Talk:Opinion polling for the United Kingdom European Union membership referendum

Page views for this article over the last 30 days
Detailed traffic statistics

Colour Bias

I would argue that green for remain and red for leave is somewhat biased and isn't very helpful with impartiality. Green usually represents positivity and red represents negativity. Should these be changed? (86.17.120.75 (talk) 02:03, 11 June 2016 (UTC))

Yes: green is soothing and easier to look at for longer periods (that's why they use it in hospitals); red represents anger and discordance and is harder to look at for long periods. I imagine the colours were chosen for precisely that reason: Remain bias. You won't be able to persuade people to want to change them, though. They will argue the hind legs off a donkey to keep them as they are, or if they change them, to keep them as bad or make them worse. I'd advise against wasting your time. I gave up when someone kept a straight face and said the average of the last eight polls here, which is very easy to work out, shouldn't be included, and then when I pointed out that an average (together with an utterly ridiculous version of a 95% confidence interval) is displayed on the graph at the top of the page, he said the rule against "original research" didn't apply to "images". Best not to argue with such people, my friend.Elephantwood (talk) 11:24, 11 June 2016 (UTC)

Actually, I think they were just chosen by copying the tables and colour scheme used in the Scottish independence referendum article (where there was a yes/no answer - at the time this article was created, it was believed that it would also be a yes/no question, and it was only later changed to leave/remain) rather than a deliberate scheme to bias the election. I wouldn't object to changing it, although there's nothing especially obvious that presents itself. Smurrayinchester 13:22, 11 June 2016 (UTC)

I just changed the 'Leave' colours to Blue, but all my edits were immediately reverted. Can we get an agreement to change the Red colouring here, please? — Preceding unsigned comment added by 82.39.97.245 (talk) 10:55, 14 June 2016 (UTC)

We wont get any agreement or change, various people have raised this issue and called for a change more or less since the article was created. There was even daily coulour edit wars when the article wasn't protected. Maybe after the referendum is over they will allow change. — Preceding unsigned comment added by 2A02:C7D:1220:9700:F95A:B4BC:4B15:9B71 (talk) 21:02, 17 June 2016 (UTC)

Question Time specials with Gove and Cameron.

Are they being added into the table? Like with the ITV and Sky debates?  — Calvin999 20:12, 16 June 2016 (UTC)

BMG Poll ( Herald)

The 'excludes don't knows' note against this poll could be misleading. The Herald article () says: 'The results include those who say that they will vote but have either yet to decide how or don’t want to say.' and later in the article that they 'will break 2:1 in favour of Remain, with 66 per cent voting to stay in the EU and 34 per cent to leave.' It also mentions a second BMG poll, this time an online one, done at the same time, which 'found Leave on 55.5 per cent, with Remain on 44.5 per cent, once so-called ‘don’t knows’ were excluded'. Perhaps that one should be added to the table too? Jrc14 (talk) 04:54, 18 June 2016 (UTC)

Having had no answers to this one, I'll go and edit the tables to include the BGM online poll in addition to the telephone poll, and to rephrase the wording about how 'don't knows' are treated. Poll stats taken from BMG's published data at Jrc14 (talk) 09:41, 18 June 2016 (UTC)

References

Semi-protected edit request on 18 June 2016

Please remove polls conducted by Qriously. It's not an official polling agency, these polls differ by a large margin from official polls.

Kyle84UK (talk) 05:58, 18 June 2016 (UTC)

Not done: please establish a consensus for this alteration before using the {{edit semi-protected}} template.  ^B E _C K ^Y _S A ^{Y L} E _S  08:08, 18 June 2016 (UTC)

Qriously Polls - Semi-protected edit request on 18 June 2016

Qriously polls need to be removed as they are in violation of WP:NOTRELIABLE, WP:SPS, and WP:NOR. This company has not conducted election polls before, and these results have not been published in a secondary source.

|- | data-sort-value="2016-06-17" |17 June |32% |style="background: rgb(233, 107, 103);"| 52% |16% |style="background: rgb(220, 36, 31); color: white;"| 20% |1,002 |Qriously |Online |Measures only those "likely" to vote

|- | data-sort-value="2016-06-16" |13–16 June |40% |style="background: rgb(233, 107, 103);"| 52% |9% |style="background: rgb(220, 36, 31); color: white;"| 12% |1,992 |Qriously |Online |Measures only those "likely" to vote"

Zurich gnome (talk) 06:50, 18 June 2016 (UTC)

I'm not sure that it's our job to say who is, or is not, a 'proper' opinion polling organisation. What criteria should we be using to make that decision, and how would we describe those criteria in the article?

