Talk:Opinion polling for the 2015 United Kingdom general election/Archives/2013/January

Lib Dems and UKIP.
This seems absolutely ludicrous to me......UKIP has a growing presence (challenging for 3rd place) these dyas, yet it is absent from most graphical presentations "becasue it wasn't there before"????? Time to rid Wikipedia of these Luddite practices and present current facts and facets, not follow the biased attitudes of the Traditional Press. — Preceding unsigned comment added by 86.5.171.186 (talk) 11:27, 21 May 2012 (UTC)


 * So, who's more biased? Those, like me, who are asking for a good enough justification before adding them, or those who order everyone else that UKIP must be added just because they say so? It's just 1-2 months since UKIP began to surpass the Lib Dems as 3rd political force, and we're still 3 years ahead of the next general election. If UKIP's boost turns out to be a temporary one, there'll be no reason for adding them to the table. Impru20 (talk) 14:00, 21 May 2012 (UTC)

at the moment Lib Dems are averaging around 2% above UKIP which is quite likely that UKIP could become the 3rd party in a while if that happens would UKIP be included on this table? C. 22468   Talk to me  15:27, 22 January 2012 (UTC)

... your premise is simply wrong. To take as an example, the poll immediately below as I write - LDs are 11%, UKIP 5%. Which makes the LDs 120% higher, 6% in raw terms being 3* your claim. LDs are on a historic low of 9-15% depending on the organisation, UKIP rarely exceed about 4%. LDs are a party of government, with nearly 10% of the Parliament, UKIP are nowhere near winning a single seat --Saalstin (talk) 16:48, 22 January 2012 (UTC)


 * This week Lib Dems were just 2% ahead in the polls today they are 4%, at the election they were 20% realistically is could happen and the poll below was made nearly a year ago before the issues with Cameron and Europe. C. 22468   Talk to me  17:41, 22 January 2012 (UTC)


 * 'at the moment' implies a current trend, 'this week' is at best a single snapshot from a single poll. To take an alternative, in mid-december, when the debate was at its height, http://www.icmresearch.com/wp-content/blogs.dir/1/files/2011/12/2011_ST-Dec_poll.pdf ICM] found LDs at 11%, and UKIP at 3% - well under 1/3 of LD support.  This might need to be discussed when there's any real evidence of it being an issue, but currently... no :) --Saalstin (talk) 17:54, 22 January 2012 (UTC)
 * Ok C. 22468   Talk to me  18:45, 22 January 2012 (UTC)


 * UKIP are a minor party. They have little representation in UK governance. More specifically, they have none in the parliament to which these polls refer. The SNP and PC, by contrast, do have MP's and yet they are not included in this table. Thus unless and until the SNP, PC and other minor parties are included, UKIP ought not to be included either. OldSquiffyBat (talk) 12:33, 4 April 2012 (UTC)


 * I'd say that UKIP should be included if either they start significantly outpolling the LibDems (i.e. the LDs are at 8% and UKIP is at 12-14%) on a repeated basis or if their dead heat with the LibDems continues for perhaps 8 months (i.e. through the end of the year). I'd also toss them in if they manage a by-election win.  But none of these conditions have been met, and there have been other surges sort of like this in the past.  One footnote to all of this: I would be in favor of including them in a shorter-term chart if we were within, say, six months of a likely election...but that's mainly because at this level, they'd probably have a reasonable shot of making a hash of the election for the Tories and excluding them would ignore a notable part of the national picture.  Finally, I would point out that there have been quite a few cases of minor parties being included on the charts for the Canadian provincial elections, and adding a fourth party isn't excessively cluttering.Tyrenon (talk) 22:11, 13 July 2012 (UTC)

Initial serious discussions in April
In tonight opinion poll UKIP are in 3rd place (http://labs.yougov.co.uk/news/2012/04/16/update-labour-lead-11/) should the table be changes to reflect this? C. 22468   Talk to me  21:11, 16 April 2012 (UTC)
 * Perhaps. Let's see if that trend continues, and then we'll discuss it further. Bkissin (talk) 15:21, 17 April 2012 (UTC)
 * Polls for 17 April have UKIP on 9% compared to LD on 8% and polls for 18 April have UKIP on 8% and LD on 10%. I propose that a fifth column (titled UKIP) should be added to the polling "list" since the election, enabling people to see how support for Lib Dems declined, this "transitional period", then UKIP settling for 3rd place, which is what I predict will happen unless Brian Paddick does something incredible in the London election. Spa-Franks (talk) 20:53, 19 April 2012 (UTC)

Is this the best chart to put it?, even if ukip aren't in 4th they are never more than 2 points behind the Liberal Democrats, it wouldn't be a good idea to Include SNP as they are Scotland only and therefore could never get above 9%. C. 22468   Talk to me  21:08, 20 April 2012 (UTC)


 * I think we should leave it for a slight while more. UKIP may indeed vie with the Lib Dems for third place in opinion polls on a consistent basis. Then I think it would be justifiable to introduce these four columns at the point where this began. Sir Richardson (talk) 21:43, 20 April 2012 (UTC)

I would recommend giving it some time to see how things develop. Remember that we can only report what polling companies comment on, and since most polling reports only detail the three main parties (Labour, Conservative, Lib Dem) we normally are unable to read much into the "other" figure. That being said, should polling companies establish a consistent recording of Ukip's showing in the opinion polls, then there may very well be a case for permanent inclusion. Until then, should Ukip maintain its position, I would suggest footnoting the "other" figure as a makeshift solution in the meantime. What do other editors think? —JeevanJones (talk) 22:02, 20 April 2012 (UTC)


 * In the event we do need to start listing UKIP, or any other constitently high polling party, the way to go might be some sort of collapsible cell for the 'others' column. The example table below is constructed using the NavFrame code, and gives an idea of what I mean. The problem with using this is I can't find a way of overriding the colour parameter, so it would always appear blue. I'll keep fiddling with it to try and find a way of doing this in white background, but the table below gives an idea of what I am talking about. I think this method would be good as we can add UKIP/SNP/other party high scoring polls without creating an overly cluttered table with too many columns.


 * If, by any chance, anyone knows a way of overriding the colour parameter coding, please let me know. Otherwise, what are people thoughts about the possible route?Spiritofsussex (talk) 09:28, 21 April 2012 (UTC)


 * I like the idea of the collapsible box, I think it and maybe have the collapsible box with all the other parties as well in it, if UKIP are regularly the 3rd party and are getting 3 points or more above the lib dems regularly then they should be included proper. C. 22468   Talk to me  13:02, 21 April 2012 (UTC)


 * I like that collapsible box. That should avoid it becoming too cluttered. As I've said, I think we should wait to see if Ukip's poll lead remains noteworthy before inclusion, and that's both in the collapsible box or as a new column. We have to be careful in not appearing to shape the direction of politics, and only archive it. —JeevanJones (talk) 21:51, 22 April 2012 (UTC)


 * Agree with the collapsible box as per JeevanJones. See also comments below as to why UKIP (alone) cannot be added at this point in time. And the advice of  Richard  BB  for people to stop pushing this issue when it is already being discussed here. OldSquiffyBat (talk) 14:24, 30 April 2012 (UTC)

The YouGov poll published today has shown UKIP on their highest in any other YouGov poll of 10%, so please add a UKIP table.(97daviee (talk) 08:32, 29 April 2012 (UTC))


 * We're still in the process of deciding whether or not they should have their own table: if their popularity does not remain, we won't include one. Please stop adding these new sections when we have an active discussion here. –  Richard  BB  10:30, 29 April 2012 (UTC)

How long will that process in deciding to add a UK Independence Party Table last?(97daviee (talk) 16:18, 29 April 2012 (UTC))


 * No, UKIP ought not to be added separately. Why? Because until such times as UKIP break-through and have MP's elected, they remain little more than a protest party with regards to any general election. (Which is, after all, the subject of this article). However, if UKIP are to be added separately, then parties such as Plaid Cymru, the Scottish National Party, the Green party and even the likes of Respect - who do have an elected MP, unlike UKIP - must also be added. And anyone making the case for the inclusion of UKIP must also make the case for these others. To not do so smacks of POV pushing. OldSquiffyBat (talk) 14:20, 30 April 2012 (UTC)
 * You obviously don't understand how opinion polls work, opinion polls don't measure the parties by seats, they measure them by vote share. UKIP will overtake the Lib Dems permantly, and you will have to add UKIP soon before people start accusing wikipedia for being Anti-UKIP.(97daviee (talk) 19:23, 30 April 2012 (UTC))
 * "UKIP will overtake the Lib Dems permantly [sic] " -- is what you just said. You cannot possibly know this, and it seems by this statement -- not to mention your others -- that you are indeed PoV pushing UKIP. There is no rush in this debate. Consensus will be reached, but accusations such as calling Wikipedia anti-UKIP do not help. First we need to see if their popularity will remain (and I expect their fighting for third place has little to do with UKIP popularity, and more to do with Liberal Democrat unpopularity). Then we need to decide what OldSquiffyBat said: whether we should include a party that has no elected MPs. –  Richard  BB  11:37, 1 May 2012 (UTC)
 * I am not defending UKIP, I just think this article should be corrected. I did not say Wikipedia was Anti-UKIP, I just mean that's what people will start thinking. If I was just a random person who just wanted to take a quick look at the latest opinion polls via Wikipedia, and found that UKIP was nearly close to being permantly ahead of the Lib Dems, I would in a state of confusion as to why a UKIP table has not been added. It just makes this article seem bias. Personally, I think Wikipedia is one of the most trustworthy and most accurate sources of information on the web, so to avoid it from being labeled bias they should add a UKIP table. I agree with you that until UKIP has retained their popularity permantly they should not be added just yet.(97daviee (talk) 13:31, 1 May 2012 (UTC))
 * For those who still believe that the UK Independence Party should not have their own table, you are damaging Wikipedia. UKIP has had their best result in any of the Local Elections ever. So please to avoid this from damaging Wikipedia add a UK Independence Party table now.(97daviee (talk) 17:52, 4 May 2012 (UTC))


