Talk:Opinion polling for the 2015 United Kingdom general election/Archives/2014/December

UKIP Errors!

While working through the polls for 2011, I have started to notice that all the UKIP shares are wrong for the polls I check. I don't know how widespread the problem is, and I hadn't been routinely looking before I noticed an error about 8 hours ago, but since, all 11 polls (all YouGov) I have looked at in 2011 have had UKIP wrong. It might be because all 11 of these polls have been harder to find pdfs for (none were still at the url that had been linked until I changed them), and somebody rather than finding them when they wanted to add UKIP shares just made them up. Or maybe they found them, and kept making mistakes (it's easy enough to do), but either way, it looks like we have quite a bit of false data. I am, of course, correcting. Amongst the questions now is how widespread is this problem? What do we do as a consequence? Should user:impru20 remove the UKIP line from the Graphical summary until it is based on real data? DrArsenal (talk) 16:54, 30 November 2014 (UTC)

Fortunately, UKIP data shown in the graph for that year were not copied from the tables but individually checked on by me shortly before UKIP's inclusion was decided. i.e. in my data sheet UKIP is at 4% in the first days of January 2011 (which corresponds with YG data) whereas in the article a number of random 2% are shown. I can help a bit out with the correction of these figures. Impru20 (talk) 17:31, 30 November 2014 (UTC)

Impru20 it's good to know that your data is better than that table, and many thanks for the offer to help correcting. 'Others' also seem to be wrong, suggesting to me the wrong UKIP was subtracted from the previous 'others'. The Oct 2011 YouGovs are all proving difficult to find the tables of, so if you are able to correct those it would be particularly helpful (I guess you won't have Green shares for them :-( ). Thanks again. DrArsenal (talk) 18:20, 30

November 2014 (UTC)

Or maybe Impru20 does have Green shares already! Many thanks Impru for correcting and adding Green shares for Jan & Feb 2011. You may see that I have done likewise mid July - mid Sept 2011 (YG only). A few of these have had UKIP correct before I checked, ...but only about the sort of proportion that you would expect from someone guessing numbers between 3 and 7 :-( DrArsenal (talk) 23:36, 30 November 2014 (UTC)

I've been working more through mid 2011 - and again, far more UKIP shares were wrong than right. :-( But I estimate between us Impru20 and I have corrected or checked most of 2011. Still a load to go, though, and the problem of finding October2011 YouGovs. Meanwhile I checked back to see who had input all these false UKIP shares - a couple of IP editors (or one using two different IPs) in mid January 2013. Neither IP has been used for IP editing since around Jan 2013. DrArsenal (talk) 09:33, 2 December 2014 (UTC)

Next→2015?

Isn't it time to move the article to Opinion polling for the 2015 United Kingdom general election? As far as I understand, there is no way for this election to be held before 2015, or am I wrong about it? Many thanks, --Checco (talk) 10:08, 2 December 2014 (UTC)

Done. Bondegezou (talk) 10:27, 2 December 2014 (UTC)

Very good. Thanks, --Checco (talk) 12:54, 2 December 2014 (UTC)

Green Party

I hesitate to mention this, because it is potentially a lot of work. The Green Party is now showing 5% in a number of polls (including the most recent Populus and YouGov polls), and in the last Ashcroft poll, although they were not separated out in the final published figures, were at 7% before the "don't know" adjustment was made. This is not far short of some of the LibDems' polling figures, and they have shown an ability to beat the LibDems in some real elections including the European Parliament elections and Westminster by-elections. So at what point should we consider adding them to the table as a separate party, as was done for UKIP?Saxmund (talk) 10:51, 8 June 2014 (UTC)

I'm sympathetic to the Greens' inclusion, but I think the best rationale to do so would be if they begin polling consistently above the Lib Dems in the same vein as UKIP. JJARichardson (talk) 11:21, 8 June 2014 (UTC)

As per UKIP, the rationale is to follow reliable sources (pollsters themselves, not newspapers). The lib-dems are not a yardstick. Also 5% ain't at all noteworthy....this says more about lib-dems collapse than Greens boost. Iliekinfo (talk) 11:35, 8 June 2014 (UTC)