It is, as you say, the first time that Qriously have published a political poll, but they are after all a start-up, and they have some history of doing market and opinion surveys on other areas. The methodology, as published on the link you mention, does appear to be consistent with that used by the other survey firms.

And there is a secondary source, is there not, in USA Today: http://www.usatoday.com/story/news/world/2016/06/17/exclusive-poll-eu-support-falls-after-jo-cox-murder/86031038/

Jrc14 (talk) 07:02, 18 June 2016 (UTC)

Not done: please establish a consensus for this alteration before using the {{edit semi-protected}} template.  ^B E _C K ^Y _S A ^{Y L} E _S  08:08, 18 June 2016 (UTC)

To my knowledge the best measure of reliable polling in the UK is whether a company is a member of the British Polling Council. "Qriously", whom I hadn't even heard of before today, aren't. USA Today isn't a newspaper of record; granted, getting polls from them is not quite the same thing as getting weather news from the Daily Express but they're liable to publish sensationalist polling data. Those polls shouldn't be on here. CedarsHale (talk) 09:20, 18 June 2016 (UTC)

I totally agree with above comments Qriously are not members of British Polling Council - so don't use them. I favour Brexit but don't believe those polls for a moment Coachtripfan (talk) 10:46, 18 June 2016 (UTC)

I can see that the Qriously polls are also being reported by le Figaro at http://www.lefigaro.fr/flash-actu/2016/06/17/97001-20160617FILWWW00401-meurtre-de-jox-cox-le-brexit-toujours-en-tete-sondages.php I think they are a serious news source (albeit a French one). As for the general lack of UK news sources discussing this poll right now, maybe that just reflects the fact that the referendum campaign is officially on hold today. Perhaps we need to wait until we see what weight the news media will give these polls on Sunday, as campaigning resumes.

I was wondering whether 'membership of the British Polling Council' might be a good criterion for whether an opinion-surveying organisation was worthy of appearing in our tables here. But is it really for us to make that call? If we have secondary sources referring to this as a poll, I feel we should rely on that, rather than using our individual judgement.

Anyhow, if we do decide that 'membership of the BPC' is the defining characteristic of a serious polling organisation, we will have to weed some other polls out of the tables (for instance I don't think that 'Lord Ashcroft Polling' is a member of the BPC, though it is a respected source.

Jrc14 (talk) 10:12, 18 June 2016 (UTC)

The polls by Qriously should be included because they have been reported in USA Today. That is a secondary source and it makes them notable. There are various associations of pollsters, but there is no such thing as an "official" polling organisation. No licence is required to conduct a poll in Britain. To exclude these polls simply because some editors here hold the opinion that the company that conducted them is not "reliable" would go absolutely against encyclopaedic principles. This company's reliability has not been questioned in any serious and reputable source. It has been opined that the their results diverge greatly from other companies'. First, this is the only published poll based on fieldwork conducted wholly after the murder of Jo Cox and the subsequent suspension of the campaigns, and in that regard there is nothing to compare their results with. Perhaps the mood of the country has shifted a fair bit. These polls, reported in a secondary source, give objective data relating to that very issue. Certainly the murder is the first known murder of an MP for many years and has received enormous press coverage, and many commentators in reputable mainstream media have discussed it in connection with the referendum campaigns. Second, polling results from different companies often do diverge. Partly that is because their sampling methodologies differ; partly it is because they conduct their fieldwork at different times or because the content and context of their questions vary. It is completely irrelevant that some editors feel that this company's field methodology is not to their liking, on the basis of the reported results of the polling. The percentages that this poll reports for Leave and Remain should not form part of anyone's argument here.Elephantwood (talk) 10:25, 18 June 2016 (UTC)