 * This is getting a bit heated, so what we need is (before the serious point) a good politician's quote: "Calm down, dears!" Now, on to finer points. Whilst trying to keep my Conservative bias out of it, since YouGov think that UKIP is splitting the centre-right vote (as the Social Democrats did to the left in the 1980's) then I really think that they should be included seeing as all the interviewees on the BBC and Sky coverage of "Vote 2012" (aside from the Lib Dems, obviously) that this is "the beginning of the end of the Lib Dems" and/or that "minority parties are coming into their own." I do think UKIP should be included. One problem, some of the 2010 polls don't have a result for UKIP. Spa-Franks (talk) 21:16, 5 May 2012 (UTC)


 * I 100% agree with you. Most of the people on this talk page agree that UKIP should have it's own table, so why aren't they adding it.(97daviee (talk) 22:54, 5 May 2012 (UTC))


 * If UKIP is to be included, the table would need modification in order for it to don't look too cluttered. I propose doing a similar thing to what I did to Spanish opinion polls tables (in which there are four national parties with seat representation): http://en.wikipedia.org/wiki/Spanish_general_election,_2011#Opinion_polls


 * My proposal would be for the table to look like this.


 * Reduced the size of the table and the font in order to accomodate the new column. I also shortened the main party names for the same purpose. In my opinion, though, we should wait for UKIP to get representation before including them. After all, these are just surveys, which may change over time and may not translate into reality so accurately. If UKIP voting intention suddenly lowers again, we'd be adding them for nothing. I think we should draw a line about which parties should be included and in which circumstances. Impru20 (talk) 10:21, 6 May 2012 (UTC)


 * I agree completely, Impru20. The table needs to be drastically reduced in size, as the "lead" section is far too big. As for adding UKIP: again, I agree with you. Their current boom could be only temporary; and, as I've said, I think it's less to do with UKIP's support and more to do with the Lib Dems' lack of support. –  Richard  BB  10:41, 6 May 2012 (UTC)


 * In response to what Impru20 & Richard BB said about UKIP's boom possibly only being temporary, to some extent they could be right, but you can't keep on using that as an excuse for not adding a UK Independence Party table. I will shut up if you can give me the period of time UKIP's boom has to last until we add the party an individual table.(97daviee (talk) 19:00, 6 May 2012 (UTC))


 * Richard BB, I agree with you on most of this, but do we know for a fact that UKIP is sucking support from the LD? You'd think that they would be taking Eurosceptic votes from the Tories. That being said, I'm not sure how we proceed here. I certainly don't approve of accusations of Wikipedia Fraud and damage from some users. There are plenty of minor parties that could have a deciding factor on the next election. If we want to add them to the table, that is fine. I'd prefer if they had representation in the Commons before we do it, but that's my personal opinion on the matter. However, I reject the accusation that Wikipedia is pro- or anti- any particular political party. That type of nonsense should not be tolerated here. Bkissin (talk) 22:12, 6 May 2012 (UTC)

I am sorry, but people will start thinking that Wikipedia is Anti-UKIP, I don't think Wikipedia is bias or Anti- or Pro- anything. Again I will stop adding comments to the talk page if you can give me the period of time of the process in deciding to add a UK Independence Party table.(97daviee (talk) 10:27, 7 May 2012 (UTC))


 * I don't think most people will think that. Unless those, of course, which are strongly pro-UKIP. And that is mostly because they don't have representation in the House of Commons, despite how high opinion polls may place them. It's common sense, and something which is done for all countries' elections: if a party doesn't have representation, it's not considered important enough to be added alongside a country main parties in a table; mostly to avoid it being cluttered and full of minor parties, something which clearly wouldn't benefit clarity. If in the next general election it wins at least 1 seat, but scores high in the popular vote, I'd support its inclusion. At the moment, though, it doesn't have an high enough relevance to influence the country politics. Impru20 (talk) 13:56, 7 May 2012 (UTC)
 * I agree wholeheartedly. The only people who are going to suggest Wikipedia are anti-UKIP are those who are fervently pro-UKIP. Anything other than the recognition they deserve is bias against them in their eyes. If we are adding minor parties to the mix, why not add RESPECT, as they just won a Labour safe seat and have representation.
 * Anyway, if an MP crosses the floor to UKIP, and others follow him, then we can reconsider this topic, but for now, let's let sleeping dogs lie, shall we? Bkissin (talk) 23:07, 7 May 2012 (UTC)

UKIP again overtook the Lib Dems in the opinion poll published yesterday. Will everybody stop mentioning seats, opinion polls don't have any thing to do with the number of seats, they are done using percentages, and UKIP have a higher percentage than the Lib Dems. It's very scary how 25% of the people on this talk page who are opposed to a UKIP table don't understand how percantages work. Pleace can we add a UK Independence Party table.(97daviee (talk) 16:09, 15 May 2012 (UTC))

Tonight at 10:00PM YouGov Published their Opinion Poll, it showed UKIP again ahead of the Lib Dems, you can no longer avoid not to add a UK Indpendence Party table.(97daviee (talk) 21:31, 15 May 2012 (UTC))

This is the first time I have seen this article. I am shocked that UKIP does not have it's own table, the two latest opinion polls show UKIP in front of the Lib Dems. I normally find Wikipedia a reliable source of information, however, this article is shaping British politics as if there is never going to be a political change.(Politico100 (talk) 17:43, 16 May 2012 (UTC))


 * Past March, Spanish opinion polls predicted a landslide victory of the current governing party, the PP, in the autonomous community of Andalusia. In the end, it ended winning by a rather small margin of difference, and far from being able to rule the region. In 1992, UK opinion polls predicted a landslide Labour victory; in the end, the Tories won. You know, you can't trust 100% in opinion polls. They're just advisory and a good compliment for an election article, but they are not the official election results. Impru20 (talk) 18:15, 17 May 2012 (UTC)
 * I'm not sure I get your point. Regardless of what the results end up being, the polling is part of the story of the election, and UKIP polling at higher than normal levels is a part of that story too. People should be able to see that the party is polling higher than the Lib Dems, as stated, but more important is that if people could actually see the real polling results, they could make more incisive observations about the state of play: for instance, including UKIP might allow one to better assess whether the right-wing vote is splintering and think about what that means in the broader European context given the first round of the French presidential election.


 * The point made earlier about MPs needing to join UKIP before adding the party doesn't make a whole lot of sense; there is no specific benefit to limiting coverage to parliamentary parties. Any party with significant polling, even just a point or so, should be reflected as is done at articles for elections virtually everywhere in the world. It is especially important where, as here, you have a substantial amount of support listed as "Other", and most of that support is going to one party.


 * Ultimately, the point of including polls is not to make predictions about who will win the election and which parties will win seats. It is about giving people information about the attitudes of the British voter using a crude but understandable metric. If voters are telling us they are more inclined to vote for UKIP than they have been before, why shouldn't we state it? Wikipedia, like any encyclopedia, is about sharing knowledge. What is gained by refusing to share this knowledge? -Rrius (talk) 20:24, 17 May 2012 (UTC)


 * My point is, that even if an opinion poll says tat a party will end up getting 9-10% of the vote, it may end up getting just 1%. The point made about MPs needing to join UKIP before adding the party does in fact make a whole lot of sense: just guess we end up adding UKIP to the table as they currently are, with no representation. People would then come here complaining about why were they added while at the same time leaving parties with representation out. In UK politics, it matters not how much percentage of the vote you receive, but how many seats you win. Having seats gives a party much more relevance in the political stage than one that hasn't any, and that's what people will claim: Why UKIP was added and why the Green Party wasn't, for example. And we can't just add all of them. Should there be any people wanting to see all of the data for the opinion poll, just check the link for it. That's what it's meant for. The table is meant to provide a quick view of the political situation of the main parties, not to provide a sum up of the % evolution of all parties.