I had a look at the archive discussions about the inclusion of UKIP (which I wasn't involved with at the time), and yes the tipping point seems to have been when they started consistently polling level with or above the LDs. This effectively makes the table inaccurate - it seems odd to include parties no 1, 2 and 4 but not 3. There would be a clear implication that all the "others" are more minor parties, which would not be true. So I disagree with Ilikeinfo: I think the LibDems are a yardstick as they are currently consistently the 4th party. If they started being 5th on a regular basis it would seem odd to give separate scores for the 1st, 2nd, 3rd, and 5th parties but not the 4th. And I am not sure why you referred to "following newspapers" as the data I mentioned was from polls (brought to my attention by comments on the ukpollingreport website, as it happens.)Saxmund (talk) 11:53, 8 June 2014 (UTC)

Also - 5% might not be particularly noteworthy, but of course the decision having been made to backdate UKIP results to 2010, there are now plenty of instances where UKIP scored well below 5% and are listed. In some of these cases, the Greens may have scored more than the 4th party listed, but are relegated to Others.Saxmund (talk) 11:58, 8 June 2014 (UTC)

Mentioned following newspapers because I generally despair at the way (most) wiki editors think of newspapers when they hear RS, and tend to think WP would be better if they weren't considered such. (does anyone smart consider them reliable IRL), I didn't mean to imply you were doing so....it was a preemptive strike if this argument drags on & on. "it seems odd to include parties no 1, 2 and 4 but not 3." Why? For years we've had parties 1-3, but not 4. "I think the LibDems are a yardstick as they are currently consistently the 4th party." Why? No-one suggested the 4th party was a yardstick before the lib-dem collapse. The UKIP debate was one group arguing for UKIP as soon as they had one poll above lib-dems, and one group arguing against UKIP until they won an MP. IMO both are wrong, WP should follow RS, the point of addition should have been when pollsters separated UKIP from others....the same way tables of polls of NI, Wales, Scotland, and individual constituencies work (I think, they should). I can't see any justification for using 'beating the Lib-dems' as a rule, there's nothing special about them. Unless we agree (as I think was proposed before) a firm % cutoff for a party to beat on average over a firm agreed time (mebe 3 months averaging over 5%???), I think we should just follow the pollsters. EDIT CONFLICT: Yeah, the Greens, the BNP, the SNP, PC, ED, & an assortment of NIish. IMO UKIP shouldn't have been edited back to 2010, but that's probably a different discussion. Iliekinfo (talk) 12:13, 8 June 2014 (UTC)

Trying to be politically neutral it seems reasonable to list the Top 3 or Top 4 parties and lump the rest all in together as "others". But not, say, the Top 2 and the 4th and lump the 3rd in with others. It seems to be making a judgment about party no 3. Similarly I think we ought to include the Greens if they start consistently scoring at or above the 4th party score, whether that party is the LDs or UKIP.Saxmund (talk) 12:35, 8 June 2014 (UTC)

Ah yes, I somewhat misread one of your comments. I can certainly see that argument, but A - we need an agreed upon method of ranking the parties which isn't going to flip flop every poll, & B - If the time came, do you want to be the one to remove the lib-dems from the table and see what happens? If we were supposed to have a fixed number of parties...lib-dems would've been scrapped by now...:) Iliekinfo (talk) 12:43, 8 June 2014 (UTC)

I'm not talking about having a fixed number of parties! Simply that it would seem odd to mention the three largest and a 5th, when another party regularly comes 4th but is relegated to "others". It hasn't come to that yet and maybe once the results of the EP elections are out of the way the Greens' poll rating will subside. However worth noting that Greens and LDs are level in this week's Ashcroft poll and, as he has provided us with a comparison with last week, can now be said to have been ahead of the LDs in that poll on published headline figures (if retrospectively).Saxmund (talk) 18:49, 9 June 2014 (UTC)

I know that's not what you're proposing, but I'm saying that it's the logical outcome of argument from ranking. The point that if the greens overtake the lib-dems (or UKIP, or whomever) they should be included because they're "beating" LDs, implies that inclusion on the table should be based on ranking, (a very logical position, even if I disagree) in which case we should agree on which ranking qualifies. 3rd, 4th, 5th? The idea seems to be that we should add the Greens if they overtake LDs, but not remove LDs (wikipolitically impossible I'd imagine). My point is that if LDs would/should still be included as 5th placed, then the current 5th placed party (Greens) should be in now. The vague qualifying standard should not be changed as needed to fit the lib-dems....the idea that LDs qualify as 4th/5th but UKIP/greens didn't/won't is inherently unintentionally biased. I'm just trying to say that if we don't go with RS we need a clear consensus on what qualifies a party for inclusion, and it needs to be applied neutrally. IMO the logical endpoint of your (entirely sensible) argument is that we either add the greens now, or are prepared to remove the lib-dems should they slip to 5th. An alternative suggestion is we could agree that a party that consistently polls above X% warrants inclusion...either way, if we, rather than pollsters, are to decide who makes the table, we need a clear system for doing so. Iliekinfo (talk) 03:41, 10 June 2014 (UTC)