To answer jrc14 Zurich Gnome's concerns: these poll results are not in violation of WP:NOTRELIABLE (that policy requires exclusion of sources that have a poor reputation, an apparent conflict of interest, etc.), they are not in violation of WP:SPS (USA Today is a secondary source, independent of Qriously); and as for WP:NOR, did jrc14 Zurich Gnome not click on the link to find out what the source was? Did he think a Wikipedia editor had calculated the poll percentages for himself?Elephantwood (talk) 10:35, 18 June 2016 (UTC)

Not jrc14's concerns at all :-) I think the question was raised by Zurich gnome - and I agree with your view as to why we should consider the Qriously polls to be notable and reliable, on the strength of the information we have at present.

Jrc14 (talk) 10:47, 18 June 2016 (UTC)

My apologies. I looked for the attribution and the one to you was the first one I saw because the others had other text following them on the same line. I have corrected.Elephantwood (talk) 11:01, 18 June 2016 (UTC)

Edits by Marko801 - vandalism?

Should we consider reverting the edits by Marko801? This user has just deleted a chunk of the table of polls, without any discussion in the talk pages (in spite of the fact that there is a discussion of that section of the table already happening in the talk pages). Jrc14 (talk) 07:58, 18 June 2016 (UTC)

Having looked at the user's page, I have taken the liberty of reverting the edits myself. Can we discuss these two polls here, before making a decision whether to remove the Qriously polls, please. Jrc14 (talk) 08:06, 18 June 2016 (UTC)

BMG poll, 15 June

The BMG website has slightly different numbers to the Herald article currently used in the wiki article; Remain 52%, Leave 48%. They also mention an online poll: Remain 41%, Leave 51%.82.1.16.12 (talk) 08:43, 18 June 2016 (UTC)

Qriously Polls

I see that two poll results have been added for an organisation called Qriously. Are there criteria for judging which polls are sufficiently serious for inclusion? I ask because (a) the poll results seem so out-of-whack with those of other polls, (b) when I visited qriously's website I couldn't see the underlying data (which I can for all of the other organisations), (c) their methodology (in so far as I understand it) seems to be based around injecting questions into web-pages.

In other words, are these serious polls or is this a marketing ploy?

Sorry, I now see that this duplicates one of the previous posts questioning the validity of Qriously. Please forgive this error of a neophyte. I would support the suggestion made in that post of removing these polls.

MrsGussieFinkNottle (talk) 08:51, 18 June 2016 (UTC)

No idea how this all works but I totally agree with this guys doubts... — Preceding unsigned comment added by Oneoffedit (talk • contribs) 13:25, 18 June 2016 (UTC)

Ditto here. This is not a reliable poll from an established and notable polling organisation Jw2036 (talk) 13:41, 18 June 2016 (UTC)

Semi-protected edit request on 18 June 2016

Please delete polls by Qriously. This is not an established polling agency and polls vary by a large margin from official ones.

Kyle84UK (talk) 08:55, 18 June 2016 (UTC)

@Kyle84UK: Please join the discussion below... Firebrace (talk) 13:55, 18 June 2016 (UTC)

RfC on Qriously polls

Should the Qriously polls be included or not? Firebrace (talk) 13:35, 18 June 2016 (UTC)

Survey

Speedy Delete. It's not been reported or verified by any other media source. 'Qriously' doesn't even have a page of it's own on Wikipedia. It has been only self-published on their website. Methodology is suspect - appears to at least in part be targeted marketing rather than an independent poll. Fails on notability and reliability, and possible NPOV issues. Jw2036 (talk) 13:38, 18 June 2016 (UTC)

 Comment: The polls should not be removed again while the RfC is active. Firebrace (talk) 13:48, 18 June 2016 (UTC)

Reply. Why should what is obviously Original Research be included when its normal policy to remove it Firebrace? - Galloglass 15:00, 18 June 2016 (UTC)

@Galloglass: Aren't all polls original research? What makes this one different? I'm not sure who it was commissioned by... Firebrace (talk) 15:10, 18 June 2016 (UTC)