 * Also, about the splintering of the right. You know, I don't know how things in the UK are when it comes to the making of opinion polls. However, here in Spain some people tend to 'lie' in the surveys and refuse to show their support for a party when things are going bad, when in fact they will still end up voting for them anyway. That's what is called the "hidden" vote, and it's an usual phenomenon when elections are still far away in time. Surveys usually try to guess the amount of hidden vote for each party, but it raises the chance for errors appearing. If in the UK that's the same, there will be surely many people from the right who don't dare to show their support for the Tories in the survey and, instead, say they'd rather vote for UKIP. Elections are still a long way off from happening. Once the date gets close, many right-wing voters may return to the Tories' fold once again, making any gains made by UKIP during all these years pointless. Impru20 (talk) 21:17, 17 May 2012 (UTC)


 * In order to give balance views we have to add a "UK Independence Party" table. There is no better estimate than opinion polls, so you just have to follow what the opinion polls are saying.(97daviee (talk) 20:26, 17 May 2012 (UTC)


 * UKIP for first time ever is on a 2 point lead over the Lib Dems, so it's about time we add a UK Independence Party table now.(97daviee (talk) 21:07, 17 May 2012 (UTC))


 * There is a better estimate than opinion polls, and that is oficial election results. Do you want to see UKIP's performance in opinion polls? Just check the link. Give me a more reasonable reason than "it's what the opinion polls say" and I may be convinced of it. Opinion polls may fail. Read my above comment. Impru20 (talk) 21:17, 17 May 2012 (UTC)


 * UKIP Britain's fastest growing politcal party, which achieved 14 per cent of the vote in the last local elections. UKIP is currently in a two point lead over the Lib Dems. Opinion polls aren't used as predictions for the next election, they are used as a way to measure how the British people's view about politics in the present time. You have to add a UKIP table because UKIP is currently more popular than the Lib Dems, I am suprised that some people aren't here to demand the removal of the Lib Dem table, and to add a UKIP table as it's replacement. Most of the people on this talk page are in favour of the introduction of a UKIP table. Every TV station in the UK has to give out balanced coverage of politics, and what makes Wikipedia justified to not give out unbalanced coverage. You should have added UKIP table months ago. So please on behalf of the people of this talk page add a UKIP table, and don't continue to make up arrogant excuses.(Politico100 (talk) 12:16, 18 May 2012 (UTC))


 * You're going too far here. You know, this article is a compliation of all opinion polls for the 2010-2015 period. Even if UKIP is experiencing a boost in support now in 2012, it was an insignificant party in 2010, when Lib Dem in opinion polls was around 20%. UKIP current boost could be temporary, and we can't just replace an historically important party as Lib Dem from the table just because UKIP is experiencing what could be a temporary rise in opinion polls. Just think what'd happen if UKIP gets down again and the Lib Dem rises in 2013 or 2014, before the elections; replacing them in the table would have been pointless.


 * Also, this article includes polls just for the general election, so, using the local elections as an argument for adding UKIP to this article is pointless, as these are two different kinds of elections. In Spain, regional and other minor parties tend to score much higher in local elections than in general elections; I suppose the same can happen in the UK. Impru20 (talk) 13:29, 18 May 2012 (UTC)


 * Completely agree that UKIP should have their own table. The point regarding 'historically important parties' (despite that this itself is subjective) is entirely moot. The purpose of this article is to reflect opinion polling, not some ostensible historical merit. A party doing worse in the opinion polls should simply not have a table while another one doing better does not, especially when the percentage being excluded as a result is so significant. 過労死 (talk) 12:30, 20 May 2012 (UTC)


 * Again, while the Lib Dems are doing bad now, they weren't so bad in 2010 or 2011. In fact, they have been surpassed by UKIP just recently, and sometimes UKIP still falls below the Lib Dems. I may agree to add UKIP if it continues to do so well, but never to replace the Lib Dems by them (we can't have Lib Dem on some tables while leaving them out in others. It's one way or the other). Impru20 (talk) 16:05, 20 May 2012 (UTC)


 * Doesn't that logic then show bias towards a specific party, ie. the Lib Dems? Isn't Wikipedia supposed to be about pure information and not bias?  If the information show that UKIP is rated higher in opinion polls (what this article is about), it should be either included in the table or replace the Lib Dems with the UKIP.  For consistency then, other similar such tables should also be updated and not be shied away from in the interest of either bein bias or lazy.  Thank you. The joyous one (talk) 20:59, 20 May 2012 (UTC)


 * No. Lib Dem and UKIP in 2010 opinion polls rated 15-20% and 2%, respectively. In 2011, and even in early 2012, it was 10% and 4-6%. It's not bias towards the Lib Dems; just because they are rating bad now, doesn't justify for removing them completely, which is what you seem to suggest. Pure information, as you point out, states that for the past 50 years, the Liberals/Lib Dems have been the third political force in the UK. How do you pretend to simply replace them for a party which has no relevance at all (as of yet, at least) at the political level? And tell me: should we include UKIP, why shouldn't we include other, more relevant, parties? People would start complaining about it if we add UKIP with no serious justification.


 * By the way, I'm getting tired of this attitude of many people here, who think they can come here giving orders and saying you must do this because I say it.' You ask for consistency, but your arguments aren't consistent. If you force me to choose between 50 years of election results or 2 months of opinion polls, it's obvious what my decision is going to be. Impru20 (talk) 21:34, 20 May 2012 (UTC)


 * And who made you supreme dictator? Again, this article isn't about election data, it's about *opinion polling* data.  The page should reflect opinion polls, it has nothing to do with past election results, there's a specific page for *that*.  — Preceding unsigned comment added by 89.101.92.255 (talk) 09:56, 23 May 2012 (UTC)


 * I'm not supreme dictator of anything. I'm just putting enough arguments for not adding UKIP, while the only argument you seem to put forward is simply "put UKIP in because opinion polls say it". There are other parties in the opinion polls, as you can see in the last You Gov opinion poll for 22 May: http://cdn.yougov.com/cumulus_uploads/document/rv1tune9md/YG-Archives-Pol-Sun-results-220512.pdf


 * Headline Voting intention
 * Con: 32
 * Lab: 43
 * Lib Dem: 8
 * Other: 17


 * Other Parties Voting intention
 * UKIP: 9
 * SNP/PCY: 3
 * Green: 3
 * BNP: 1
 * Respect: 0
 * Other: 1


 * Thus, even opinion polls put UKIP within the 'Others' category. But that is something you want to simply skip. Where's the line of which parties should be in the table and which shouldn't? You don't seem to have the intention of drawing one, but I do. You're right, this table isn't about election results. Neither did I say that. I just said that election results should be used as reference. For example, if a party ended first in the last general election, it should be shown first in the table; if a party ended second, it should be shown second; and so on, regardless of their current opinion polling. We all know that Labour has long surpassed the Tories as the first party in opinion polls, yet they are still shown second and nobody complains about it. However, everybody complaining about the matter want to replace the Lib Dems by UKIP. It's nonsense.


 * If the Lib Dems are currently more relevant than UKIP despite them polling similar in opinion polls it's because several things: firstly, the Lib Dems have seats in the Commons while UKIP hasn't. Secondly, the Lib Dems have been the third political force in the UK for the past 50 years, either as Liberal Democrats or as the Liberals or whatever name they beared, and that's the place they scored in the last general election. And thirdly, the Lib Dems are currently part of the UK government. That's why the Lib Dems must have priority in the table regardless of all of you wanting to replace them by an (as of yet) irrelevant party. Once UKIP scores high in a general election, they'll be added to the table. It's general elections the ones which have the capacity of changing the fate of a country, not opinion polls; thus, it's foolish to simply put them aside. Impru20 (talk) 13:54, 23 May 2012 (UTC)

I would argue that UKIP should be included on the table because they have been consistently polling equal to, higher than, or within 2% of the Lib Dems since January now. To me that is a fair justification to include UKIP on the table as their support is almost the same as or greater than the Lib Dems. Jackyb92 (talk) 17:20, 21 May 2012 (GMT)
 * Another poll had UKIP third again last night (http://cdn.yougov.com/cumulus_uploads/document/okc1i8gnlc/YG-Archives-Pol-Sun-results-210512.pdf), it is getting rather ridiculous that they aren't included on the page outside of 'others'. I would also argue that if UKIP class as 'others' then so should the Lib Dems as they are polling at similar levels. Jackyb92 (talk) 10:50, 22 May 2012 (GMT)


 * Seats and the past have no relevance with opinion polls, it's about the people's political opinion and the present. UKIP over this month has overtaken the Lib Dems many times, when UKIP haven't they've never been below 1% to the Lib Dems. People how are opposed to a UKIP table are now just simply being callous and rude.(97daviee (talk) 17:46, 25 May 2012 (UTC))


 * I'm ending the discussion here. Not only did you need to create a second account to, supposedly, "justify" your statements, but also you dare to come here giving orders such as "UKIP must be added" and insults like "otherwise, Wikipedia is anti-UKIP" and the such. Now, after being unblocked, you still dare to come here and still acting the same. I've explained carefully why UKIP shouldn't be in the table, not one, but multiple times. Care to read some of them before posting the same over and over again with no accurate arguments, and before coming here to call people "callous" and "rude", when the only one who came here imposing his will from the beginning was you. Impru20 (talk) 20:32, 25 May 2012 (UTC)