I don't think it's the logical outcome at all. I suspect that the tables started with what would universally be regarded as the "top three" political parties. Easy to form consensus around that. UKIP broke into it in about 2012 and have been subsequently added to the tables, they are now polling up to twice the LDs (three times in last week's Ashcroft poll) and can for the time being be regarded as a "main party". You are right, there is no way we can remove parties once added, it would mess up the tables and the graph even more, and upset their supporters, and of course their poll position might improve (as might the LDs when we get closer to the election). My argument is that if we are showing the top x parties and another party moves into the top x, we should start showing the top x+1. How far back we go is another matter. I am not sure how many pollsters have been separating out the Greens in their final table. I suspect when UKIP made it into the Top 3, there were UKIP partisans willing to do the work to backdate to 2010 - I see no sign of Green partisans on here (it's just you and me arguing at the moment) and I raised it as an intellectual query, I'm more of a Tory. PS what is "RS"?Saxmund (talk) 08:14, 10 June 2014 (UTC)

"My argument is that if we are showing the top x parties", if that's the idea, what is x? "you and me arguing" Friendly pedantry? RS = Reliable Source. Iliekinfo (talk) 14:23, 11 June 2014 (UTC)

At the moment, x=4. Before UKIP was included, x=3. If we include the Greens, x=5. But the argument holds for any value of x. Saxmund (talk) 18:25, 12 June 2014 (UTC)

"But the argument holds for any value of x." If the system is to be "the top x parties", then yes, but under that system changing the value of x to accomodate/exclude certain parties is irrefutably biased. If the rationale for inclusion is to be the 'top x', then we obviously need to know what x should be. It is my unshakable opinion that under that system (which is not my favoured system) changing x from 4 to 5 in order to continue to include the lib-dems (or whomever) would be indefensible. If the system is to be the top x parties, then x should be fixed. Iliekinfo (talk) 03:29, 13 June 2014 (UTC)

Why? the principal seems sound to me. We don't exclude anyone who is already being included, but we add a party if its score moves it consistently above a party that was already being featured. Where is the bias? Saxmund (talk) 22:23, 13 June 2014 (UTC)

That would be bias in favour of existing parties. Obviously. Iliekinfo (talk) 17:35, 3 September 2014 (UTC)

The Greens now seem to be neck and neck with the Libdems (see latest YouGov). Time to include them? — Preceding unsigned comment added by 147.83.123.130 (talk) 09:28, 30 September 2014 (UTC)

The last yougov I saw had them on 5, lib-dems on 7, on average polls are showing ~lib-dem 8, green 5. More to the point, the score relative to the lib-dems is not the quality for or against inclusion. Wikipedia follows reliable sources, and the table should be either the "main/headline" parties as reported by RS, or all of the parties judged by Opinion Pollsters worthy of reporting individually (rather than "others"). Currently this article has two tables, once for the headline parties (which currently doesn't include the greens), & one for all parties judged noteworthy by pollsters (which does). It is possible that the Greens will become a headline party if they keep growing....but bear in mind that most of their "catching up" to lib-dems is lib-dem falling, rather than green growing. Iliekinfo (talk) 18:45, 1 October 2014 (UTC)

"As per UKIP, the rationale is to follow reliable sources" . Iliekinfo 11:35, 8 June 2014 Well, RS presumably includes UKPollingReport (UKPR), where the most recent post, 9020, gives Green share in the 1st sentence. Similarly the Sun, who commision YouGov polls tweeted "CON 30%, LAB 35%, LD 9%, UKIP 15%, GRN 5%" tonight (I'm not saying the Sun is -or isn't- generally a reliable source, just that on the particular issue of the polls they have commissioned they are one). UKPR post 9014 gives Green share for 3 of 4 pollsters, 9013 gives Green share for 1 of 3 pollsters, 9010 for 2 of 3 pollsters, 9008 for the only pollster mentioned (I haven't looked back before indyref). And whether you judge most of Green catching up comes from LD falling of Green growth in share depends on your starting point. Since around the turn of the year, Green share has probably gone up more than LDs has fallen, and movements tend to have mirrored eachother - see http://www.statgeek.co.uk/wp-content/uploads/2014/09/ld-green-uk.png. DrArsenal 87.112.138.68 (talk) 22:18, 9 October 2014 (UTC)