Was it commissioned by anyone? We don't know. Who is this 'Qriously' Have you come across them before. I certainly have not and you might note they are not members or the British Polling Council either. Including this just makes the Encyclopaedia look foolish. - Galloglass 15:20, 18 June 2016 (UTC)

Polls by non-BPC members had their own section at Opinion polling for the Scottish independence referendum, 2014... Firebrace (talk) 15:26, 18 June 2016 (UTC)

This seems like a good solution to me Firebrace, I'd support this table being set up in a subsection below the main one Jw2036 (talk) 15:35, 18 June 2016 (UTC)

Keep it. This has been published by "USA Today" [1] 77.9.78.66 (talk) 14:09, 18 June 2016 (UTC)

Keep: Not in violation of any Wikipedia policies or guidelines. Antiochus the Great (talk) 14:45, 18 June 2016 (UTC)

Speedy Delete. Self published poll. 100% failure on notability and reliability. Including this alongside reputable polling organisations is simply unsustainable. - Galloglass 15:00, 18 June 2016 (UTC)

Keep it. In addition to the USA Today reference noted above, it's also been published by le Figaro [2]", so it evidently does appear in secondary sources. The company have not done political opinion polling before, but they've been in the polling business for some years, and I didn't find any secondary source that challenges their reliability. I don't think there is anything in Wikipedia policy to justify excluding their polls from the table. Jrc14 (talk) 15:05, 18 June 2016 (UTC)

Do you have any source to support their reliablity? I certainly can't find any such source and they are not members of the British Polling Council. - Galloglass 15:14, 18 June 2016 (UTC)

Speedy Delete Not an reputable organisation, this company has never made an opinion poll for an election/referendum before. Their report contains very little data compared to the other companies regarding samples. weightings etc. — Preceding unsigned comment added by Zurich gnome (talk • contribs) 16:35, 18 June 2016 (UTC)

Strong keep. All of the reasons given for deletion are opinion only. The results were not self-published, and it is simply not true that there is "100% failure on notability" when the polls have been reported in USA Today. Nor are there any reliable secondary sources for being sceptical about reliability any more than there are for polls conducted by other companies. (I have read the article by Anthony Wells of the rival company YouGov, in which he says "As far as I can tell the poll was conducted by embedding survey questions in adverts on smartphone apps". Well first, he is not independent. Second, he only says that the use of that methodology is "as far as (he) can tell" - in other words, he is guessing. Third, it is not for us to edit this article according to our own opinions on whether or not that is a good or bad methodology. Are we supposed to read criticisms that polling companies have made of each other's methodologies on the basis of guessing and then go through polling results deciding which ones deserve on the basis of what rivals have said about them to be excluded from mention?)

For this reason, I have deleted the notes to the table entries for the Qriously polls that were sourced to what Anthony King of YouGov guessed about their methodology.

Last, of course there is no "original research" issue here. Be serious.Elephantwood (talk) 18:32, 18 June 2016 (UTC)

Discussion

I have just started Qriously, including something from them about "methodology". Expanding that article might help us to answer this RfC. Edwardx (talk) 15:22, 18 June 2016 (UTC)

The fact it's described as an 'ad-tech company' rings serious alarm bells regarding it's neutrality and reliability! Jw2036 (talk) 15:33, 18 June 2016 (UTC)

If it's true Qriously "serves up questions instead of banner ads on mobile devices and then retargets users with relevant ads" and that was how the polls were conducted then I don't think we should be entertaining them at all. Firebrace (talk) 16:08, 18 June 2016 (UTC)

The poll has been removed pending the outcome of this discussion - Galloglass 15:24, 18 June 2016 (UTC)

Support this. It should be removed until a decision is made, or else it could corrupt the reliability of the info given. IMHO this page should be restricted to pollsters who are members of the British Polling Council who are the body representing and regulating the industry. Jw2036 (talk) 15:31, 18 June 2016 (UTC)

The BPC don't regulate the "industry". What gave you the idea that they did? Companies that don't belong to the BPC are allowed to conduct polls without BPC "regulation". They don't represent the "industry" either - they only represent their own members. The BPC is a private association that companies can apply for membership of and can also be thrown out of. It has no official role. Polling companies do not have to belong to it or have anything to do with it. How many times does this point have to be made?Elephantwood (talk) 18:40, 18 June 2016 (UTC)