 * Impru20, your account states that you're a liberal, I know now why you don't want a UKIP table to be added, that's because you are terrified that if the British electorate saw clearly UKIP ahead of the Lib Dems that you're liberal dream would end in Britain. The point of a talk page is to exchange views, if you can't stand other peoples views then that just sums up that you are only against a UKIP table due to the fact that you are personally against UKIP. I've read you're comments twice actually, and I just believe that it's based on your opinion not facts. It's a fact that UKIP is ahead of the Lib Dems in the polls, it's a fact that the majority of the people on this talk page support the introduction of a UKIP table, so that's why I support a UK Independence Party table.(97daviee (talk) 21:58, 25 May 2012 (UTC))
 * First of all 97daviee, a talk page is NOT the area to be discussing political issues. It is not a discussion forum on why you like or dislike a particular political party. It is to discuss changes to the article at hand. Second of all, your use of sock-puppetry on this site is grounds for you being blocked. Third of all, there is NOT a majority in favour of changing the layout and adding UKIP, as I am against it as well, and the other sources in favour are accounts YOU created. Next, name-calling, finger-pointing and dictating orders to other editors, despite the fact that you are not an administrator on this site, is harassment. Additionally, calling into question one's political affiliation is out of line in this conversation, as 1. Impru20 does not live in the United Kingdom, and therefore has no vested interest in British politics 2. None of us have called your political affiliations into question, attempting to converse with you civilly and 3. Impru20 is merely trying to neutrally look out for the best interests of the article. Decisions are made by consensus here, and your edicts to other editors is far from that. Bkissin (talk) 02:34, 26 May 2012 (UTC)


 * That's what I said exchanging views on how this article should be changed. The only people on this talk page who are against a UKIP table is you & Impru20. I have not been "name calling, finger-pointing and dictating orders to other editors", it just seems that whatever you or Impru20 says it seems to happen, and if anybody on the talk page who is in support of changing this article in someway which you or Impru20 don't support you just seem to act in an autocratic way. UKIP is ahead of the Lib Dems yet they haven't had their own table added, that's obviously not neutral. Then why not agree with the majority of the people on the talk page in adding a UKIP table.(97daviee (talk) 07:59, 26 May 2012 (UTC))


 * Daviee, please remember to assume good faith. Accusing other editors of having a political agenda is not good for the community: we just want what's best for the article. Given your recent sockpuppetry, it would be wise to be more diplomatic in this discussion and not accuse others of having an agenda (which could be construed as a personal attack. Finally, as Bkissin has stated, there is not a majority in favour adding UKIP yet, and, on the contrary, it seems that the consensus is indeed against a change. It is not just Bkissin and Impru against it, I am also against it, and scrolling through this page seems to demonstrate that several other editors are, also. –  Richard  BB  12:09, 26 May 2012 (UTC)


 * 86.5.171.186, C. 22468, Spa-Franks, Sir Richardson, Spiritofsussex, Rrius, 過労死, Jackyb92, The joyous one and 89.101.92.255 all agree that we need to add a UKIP table. You must be mad to think that's a personal attack. It's shame that we have a minority of people against a UKIP table, because they are damaging Wikipedia from the inside.(97daviee (talk) 13:43, 26 May 2012 (UTC))


 * As a heads-up I'd like to say that I've seen this kind of POV-pushing before by UKIP in very many forums and places on the internet and even several times before on Wikipedia. The way they work it is as follows: usually more than half of those who 'support' whichever particular thing UKIP are currently trying to push are unregisterd users and/or anonymous IP addresses. These - and some of their more persistent registered users - will ignore what has been said/wikipedia policy/common sense/you name it and go on and on and on (and on!) until ordinary folk are just plain old sick and tired of hearing them. Then they'll 'get their way'. That - at least - is how it usually goes.... OldSquiffyBat (talk) 08:16, 4 June 2012 (UTC)


 * As I've been mentioned directly in the previous post but one, I'd just like to use my right of response. Unlike what was suggested when my name was mentioned, I do not believe I have ever said there should be a UKIP column - I believe what I indicated my position to be was that if UKIP continued to regularly and continually outrate the Liberal Democrats for a consistent period of time, then as a community we should consider adding a column, and I also suggested how this perhaps could be achieved without producing an overly cluttered table, namely in the form of collapisble cells. For what it is worth, I personally do not believe that this has happened - yes, there are occasions when UKIP are outrating the Liberal Democrats, but it is not happening regularly, and it is not a pattern being replicated across every polling organisation.


 * However, this is now, for me, irrelevant. I have not updated or edited this article for some time, and I do not feel any desire to start doing so again. I first came across the opinon polling article for the previous election in quite a poor, outdated state in 2008, and I decided to update it, and continued to do so. In all that time, although there have been some disagreements, I have found all the editors were able to get along well, and work for the benefit of the article - i.e. just what I believe Wikipedia is all about. More recently, however, especially surrounding this UKIP issue, there has been an increasing sense of certain users casting aspersions on the integrity of other users, and almost suggesting they are trying to use wikipedia for political purposes. I am happy to publically state my political allegiance, and it is displayed on my user page - I do not believe this should leave me open to the aforementioned aspersions, especially when all the edits I have made to this page, and others, are always made in an attempt to improve the individual articles for the readers of those articles. I am not a member of wikipedia to get into squabbles and arguments, and I believe that the vast, vast majority of members feel the same; but this page has now become nothing more than a slanging match, and I do not intend to be part of it any longer. I therefore do not intend to edit this page any further; I hope others feel able to, but I have quite frankly lost all motivations to do so because of these constant arguments and aspersions. I hope that this issue surrounding UKIP can be resolved in the near future, and that the page can return to the co-operative, convivial atmosphere that existed before, and at that time, I hope to be able to rejoin the editors, all of whom I have a great respect for because of the hard work they all continue to put in.Spiritofsussex (talk) 11:06, 4 June 2012 (UTC)
 * Hear Hear Spiritofsussex! What we are attempting to build here should be a collaborative effort, in which all users treat each other with respect. Accusing each other of trying to use WP for political gain is a black mark on ourselves, and on what we are trying to create. Bkissin (talk) 21:36, 4 June 2012 (UTC)


 * I was utterly disgusted today, that UKIP is 2 per cent ahead of the Lib Dems, but UKIP doesn't even have it's own table. Shame on Wikipedia, you've devalued the trust everybody has with Wikipedia on political articles. Let's add a UKIP table before this damages Wikipedia's reputation further.(Politcalforlife (talk) 10:38, 10 June 2012 (UTC))


 * It's fascinating to see UKIP this high, by adding a UKIP table you will be doing what the point of encyclopedia is for, and that's giving detailed information. By not adding a UKIP table, you're shaping British politics for the benefit of the Lib Dems. An encyclopedia has to be broad on politics, but unless of course Wikipedia is officially declaring it's political affiliation liberal which I doubt it will. I believe Wikipedia has to include every party fairly, and that includes UKIP.(Politcalforlife (talk) 21:17, 11 June 2012 (UTC))

NOTE: ''Politcalforlife has since been confirmed as yet another sockpuppet of 97daviee. I would urge all users on this talk page not to mistake this rabid sockpuppetry for multiple users voicing a consensus. I would also urge all users to keep an eye out for other potential sockpuppets pushing a pro-UKIP bias.'' –  Richard  BB  11:13, 12 June 2012 (UTC)
 * Thank you User:Richard BB! I reported his username to this OTHER sockpuppet investigation! Bkissin (talk) 19:09, 12 June 2012 (UTC)


 * Unlike 97daviee, I am happy to wait until a consensus is reached. However, I think it should go back all the way to the election, but there aren't many "UKIP" results for the 2010 polls. Spa-Franks (talk) 21:26, 13 June 2012 (UTC)
 * To be honest, Spa-Franks there is perhaps a compelling argument for adding them. I don't know if it's enough, but it's something. What I dislike is being ordered to change an article without consensus and being accused of political interference when that demand is not met. Thank you for at least being reasonable in all this. Bkissin (talk) 01:34, 17 June 2012 (UTC)

New UKIP debate: November

 * UKIP is now consistently polling at or around the same as the Lib Dems, either +/-2%. They should now be added as there is now a possibility they could become the 3rd party in terms over the overall vote.--CH7i5 (talk) 17:49, 25 November 2012 (UTC)


 * As it has now been over seven months since this topic was brought up, and UKIP's popularity has remained as high as the Lib Dem's, I wouldn't oppose them being added to this page. –  Richard  BB  17:56, 25 November 2012 (UTC)


 * If we were to add them, how far back should we show their figures? In May 2010 they were only getting 2%. It was only from early to mid 2012 that they started to get as high as support as the Lib Dem's. I would suggest just starting from 2012, or possibly from now. CH7i5 (talk) 21:51, 25 November 2012 (UTC)


 * Go as far as the polls allow. Slowly, people will check the polls. If we each did 2 polls a day, itd be done in like 2 weeks tops.--Metallurgist (talk) 08:07, 26 November 2012 (UTC)


 * What design of table should we use? I would support this one suggested by Impru20


 * Its not to large and fits everything that needs to be there in.CH7i5 (talk) 13:15, 26 November 2012 (UTC)
 * Looks perfect to me! –  Richard  BB  13:16, 26 November 2012 (UTC)