They're certainly beginning to get mentioned in some headlines, but only as being notably unusually high. I agree this is the sort of thing I was talking about, let's see if it becomes standard rather than the exception. A (non-scientific) google of latest news reports of national opinion polls finds the greens not in the headlines in any I stumble across, just the same 4. They're definitely heading in the right direction though, keep an eye on it.....the decision will presumably be made at the 2014/2015 table break, so see how it goes. UKPR is an IMO excellent source, if Greens make it into that 'latest voting intention' table, that'd be a very strong point in their favour. Iliekinfo (talk) 02:22, 10 October 2014 (UTC)

Wow: even TNS are now listing Green Party shares [[1]]! DrArsenal (talk) 19:35, 2 December 2014 (UTC)

Graphical summary

Could we add the green party in the Graphical summary? --81.58.144.30 (talk) 14:05, 14 November 2014 (UTC)

Agreed. Someone who knows how to edit the graph should do so.Chessrat (talk) 00:27, 17 November 2014 (UTC)

I made an identical request a few days ago. Can anyone do it? Maybe User:Impru20? Thanks in advance, --Checco (talk) 11:12, 18 November 2014 (UTC)

I had this conversation on my talk page:

Hi, I understand from user:Impru20 that he will add if polling results for Greens are in the main table for the entire period 2010-2014. Since I learnt that from Impru, 7 days ago, I've added 7+ months of Green Party shares in 2014 and to 31 rows of the 2010 table (that I had to add the new column to first, and is harder, because fewer of teh URLs are correct after 4 years). Perhaps you want to join me in filling in the data? DrArsenal (talk) 18:20, 17 November 2014 (UTC)

OK, thanks for all the work! I'll start adding some old Green Party figures :) Chessrat (talk) 02:46, 18 November 2014 (UTC)

I'm not certain that Green Party figures should be backdated to 2010, but it won't do any harm. Chessrat (talk) 13:40, 18 November 2014 (UTC)

In fact, we don't need that, but thanks to all of you for your good work! --Checco (talk) 16:10, 18 November 2014 (UTC)

I'm not sure why you say "we don't need that". You might be right, Checco, or you might not.

Neither I nor Bondegezou are entirely convinced that Impru20's graphs are entirely in conformity with WP:SYNTH, but it seems to me that others hereabouts like Impru20's graphs (and I know there have been discussions of alternatives in the past). Other graphs already exist, like http://www.statgeek.co.uk/charts/uk-twelve-month-trend.png, but have disadvantages (that one is only a year's worth, and I'm sure people want a graph for the whole parliament). It isn't terribly difficult to cut and paste data into a spreadsheet to generate a graph, but what would be more difficult would be getting agreement for a different particular graph. And the graph is also on the United Kingdom general election, 2015 page, so getting the graph changed on that page is also relevant. I would reckon it would be unreasonable of Impru20 to insist on every single poll (especially since in 2010 many are no longer available) before changing, but at the moment, there is no data on this page for the Greens between Dec 2013 and Dec 2010 DrArsenal (talk) 19:54, 18 November 2014 (UTC)

Although there may be complications, I don't think it's an unreasonable request from Impru20. As long as the bulk of the polls are tabulated below that should satisfy that there's enough data points to include a Green trendline to the existing graph, which I believe is worth keeping. I'll try to contribute in adding the backlog. HeadlightMorning (talk) 20:26, 18 November 2014 (UTC)

For what it's worth, those Statgeek (me) charts cover YouGov only, and have just been updated having seen this talk page. With regards which parties to list, here are my own thoughts on the matter - http://www.statgeek.co.uk/polling/ - "From the perspective of my charts, the Greens get mentioned, despite their low poll rating simply because they have an MP in Westminster. I feel that this gives them a right to have some polling exposure. UKIP are consistently polling well over 5% in some regions on the UK and in the UK-wide poll itself. For that reason, they get a mention. The BNP neither poll highly, nor have an MP in Westminster, and hence they don’t make the charts. If the Greens lose their MP and don’t raise their polling scores, they will probably be removed from the charts in future." - That was dated a couple of years back, so with UKIP far higher in the polls now, it sounds a little odd, but the rule still applies. My own opinion is that we should try to include parties, rather than omit them, but only if their omission would create regular questions as to why. Camieabz (talk) 01:32, 20 November 2014 (UTC)

