‘future assessments’!
With respect, stop trolling
I accept that SSE’s relocation will give rise to issues on re-accreditation. I am sure they will be re-addressed
But challenging HMRC is not an issue. Indeed, as a practising accountant I am at present engaged in a dispute with HMRC on behalf of a client because HMRC is blatantly (in my opinion, and I expect to win) wrong on a legal issue. They make mistakes. They need to be challenged when they do. And that’s not a reason for not getting an FTM.
I’m ducking nothing. I am saying a) tax rulings happen and b) so do disputes and c) that’s why we have courts and d) the new location will have to be considered, but with respect, not by you.
And nor by you, if you are being honest about your non-involvement in the appraisal process, do we will both have the same impact.
Even you however can’t deny that this looks damaging ?
It isn’t damaging because the FTM process is annual and reappraisal is not due yet
To be clear, you are saying that a company with high value litigation with HMRC and more than a decade of open tax audits can get a FTM? SSE doesn’t look like a low risk compliant taxpayer to me.
It has a Fair Tax Mark
And partly that’s because it tells about its tax affairs. After all, how do you now what you say?
Richard, the tax tribunal decision is a public document. It is available here:
https://www.gov.uk/tax-and-chancery-tribunal-decisions/the-commissioners-for-hm-revenue-and-customs-v-sse-generation-ltd-2019-ukut-0332-tcc
The decision makes clear that SSE had tax audits open back to 2006. It also shows the amount of tax allowances in dispute was over £200m.
Are you seriously telling me that companies applying for the FTM are not asked if they have any open tax audits or litigation with HMRC?
I have already answered this question
The issue will be reviewed