 * So should we begin slowly adding them in? CH7i5 (talk) 13:27, 26 November 2012 (UTC)


 * In my opinion, yes. However, I'm not sure if you want to wait for more editors to voice their opinion. The three of us may be in favour, but there might be others who had disagreements (that said, I won't object if you start adding UKIP's data now). –  Richard  BB  14:17, 26 November 2012 (UTC)


 * No they ought not to be added. As per the discussion(s) above. And whoever has added them - without consent, and against previous consent - please have the courtesy to remove them before agreement has been reached to the contrary. Sianska79 (talk) 21:41, 26 November 2012 (UTC)


 * It was my understanding that they shouldn't have been added before because we thought their popularity was going to be short-lived (I was one of the people saying it shouldn't be added). However, their popularity has persisted to today. –  Richard  BB  21:48, 26 November 2012 (UTC)


 * They have been polling for at or around the same as the LibDems for several months now, to long for it to just be a 'short lived boom'. It wound be biased to not include them but at the same time include the LibDems, who have several times polled a lower percentage than UKIP. CH7i5 (talk) 17:01, 27 November 2012 (UTC)


 * There are several reasons for not including them. Short-lived popularity may have been one and as you both say, that has proven to be not so. However, a quick look at the make-up of British (sorry, UK) government shows that other parties have parliamentary representation: the Greens, SNP, Plaid Cymru and some parties from Ireland. Whilst it may be approrpriate - due to regionality - to not include the scottish, irish and welsh parties, excluding the Greens (who have an MP) whilst including UKIP (who have not) is making a selective and subjective judgement. Greens must be included if UKIP are included.


 * On another point, UK parliament elections are not run on any kind of PR system. So the % a party may have in polls does not warrant their inclusion here. If (as is the case in Sweden) 4%+ meant a seat, then I could understand why UKIP must be included. But 4%, 8%, 12%... none of that makes any difference in the UK. It is a constituency based system. The % a party has in polls is not grounds for their inclusion - only if they are/or are likely/or have representation ought they to be included.


 * Finally, once consensus has been established - as it was here, I think - it ought not be challenged every few weeks or months. Can imagine the mess of Wikipedia if that happened all the time on every article. Conclusion: as per previous discussions, and previous consensus, they ought not be included. Sianska79 (talk) 10:26, 28 November 2012 (UTC)


 * I think you make a very valid point concerning the inclusion of other parties, such as the Greens or Plaid Cymru. Consequently, I think I'll abstain from further support of UKIP inclusion. –  Richard  BB  10:44, 28 November 2012 (UTC)


 * But this is not an election estimate, this page is for opinion polls, which is totally different, it does not guarrrente that the party with the highest percentage will win. Opinion polls represent which way people say they will vote not the actual outcome of the elections, so saying that just because UKIp doesn't have an MP is invalid, this page is not for that. Also I have no objection to adding the greens, although they are only polling around 1-2% 81.136.154.148 (talk) 14:20, 28 November 2012 (UTC)

I think it's best to keep UKIP as they are often being referred to as Britain's third party now, so, from that perspective, keep them. I personally think they should be added from 2010 or an explanation should be added, as I can imagine someone coming here before the election and going "UKIP? Where did they come from?". Spa-Franks (talk) 22:41, 29 November 2012 (UTC)

I have added a brief conscise explanation as you suggested. I think it's best as it is on the bassis that we have to work with evidence. The evidence we do have is that UKIP and the Lib Dems have been in a statistical tie for over 9months. I think it's important to have a ittle buid up to that, which is why the beginning of 2012 seems a sensible place to tart with that. Also it makes it a hell of a lot easier in terms of table formatting, we don't really want to split 2012 up in to 2 tables. I think this solution works, you guys have done a good job! — Preceding unsigned comment added by 217.41.32.3 (talk) 05:55, 1 December 2012 (UTC)


 * This is a little strange. Against consensus the UKIP figures were added to this article. And not yet removed despite being asked to remove them. Until consensus says "Hey, include them", their figures must be removed from the article. And there are good grounds as to why they cannot be included: 1. no-one has offered any argument yet as to how or why a party without represntation deserves to be listed (UKIP) whilst parties with representation (Greens, SNP and others) are not listed, 2. UK government works on a First Past Post system - so % are not relevant to what makes a "3rd party" - UK "3rd party" is clearly Liberal Democrats as they have very many seats and are in government. How can anyone argue that UKIP are "3rd party".


 * Anyway, the correct procedure - which is not being followed here - is to remove disputed material (which was added against consensus) and then to argue. Not to argue with that disputed material in place. So, I'd like to ask the user who added UKIP to the tables, again, to please now remove them. Then maybe arguments can be made either way. Thank you!! Sianska79 (talk) 15:10, 2 December 2012 (UTC)

UKIP's polling is way too high to ignore by lumping it in with "Others". Not having it distorts the information being presented. There is really no reason to include the Greens since they seem to be consistently polling at 2%. What exactly is to be gained by excluding a party that has consistently had polling figures above nominal levels for some time now? It would be one thing if the combined minor parties were taking five to six points of if there were 7 to 10 parties getting two points each, but neither is the case. Looking at the last YouGov, of a total of 17% for what we would lump together as "Other", UKIP had 11 against 3% for the SNP and Plaid, 2% for the Greens, and 1% each for the BNP and what they call "other".


 * As a reader, I came here today hoping the UKIP numbers were there, specifically because I wanted to check how if the Tory rebellion or by-election had affected their position. Why should I be denied the ability to do so in future? Including another party with significant polling numbers does not prejudge the election. All this table is meant to do is tell the reader the opinion polling results, and not just those for the three parties with double-digit numbers of seats. What lumping UKIP and the Greens into "Other" does is tell the reader that there is nothing interesting there to report. But we are talking about a significant share of the vote, and ignoring that is hard to understand.


 * Restricting our presentation to the parties that did the best at the last election seriously misunderstands the point of listing the polling for the next one. This page is not meant as merely a progress report on the Conservatives, the Lib Dems and Labour. It is a list of polling conducted in anticipation of the next election. One can choose to view it as a the former just as one can use it for any number of other purposes, but it has to be kept in mind that the latter is what it is really for. So not showing something major that is coming out of the polling is a failure to do what we are supposed to be doing here.


 * The argument made about restricting to parliamentary parties is frivolous. I refer to comments in an earlier discussion UKIP could only be added if an MP joined it. How silly. What if UKIP passed the Conservatives, taking second position? Would we still ignore them because they have no seats? If that is truly going to be a rule, if we are going to ignore sustained opinion polling results for parties for the next election because they didn't win seats at the last one, then this article should really be moved to Opinion polling for major United Kingdom political parties, 2010–2015 because it clearly has nothing to do with the next election. -Rrius (talk) 15:56, 2 December 2012 (UTC)


 * Sianska79, 81.136.154.148 did offer arguments you asked for above, which you obviously did not read. Besides based on the consensus from this debate you are the only one objecting to keeping UKIP in. As this is a new debate previous consensus should not be taken into account, as that is not they way it works. So there is no need to remove UKIP as it is going with the current consensus. YOU are the only person objecting

For CH7i5 Metallurgist 81.136.154.148 217.41.32.3 Rrius Spa-Franks Against Sianska79 Abstentions Richard BB (original vote for) CH7i5 (talk) 17:31, 2 December 2012 (UTC)

Also the newest shows that UKIP is 4 points ahead of the LibDems, to just shove them into the 'others' section would not be an accurite display of the facts. Opinion polls are different from the election itself, they show what the public OPINION is (surprise, surprise -sarcasm-). To say something along the lines of 'oh, but they might not get any seats' is irreverent! The libdems might not get even 10 seats, they might possibly not get any seats either, the way public opinion is. CH7i5 (talk) 18:11, 2 December 2012 (UTC)

I have to say, it seems a very strange idea to consider removing them no matter which we you look at it: UKIP and the Lib Dems do seem to be in a statistical tie (within the margin of error) and have been for 9months. I can get why people resisted at first (and agree with their resistance) but we have clearly reached a point where we can see that in polling terms (the only terms that matter here), this is clearly not a flash in the pan! I mean to simply shove UKIP in others no longer seems credible to me, for one thing they don't fit! Dunno if you noticed but there have been a number of occasions where the others column has been more than double the Lib Dems column! When one party is having that much effect for such a long and sustained period, we can no longer call it an anomoly! I just want to also point out that all the polls now include UKIP within their headline poll and not in others anymore. To switch back to not having UKIP with it's own column would be to go against what officially registered pollsters are doing! We'd be breaking the precedent, which would make wiki look biased, surely! I can see absoutely no way that the reversal of this decision can be justified! I mean it's not like you can pull the argument "Oh but they not get any seats".... a poll is a poll, you can't sit on a poll (it'd rather hurt, if you'll pardon the pun)! the point here is consistency, 9 months of a statistical tie is a pretty lear pattern that shouldn't be ignored. I whole heartedly object to the removal of UKIP column, in fear that wiki will look as if it has it's own agenda! sheffno1gunner (talk) 02:17, 3 December 2012 (UTC)