Thanks Camieabz both for your charts and for explaining that they cover YG only. I was wondering how the lines were coming out at different places from where I would expect from my calculation of 10 poll rolling means of the polls on this page! Thanks also for your comment on the principles of which parties should be included. I am not arguing that you should put in the work to implement on your own charts (since they are your own to do what you wish with), but wonder whether SNP really ought to be included on your criteria? DrArsenal (talk) 11:34, 23 November 2014 (UTC)

The SNP have seats in Westminster, and if current polling carries through to May will probably increase that seat count (perhaps even so as to be the 3rd largest party in the UK). My own criteria is limited by YouGov's data. I can't account for Plaid Cymru, as YouGov use Midlands & Wales combined (I believe they do have them separately, but combine them, as the Welsh sample would be minimal (less than 100)). Like I said previously, the inclination ought to be to include all we can, rather than look to omit. Otherwise UKIP wouldn't have been included until after Clacton, or the Greens with their one seat, despite both being ahead of the Lib Dems (the latter very recently). Camieabz (talk) 01:54, 28 November 2014 (UTC)

I think I've just found most of the 'lost' tables for 2010 YouGovs by replacing the urls in the way I did for 2011 :-) DrArsenal (talk) 13:23, 3 December 2014 (UTC)

Almost all the green figures have been added now, Ive been doing a few here and there. We should be ready for the addition to the graphical summary soon enough...--ERAGON (talk) 13:00, 6 December 2014 (UTC)

Graph label

The average polls graph appears to have been updated, but the label is still '6 May 2010 - 7 Nov 2014'. Just FYI. 86.2.12.245 (talk) — Preceding undated comment added 09:31, 14 December 2014 (UTC)

London (and other) regional polls

At Talk:Opinion polling in United Kingdom constituencies, 2010-5 there have been a couple of suggestions about what to do with London (and other) regional polls.

Do they belong on this page, perhaps re-naming the "Scotland, Wales and Northern Ireland", or perhaps in a new section, or do they belong on Opinion polling for the 2015 United Kingdom general election in individual constituencies, in a new section (any suggestions what to call it)? DrArsenal (talk) 23:56, 16 December 2014 (UTC)

I see rather than discuss, An IP editor, 123.2.85.195, has changed the section on this page to 'regional polling'. I wouldn't be amazed if this is User:Onetwothreeip. I can see the logic for the change, but I'm pretty sure some from Scotland, Wales and Northern Ireland (I suspect User:Clyde1998, User:Crazyseiko and User:CFindlay12 may be among them) might not be happy with the new section name. I'm convinced need to discuss what to do about London polls, and which page is the more appropriate. DrArsenal (talk) 11:40, 17 December 2014 (UTC)

You are correct! I see no logic for the name change whatsoever. However if there is a consensus that London polls should be included on this page, renaming the section "Constituent Country and London Polling" (or something along those lines) would be accurate and could be an option. CFindlay12 (talk) 12:29, 17 December 2014 (UTC)

I have no idea what you lot are on about, and I would not be that london sqe to do that either. --Crazyseiko (talk) 14:55, 17 December 2014 (UTC)

I don't why this would be such a controversy. I'm the one who created the section and named it "Scotland, Wales and Northern Ireland". I've now changed the section title to regional polling, as Scotland, Wales, Northern Ireland and London are all regions of the United Kingdom. There shouldn't be a problem in placing the London polls here, since it's not one individual constituency and is a regional group of them. 123.2.85.195 (talk) 23:56, 18 December 2014 (UTC)

Update: London added. Onetwothreeip (talk) 04:36, 19 December 2014 (UTC)

I'd suggest "sub-national polling" or similar heading might be more appropriate. Agree the best place for them is this page rather than the constituency one. Andrewdpcotton (talk) 11:06, 19 December 2014 (UTC)

The main problem is the ambiguity. Scotland, Wales and Northern Ireland are all reasonably referred to as "nations", while London is not. While the three constituent countries are indeed constituent countries, all four of these areas are regions, which is the only term that could fit all four polls. Onetwothreeip (talk) 23:05, 19 December 2014 (UTC)