But now people who are new to the RfC can't see what we're talking about... Firebrace (talk) 15:36, 18 June 2016 (UTC)

Not sure I agree with the Poll being removed while the RfC is active. Unless the material blatantly violates policy (and in this case it does not), it would be considered best practice to keep the material in the live version of the article, so users are clear as to the content of the RfC. Antiochus the Great (talk) 15:48, 18 June 2016 (UTC)

Antiochus, this is not a real poll. This has been put together by an Ad Agency with people clicking on adds that appear to be polls. There is no weighting, no structure, no click limit to what's been produced. To even call it a poll is stretching credulity to its limits. - Galloglass 15:59, 18 June 2016 (UTC)

Galloglass, It is a real poll and on the referenced site they do describe their sampling and weighting methodology. There could be weaknesses (I am not myself an expert on opinion polling ... but I don't think you are either). Since we have secondary sources that do refer to it as a poll, I don't think we can simply assert that it isn't one. - Jrc14 (talk) 16:10, 18 June 2016 (UTC)

Jrc14, for a polling experts view on this poll can I suggest you go read Anthony Wells comments at Ukpollingreport [3]. you might note that not even he has heard of them. - Galloglass 16:28, 18 June 2016 (UTC)

If we're going to restrict the page to pollsters who are members of the British Polling Council, we must be consistent about it. That would mean also removing, for example, poll results from Lord Ashcroft Polls. That would feel wrong, as that is a well respected polling organisation, albeit not a member of the British Polling Council. Jrc14 (talk) 15:38, 18 June 2016 (UTC)

I reckon Firebrace's solution above of a separate table in the style of the Scots indyref page is the solution. The Ashcroft polls are in another subsection so can remain Jw2036 (talk) 15:40, 18 June 2016 (UTC)

I may be wrong, but I am not aware of any consensus that only pollsters who are members of the British Polling Council should be used in British related articles. It feels like an arbitrary reason to bury the poll elsewhere in the article (out of sight out of mind) or just delete it entirely. Antiochus the Great (talk) 15:53, 18 June 2016 (UTC)

You are aware that this is not a polling organision but an Ad Agency Antiochus? - Galloglass 16:02, 18 June 2016 (UTC)

If you read the latest version of Qriously, they are not what I would call an ad agency, and they do cite some sort of methodology. Edwardx (talk) 16:10, 18 June 2016 (UTC)

Edwardx, would you describe them as a polling company? - Galloglass 16:12, 18 June 2016 (UTC)

Galloglass - no, I would not. Personally, I doubt whether their methodology is as robust as those of the British Polling Council members. And the poll may be somewhat self-serving. Those last two sentences are WP:OR, of course. Nonetheless, Qriously have at least stated in broad terms what their methodology is. I think Firebrace's proposed solution represents a good way forward on this. Edwardx (talk) 16:19, 18 June 2016 (UTC)

They state the nature of their business as "market research and public opinion polling". [4] Firebrace (talk) 16:32, 18 June 2016 (UTC)

Semi-protected edit request on 18 June 2016

Please add recent Ipsos Mori poll: 53% leave, 47% remain, from Thursday 16th June: http://www.standard.co.uk/news/politics/brexit-poll-most-brits-do-want-to-leave-one-week-to-go-before-eu-referendum-vote-a3273141.html

80.6.242.105 (talk) 14:41, 18 June 2016 (UTC)

Poll already included in article. Antiochus the Great (talk) 14:49, 18 June 2016 (UTC)

Not done: The poll was actually done on 11-14 June, it was just published on 16 June. You'll find this on the article. st170e^talk 14:50, 18 June 2016 (UTC)

BMG Poll

The phone poll which shows Remain in the lead assumes that Don't Knows split 2:1 in favour of Remain. How? Why? (####) — Preceding unsigned comment added by Coachtripfan (talk • contribs) 17:26, 18 June 2016 (UTC)

Please read WP:NOTFORUM. Firebrace (talk) 18:11, 18 June 2016 (UTC)