There are some on the Next UK election page talking about this page and seem to be making an implied threat to remove UKIPs column. I would suggest they are being daft and that this would be a breach of wiki rules because it's not interpretation, it's basic reporting of the numbers. In making the decision to get rid of UKIPs column, we are actively making the decision to ignore how pollsters report their polls, in my view this would be original research because it requires self interpretation and adjustment of figures! I think someone said on this page that "There are no seats in polls, they're just national percentages, you can't sit on a poll". To my mind it would be original research to remove UKIP (not the other way round) because it would be ignoring the reporting and methodology of officially registered polls. We can't do that.... a poll is a poll, not based on seats, its a national %. E.g. the Lib Dems got 23% of the vote but no where near 23% of seats in 2010! Sorry but we're just reporting the polls as they're being reported. Interpreting or changing them would be original research. This seems to be what your recommending! I suggest you don't! As far as I'm concerned I'm on the side of keeping the UKIP column, no concensus for removal. Nick  ' 09:48, 3 December 2012 (UTC)

I have taken the liberty of spending a whole hour going through all of the sources for polls where UKIP's figure has been part of "others" figure in the headline poles, in all but a few their figure is several pages into each source. When I wrote UKIp's figure in their column, I ofcourse subtracted that figure from "others column". This should not be a problem in the future as UKIP are now included in all pollsters headline polls because of the statistical tie! Hope this helps! Nick  ' 16:17, 3 December 2012 (UTC)

New UKIP debate: Idea
Saw the new UKIP debate just by chance and wanted to post my opinion about it. So, back in May-June I was, at first, against including UKIP in the table, because its 3rd place wasn't consistent then. Now I see that UKIP has grown to a size in opinion polls that can't just be ignored. I've read most of the debate here, and came to the conclusion that, while there seems to be consensus about the ever-growing presence of UKIP in the polls, there isn't one when it comes to reach a reasonable justification about UKIP's addition. As of lately, I've been quite busy in the creation and maintenance of Spanish opinion polling's articles (which I've done mostly alone), and by the way, I've encountered some issues when deciding which parties should be in the table, and which shouldn't. Ok, I'll explain the example here. For those who just want to skip directly to the idea, just skip the text in italics below: the idea is in bold text.

''So, the issue came when doing the opinion polling for the 2011 Spanish general election. Parties are usually put in the table in the place they score in the previous election, but in that case, I had to make an exception: specifically, for PNV and UPyD. PNV is a mostly boring party when it comes to opinion polling in a national level (it almost always gets between 1.0% and 1.8% in opinion polls, and that when it does appear in the surveys). However, UPyD is another issue: they've grown from nowhere to become the 4th political force in Spain in just 4 years, so its relevance is quite clear. The issue was that, in the 2008 election, PNV scored 5th while UPyD scored 6th. However, right from then and from the very first survey UPyD would surpass PNV in all opinion polls (with modest percentages at first, later growing to respectable figures through 2008-2011). Thus, I wanted to find a way to include UPyD and ignoring PNV while at the same time not looking extremelly weird. If you take a look at the article (here), in the graph note you will see how I solved this.''

'The idea is to set a % limit at which parties can be considered important'' enough to be added to the table. If a party polls consistently above a specified % threshold in opinion polls, it should be considered relevant enough to be in, regardless of the number of MPs it possesses.''' In the case of Spain, I've set it at 2.0%, which is a percentage that very few parties have ever managed to surpass there (and the ones which have done so have usually won enough parliamentary representation to become key figures in all Spanish parliaments until now). For the UK, however, this threshold could be risen to 3% or 5%; this would let UKIP in while leaving out other minor parties at the same time (Green, BNP, SF, PC...), regardless of parliamentary representation. What do you think?

As a side note, the example I've given, UPyD, has grown from a mere 1.9% in 2008 to the area of 8-10% in 2012 opinion polls (see here), benefiting mostly from PP's electoral bleeding (for those of you who don't know, PP is the main conservative and centre-right party of Spain, as well as the current ruling party), which is down to 30-33% from the 44.5% they got in the 2011 general election. It looks something analogous to what is happening in the UK between the Tories and UKIP. Thus, if I had to decide right now, I'd be in favour of including UKIP in the table, as long as it's done with coherency: that is, if UKIP is to be added, all of its data up from the 2010 election should be included, just as it's done with the other three parties, not just the data for the last year. Impru20 (talk) 18:48, 5 December 2012 (UTC)

I don't think there can be any doubt about it. Our job is to report reality. Most official opinion polster have been including UKIP in their headline poll, so not to include them would involve deliberately subtracting UKIPs figure and then adding it to figures. That would be fiddling with the figures, we can't do that! Also the statistical tie is the next compeling argument, 9months is not a flash in the pan, three yea fair enough but not 9! As far as adding UKIP to the table from the beginning of the Parliament, I have my reservations. If it's all or nothing I'd vote to keep them in but really I don't think they had earnt it back then. But because we divide the tables into years it's not an issue! I don't see why it ha to be all or nothing, what we have now works. One thing I do think we need to do is look at plotting a new graph. I'd be perfectly hppy to have UKIP on a graph from the begining of the Parliament. I think it would be an illustrative justification for adding the party. So yea, I'd vote for one graph, I thik tat's the optimum solution: One graph and tables by the year, which is consitent with what we had before. I don't think we can completely compare Spanish Politics with British Politics, I get what your saying but I don't agree in this instance! sheffno1gunner (talk) 22:32, 5 December 2012 (UTC)

Yea, we don't need to do that, the table breaks up into years which makes this a non-issue, UKIP were not polling anywhere near enough in 2010 but since March 2012 things have changed. The statistical tie is the main argument for me. A poll is a poll, you can't sit on a poll! As far as the table goes, I like that idea! I think it's a comprimise between the 2. It means we only give UIP a table from when they became relevent and also it means we have 1 graph for the Parliament and yes it visually demonstrates why we have added UKIP. Yea, lets do that it makes sense, it satisfies both of your concerns and importantly it's consistent. Consensus? Nick  ' 22:48, 5 December 2012 (UTC)

That sounds good to me Nick! Can we move on to talking about the graph yet? — Preceding unsigned comment added by 81.133.12.45 (talk) 22:51, 5 December 2012 (UTC)

Further Evidence - BBC Poll Reporting
Notice that our new table format merely reflects the same format the BBC have used since 14/05/2012. This is further justification for the decision to include UKIP in it's own column, I think this argue also carries some weight with European Elections also. http://www.bbc.co.uk/news/uk-politics-18264385 — Preceding unsigned comment added by 81.133.12.45 (talk) 01:30, 10 January 2013 (UTC)

Add UKIP 2010/11 data? 3rd party lead in depth

 * I don't agree with the 3rd party lead box, either. I mean, there is no sense in having a lead column between the 3rd and the 4th parties. With the same arguments used here for having it in, it would make even more sense for a "2nd party lead" column, yet no one seems to have put that possibility forward. Moreover, the following is being organised in a wrong, weird (or both) manner:
 * 1. The 3rd place graph (the image) is pointless. All data shown in it is alredy accessible from the main graph. I think this is an innecesary duplicity.
 * 2. In contrast, the main graph shows data for UKIP down from 2010 BUT in the tables we only have the data down from 2012. I believe it'd make sense to have UKIP data in the table from the beginning, or else it would make little sense to show it all in the main graph.
 * The whole thing looks (to me, at least) like a weird attempt to highlight UKIP's recent poll numbers by having an excess of information in some parts of the article and a shortage of information in other parts. Anyway, I will accept whatever the majority decides. Just thought I should post my opinion because I was the one who introduced the (1st place) lead column to the table in the first place. Impru20 (talk) 21:33, 21 January 2013 (UTC)

I will gladly accept the argument of duplicity. My reason for adding it was not to highlight UKIP, it was because another editor had gone to so much trouble to make a graph and I wanted to find a use for it because I had made his graph redundant by going through all of 2010 and 2011 to get the data for UKIP, I mean it took me hours of opening every poll etc! We're now able to have 1 graph for the whole Parliament, it was partly on that basis that we have not split the article! So I decided to chop his graph in half and use it on UKIP's page because it was note worthy. However, I have come round to your point of view in that this graph is not relevant for this page. We are in effect double duplicating. 1) We have the same graph on more then 1 page and 2) we have a segment of a graph below the entirety of the main graph. This is not right, therefore I shall remove it. it is better placed in UKIP's article and also would make this article look more balanced.

As for the next point: I see no problem in including UKIP in the graph from 2010 but not in the tables until 2012. The reason being is that we can't simply suddenly take UKIP out of the "others" column half way through the graph, it would look very odd, especially when you consider the sudden collapse of the "others" on the graph. UKIP shouldn't be included for 2010 and 2011 because they did not become relevant in polling terms until April 2012, they are included from the beginning of 2012 for formatting reasons as previously agreed, also it helps to give a bit of perspective as opposed to them suddenly springing up from zero.