While it is possibly pedantically correct to say that England, Scotland, Wales and Northern Ireland are all regions of the UK, the current practice in Government is to refer to them as (Constituent) Countries. London is a region of England.Saxmund (talk) 00:13, 20 December 2014 (UTC)

I understand that the four countries of the United Kingdom are all constituent countries, but it is not necessary to say so when it does not fit the syntax. While London is a region of both England and the UK, the polling conducted for London is not relevant to being a region of England but rather a region of the UK, as is the polling for Scotland, Wales and Northern Ireland. ((It's also confusing as people often think Scotland, Wales and Northern Ireland are "constituencies" by any sense of the word.)) While it would be unfactual to say all the constituent countries are polled, I believe that National and regional polling is a compromise that fulfils that readers do not mistake Scotland, Wales and Northern Ireland for regions of England. Onetwothreeip (talk) 00:27, 20 December 2014 (UTC)

"Country" is still the accepted term for one of the constituent parts of the UK, not "nation", The word "national", when unqualified, normally refers to the whole of the UK. United Kingdom uses this terminology. Saxmund (talk) 08:01, 20 December 2014 (UTC)

"Sub-national" would fit both Scotland and London then. Onetwothreeip (talk) 09:43, 20 December 2014 (UTC)

As a Scottish editor I'd say sub national is reasonable as it implies below the UK level but regional is potentially offensive as it implies that Scotland is a region rather than a constituent country. Other editors may have different feelings on this but regional definitely feels wrong for a section covering Scottish polling.Andrewdpcotton (talk) 10:06, 22 December 2014 (UTC)

Detailed poll results (December)

Is there any point in keeping the detailed poll results now? There aren't even that many poll results put there, and they haven't been updated for over a month either. They were initially a compromise to have Green results on the page, but that's now redundant since they're on the main table now. SNP and PC results are also redundant due to separate polling for Scotland and Wales respectively, leaving only the BNP which often has less than 10 supporters in each poll. I hesitate to remove significant parts of articles, but I'm pretty sure most people would find this section unnecessary too. Also how about setting the tables to default hide again? Onetwothreeip (talk) 02:19, 21 December 2014 (UTC)

WP:IMPERFECT is very clear on this. We don't remove material just because it's incomplete and needs more work doing. Bondegezou (talk) 22:40, 21 December 2014 (UTC)

I'm suggesting it's unnecessary and redundant, even if perfected. Onetwothreeip (talk) 23:01, 21 December 2014 (UTC)

I think I added most rows, and the vast majority that were added after the initial burst. Since I stopped adding them, I think only one has been added. The problem, Bondegezou, with 'incomplete and needs more work doing' is I can't see any prospect of that work being done, and in places it is inferior to the tables in the main listing, where updated links to data have been found since the 'detailed' table was created. I would suggest setting the tables to default hide, as Onetwothreeip says, may be sensible for now (perhaps interest will be re-kindled if/when SNP start polling above LDs). DrArsenal (talk) 23:31, 21 December 2014 (UTC)

Onetwotheeeip: it contains information not otherwise covered in Wikipedia, so I don't see how it is redundant. DrArsenal: I've nothing against default hide. But WP:IMPERFECT is pretty clear here and it's policy. Bondegezou (talk) 00:26, 22 December 2014 (UTC)

I was suggesting the default hiding of the main tables except for the current year. I don't know why that was changed.

The only practical information provided by the detailed table that isn't detailed anywhere else is polling for the British National Party, which wrongly implies that they have any supporters anymore. Onetwothreeip (talk) 00:34, 22 December 2014 (UTC)

The detailed table adds information on the SNP, PC and BNP. The decline of the BNP since the last election is notable. That the BNP has minimal support now is significant given that, at the beginning of the period covered, they were the 5th largest party (by votes cast at the 2010 election). UK-wide polling has been more frequent than polling in Scotland or Wales, so the performance of the SNP/PC in national polls adds information too. Bondegezou (talk) 14:55, 22 December 2014 (UTC)

I doubt many will think a decline from 2% to 0% is one of the notable facts from these polls. BNP was politically irrelevant in 2010, and they're politically irrelevant in every year since. The information on SNP and PC in this table is very unreliable. Numbers like 3 to 5 percent of the British population is not enough to make a proper analysis of their performance, and are superseded by far superior polling in Scotland and Wales individually. Onetwothreeip (talk) 23:20, 22 December 2014 (UTC)