And finally: the 3rd party column is indeed useful and not at all confusing. Highlighting 2nd place would be irrelevant, notice in the explanation above the data, it clearly explains that the Conservatives and Labour are the ONLY 2 parites to have held either 1st or 2nd place throughout the Parliament, they alternate. The same is clear for The Liberal Democrats and UKIP. The note above makes everything clear. The reason why the 3rd party lead column is necessary is because the position of 3rd place keeps changing and this is of significance, all of our reliable sources seem to think so. Among our reliable sources there is just as much focus on 3rd place as there is 1st at the moment, if not more! You argument for 2nd places does not stand because there is generally over a 15% gap. 2nd and 3rd place have never, not once, alternated. No reliable sources have made a thing of this either!

I will remove the 2nd graph, I will also move the text to UKIP's page for the reasons outlined above! But I see no reason for any other change!Sheffno1gunner (talk) 01:06, 22 January 2013 (UTC)

That seems a reasonable solution and takes into account the valid concern of duplication. Let's consider this matter closed! Nick  ' 11:44, 22 December 2013 (UTC)

Ok, so the duplicity issue is settled in a manner in which everyone ends up satisfied. One point less to discuss. I don't know if I understand your 2nd point. You're saying you defend having UKIP in the graph from 2010, but not in the tables. Why? Wouldn't it make sense to have UKIP in the tables too? Not only to show their polling data from 2010 in something the readers can easily check (as it is now, readers have no way to guess where the 2010-2012 graph data for UKIP has been obtained unless they open every poll from 2010 and look for the data themselves, something which would make no sense if it's pretended to have UKIP as a relevant party), but also to enable having UKIP 2010 election result in the table. This would allow Wikipedia visitors to compare current UKIP polling numbers with actual results back in 2010, thus giving them a larger picture of British voting behavior's evolution in the past 3 years. Also, as it is now, the tables for each year seem to use different formats, when all of them should use the same (as a side note, I don't even understand why years are separated in different tables in the first place, but that's another issue). You now try to justify not having UKIP in the tables in the years 2010 and 2011 because "back then, they weren't relevant enough". Ok, so, what if in the near future they (UKIP) climb down back to lower vote %? Should they still be considered as relevant enough to be in the table, or would them be removed from it? Or what if it's the Lib Dems who climb down to those levels? Would them stop being considered relevant enough to be in the table? I think the tables should reflect some consistency in this sense. I believe we can't add parties randomly just in the periods they are considered "relevant enough", because it would look just like that, random. If at some point it is decided by majority for a party to be in the table, that party should be in for the entire 2010-2015 period. As far as I know, no other opinion polling article but this one shows a party "cut down in pieces". It is my thought.

As for the 3rd point: I'm not defending in any way to have a 2nd party lead column. I used it to explain why the reasons for having a 3rd part column are very redundant. Just an example: Let's say that Tory vote falls sharply in next polls, and UKIP climbs up as a result proportionally, having both of them in the area of 20%. Then, using the same arguments used to justify the 3rd party lead column, a 2nd lead column would also make sense, but in the event of that happening, we would have up to three lead columns. And I'd like to note that the event of having a 3rd party polling extremelly close to a 2nd party is not a flying pig: it already happened in 2010, with some surveys showing a three-way race between Lab/Con/LibDem. Still, only 1 lead column was used back then, because it didn't make sense to have more of them. Impru20 (talk) 17:13, 22 January 2013 (UTC)


 * Can we respond to comment in order please. It looked like Nick was responding to you, until I checked the times. I'm sure this was an honest mistake.


 * 1. I really don't see the problem in having UKIP in the graph all the way through but not in the table for the years where they are not relevant, to include them from the begining of 2012 gives a fair amount of perspective. It is not possible to have 1 single reader friendly graph to cover the whole parliament without including UKIP's line from the begining, it would screw up the other's line and look an absolute mess! That said, I think you do have a point about the actual election result in 2010, this would add to perspective. This is why I propose that we also put the result for the previous election at the top of the page. It seems to me that it's more helpful at the top of the page as well, it's not doing much down there: The 2013 table would look like:
 * 2013
 * {| class="wikitable sortable" style="text-align:center;font-size:95%;line-height:14px"

!! width="100px" | Date(s) conducted !!width="350px"|Polling organisation/client!!class="unsortable" width="65px" | Sample size!!class="unsortable" style="background:#0087DC" width="90px"| Cons !!class="unsortable" style="background:#DC241F" width="90px"| Lab !! class="unsortable" style="background:#FDBB30" width="90px"| Lib Dem !! class="unsortable" style="background:#70147A" width="90px"| UKIP !!class="unsortable" style="background:gray" width="90px"| Others ]]!! class="unsortable"| 1st party lead!! class="unsortable"| 3rd party lead
 * style="background:#D5D5D5"|6 May ||style="background:#D5D5D5"| General Election Results (GB only) ||style="background:#D5D5D5"| 29,691,380 || style="background:#B3E0FF" | 36.9% ||style="background:#D5D5D5"| 29.7% ||style="background:#D5D5D5"| 23.6% ||style="background:#D5D5D5"| 3.1% ||style="background:#D5D5D5"| 6.7% || style="background:#0087DC; color:white;"| 7.2% || style="background:#FDBB30; color:black;"| 20.5%
 * 20–21 Jan ||YouGov/The Sun || 1,675 || 33% || style="background:#F8C1BE"|42% || 10% || 10% || 4% || style="background:#DC241F; color:white;"| 9% || style="background:#FFFFFF; color:black;"| Tied
 * 17–18 Jan ||YouGov/Sunday Times || 1,912 || 33% || style="background:#F8C1BE"|42% || style="background:#FEE4AF"| 11% || 7%|| 7% || style="background:#DC241F; color:white;"| 9% || style="background:#FDBB30; color:black;"| 4%
 * }
 * This would definitely help give the reader a sense of perspective. They would emidiately be able to see the difference between the actual previous result and the polling figures. Which in a sense further justifies their inclusion for 2012 and 13.
 * 17–18 Jan ||YouGov/Sunday Times || 1,912 || 33% || style="background:#F8C1BE"|42% || style="background:#FEE4AF"| 11% || 7%|| 7% || style="background:#DC241F; color:white;"| 9% || style="background:#FDBB30; color:black;"| 4%
 * }
 * This would definitely help give the reader a sense of perspective. They would emidiately be able to see the difference between the actual previous result and the polling figures. Which in a sense further justifies their inclusion for 2012 and 13.
 * This would definitely help give the reader a sense of perspective. They would emidiately be able to see the difference between the actual previous result and the polling figures. Which in a sense further justifies their inclusion for 2012 and 13.


 * I agree that parties can't simply dip in and out of the tables and graph, that would be illogical. But if we're honest it was 7/8months before UKIP were added. UKIP's first polling tie was 5/6th April and their first lead 13/16th April from then on there was scarcely more than 2% difference either way and yet we did not add the column until November. We waited more than 7 months so that we were absolutely sure that this was not a mere anomaly. Also, if you remember back to the discussions for adding the column, that was already discussed! If UKIP support fell away back to an insignificant level for the same period that they had been neck and neck with the Lib Dems then that would trigger reopening the discussion. The media, including our reliable sources do not accept this as a likely possibility, they point to the 2014 European Elections that has a spillover effect to the general elections. The odds of us having to discuss removing the column are really very low indeed! Also as far as adding other columns go for other parties leaving the others column, I think we have set a very sensible precedent: Wait 7 months from when the issue is first raised and see if it is a blip or if it stands the test of time! It could potentially happen with the Greens, who knows, we've just got to be open minded and treat that potential situation as this.


 * You say that no other opinion polling page shows a party cut down in pieces. Au contraire, please turn your attention to Opinion polling for the Italian general election, 2013. Notice that the party/"joint electoral lists" in 2013 do not correspond to 2012. Now I realise that I might be opening a can of worms by showing you this because there are lots of small parties included, there is very clear reasoning for this that does not apply to this article. Very small parties are included because they make up the electoral alliances and hence the governing coalitions: Naturally if a small party within an alliance is included, it is only reasonable to include a party that out polls that party, even if they are in no electoral alliance and their figure is stand alone. I could go on all night about the Italian election but I fear we have enough brick walls to bash our heads against! I'm only using the Italian example to illustrate that it is not unprecedented to have different parties in different tables for different years. We do it buy year because that is standard practice, the data can't all be in one table, needs to be broken up some how. This was the precedent/rule set a long time ago.


 * As for the 3rd 2nd Party Lead Column and the hypothetical situations, there are already existing solutions to those potential problems, that would mean that there is NO instance where 3lead columns would be necessary. What we have for now is not likely to become a problem in 2013 but if it does it can be adapted to:


 * 2013 using hypothetical numbers


 * The above layout is how we could make the current table work in the very unlikely event that 2nd and 3rd place start to alternate. We have done this before with Opinion polling for the United Kingdom general election, 2010. I seriously doubt we will need to, our reliable sources say nothing about this being a potential outcome either! If we get to the point of 2014 and we discover that 2013 gives us every reason to believe that a close battle between 2nd and 3rd place is likely to continue, then we drop the third party lead column and just keep the 1st party lead column with "lead over con/UKIP/Lib Dem".


 * The point to me showing this is to prove you wrong that there is never a set of circumstances by which we will need 3lead columns! It's about picking the right solution! Being pragmatic about a fluid situation! 2lead columns without any "lead over ???" clearly does the job for 2012. We shall have to see what 2013 brings, I doubt we will need to add "lead over ???" But if we do it's an easy edit to make! To compare this situation to pre 2010 does not make sense. There were only 3 parties polling anything that could be considered significant. Out of the "Others" the largest figure was UKIP which had 3.1%, by no stretch of the imagination a significant figure! This time we have 4parties polling significant figures, this creates a different dynamic. We currently have 2 "two horse races", that is certainly how our reliable sources are presenting it. As I have said if that situation changes, we have solutions for any possible outcome that would NEVER result in more than 2lead columns! The 3rd party lead column is highly useful at the moment and reflects our sources!


 * I hope I have satisfied your concerns.Sheffno1gunner (talk) 19:20, 22 January 2013 (UTC)


 * Can we respond to comment in order please.


 * Solved. For next answers, please, use ":" before posting what you're commenting. That'd help to bring more order to the discussion.


 * it would screw up the other's line and look an absolute mess!


 * I can't see the issue. You only have to substract UKIP's data from Others' one in the table. Like this:


 * Before
 * {| class="wikitable sortable" style="text-align:center;font-size:95%;line-height:14px"

!! width="100px" | Date(s) conducted !!width="350px"|Polling organisation/client!!class="unsortable" width="65px" | Sample size!!class="unsortable" style="background:#0087DC" width="90px"| Cons !!class="unsortable" style="background:#DC241F" width="90px"| Lab !! class="unsortable" style="background:#FDBB30" width="90px"| Lib Dem !!class="unsortable" style="background:gray" width="90px"|  Others !! class="unsortable" width="40px" |Lead
 * 22–23 Dec || YouGov/The Sun || 1,635 || 39% || style="background:#F8C1BE" | 41% || 9% || 11% || style="background:#DC241F; color:white;"| 2%
 * }
 * }


 * After
 * {| class="wikitable sortable" style="text-align:center;font-size:95%;line-height:14px"

!! width="100px" | Date(s) conducted !!width="350px"|Polling organisation/client!!class="unsortable" width="65px" | Sample size!!class="unsortable" style="background:#0087DC" width="90px"| Cons !!class="unsortable" style="background:#DC241F" width="90px"| Lab !! class="unsortable" style="background:#FDBB30" width="90px"| Lib Dem !! class="unsortable" style="background:#70147A" width="90px"| UKIP !! class="unsortable" style="background:gray" width="90px"| Others !! class="unsortable" width="40px" |Lead
 * 22–23 Dec || YouGov/The Sun || 1,635 || 39% || style="background:#F8C1BE" | 41% || 9% || 5% || 6% || style="background:#DC241F; color:white;"| 2%
 * }
 * }


 * Don't see where is the issue.


 * This is why I propose that we also put the result for the previous election at the top of the page.


 * No, no. Look, you are simply complicating an issue that could just be solved by adding all of UKIP data, which would need just one election result, the one in the first table. What you're proposing would require (in order to be consistent) to have the election results in ALL of the tables (this article has one for each polling year).


 * Also, if you remember back to the discussions for adding the column, that was already discussed! If UKIP support fell away back to an insignificant level for the same period that they had been neck and neck with the Lib Dems then that would trigger reopening the discussion.


 * Wouldn't it be easier to just remove it entirely rather than having it discussed over and over again, possibly resulting in bringing the 3rd lead column back and forth over and over again?


 * You say that no other opinion polling page shows a party cut down in pieces. Au contraire, please turn your attention to Opinion polling for the Italian general election, 2013. Notice that the party/"joint electoral lists" in 2013 do not correspond to 2012.


 * Yeah, well, what I highlighted in bold actually explains it all. In Italy, joint electoral lists and coalitions are formed when elections are near. Because of the electoral system of Italy that gives an absolute majority of seats automatically to the winning coalition, it makes sense to include ALL coalitions' member parties in the table when coalitions are formalised, because the absolute majority of seats could be won by either side by a margin of just 1 vote, and thus, parties which were irrelevant before joining a coalition can become of a vital importance when they do. This is entirely different to what happens in UK.


 * We do it buy year because that is standard practice, the data can't all be in one table, needs to be broken up some how. This was the precedent/rule set a long time ago.


 * I know it was a rule that was set a long time ago (that's why I never tried to undo it), yet I never understood the sense of it, since there are more practical and effective ways to solve the article-long span of the tables. I'm more in favour of something like what I did in Spanish opinion polling articles. For example: Opinion polling for the Spanish general election, 2011. As of currently, it's the Spanish polling article which has the most opinion polls (you can see it by yourself), yet it's made so that it only occupies a small part of the article. Maybe one day I can suggest trying to do something similar to this article, but well.


 * What we have for now is not likely to become a problem in 2013 but if it does it can be adapted to: 2013 using hypothetical numbers


 * I never liked what you show there (I know it's applied in the 2010 election polling article, but I never liked it there either). Mostly because it enlarges the lead column in a weird way. Also, should that scenario becomes true, the 3rd party issue would become more tricky. So, we will have both 1st and 3rd parties highlighted and in bold, yet the 2nd party (which has more votes than the 3rd, obviously) won't. The issue would become crazier if the 2nd and 3rd parties rotate their places in a frequent manner.


 * The 3rd party lead column is highly useful


 * If it was so useful, I bet other opinion articles would have already made use of it, or tried something similar. Even I considered doing something similar in the Spanish polling articles when it became clear that IU and UPyD would fight for the 3rd place in the next election. But no one has ever introduced it at their articles. I never introduced it in Spain article. And not because it wasn't useful at all: it could be interesting. But I think it enters into the "excess of information" category, and that it was not really necessary. I believe things should be jugded and introduced by considering their usefulness and need for it. And in this case, I don't see the need for this column. Once again, I'll comply with what the majority says. But maybe the need for such a column should be rethought. Impru20 (talk) 20:59, 22 January 2013 (UTC)


 * By the others line, I think Sheff meant the line on the graph not the "column". If you picture the graph, it would look a little bit odd. I saw that they did that for a minor party in the old Italian graph (it's been deleted for some reason, the "others" line kind of collapsed and then you'd got this random line that came from nowhere, it looked a complete mess so I'd like to rule that one out!


 * However with regard to columns I'm inclined to agree with you! Let's just add UKIP's data to 2010 and 2011 and have done with it, that solves that particular issue of the graph and the columns being inconsistent. So, I hope we get consensus on this part of it at least! Also if we do as you suggest there we don't need to put the result at the top of the page, which also avoids "duplicity" as people have been saying! So hopefully that bit is sorted? What's more we already have all the data on this talk page, which saves hours of opening and closing and scrolling through every single poll!


 * Right that just leaves the issue of lead columns! For 2012 at least (and 2013), I like the 3rd party column, it's relevant and actually helpful because it does change a lot. However it's pointless for 2010 and 2011 because UKIP are nowhere to be seen. If we use the 3rd party lead column in the right places, then we are not overloading with information. I think the article is better for having the 3rd party lead column in years where 3rd place is constantly changing. In 2012 the 3rd party column is actually more necessary that the 1st party lead column, that might well be the same for 2013! I firmly defend the 3rd party column for 2012 but agree with a o of what you've said!


 * However, your concerned about consistency, well the lead columns are not consistent on other articles (i.e. pre 2010). As far as that one goes, I think it's important to see the lead column's as a readers aid, so we've got to try and be helpful to the reader with this. It's been said that "oh the reader can work it out for themselves", true but if we follow that logic then we should disband both lead columns and bold/highlighted percentages altogether. I totally get what your saying about including UKIP throughout and comparing it with the graph etc.... I agree with you on that and why that affects consistency so much but...... adding and removing one narrow column every year where it is needed is not a big deal, it's merely a readers aid! I get why it annoys you... you do, do a lot of the work of filling the polls in, so yea it's a tiny bit more work each time. I get that but others are sharing the burden more now.


 * As far as having the lead over ??? in the column, it kind of has already been established as a solution when needed. I'm sorry to hear that you don't like it; I think it works rather well on European Parliament election, 2009 (United Kingdom). As has already been said though, the odds of that happening are incredibly unlikely! So let's not consider that for now, it sounds to me like that was only suggested as a back up plan! Tell you what how about I put a tenner in your paypal account if it does happen? I'm that sure it won't! The odds of needing 2 columns and a lead over column are so slim because if you think about it, we only need a 3rd place column when 3rd and 4th place are alternating! That would mean 3 parties (2nd, 3rd and 4th) would have to all be within 5% of each other. I will put £50 in your paypal account if that happens! lol The odds of us needing both solutions are slim and even if it does happen, is it going to happen for a long period of time? No, highly unlikely, so we don't have to widen the column, just deepen the individual row in the odd instance where it does happen.


 * This all sounds highly technical but realistically we're not going to have to go near any of these solutions and when we do, they will most likely be blips... one off's! I really hope that settles it! Most of your suggestions I have agreed with but I really do think the 3rd party lead column is of real value to a reader!217.41.32.3 (talk) 00:15, 23 January 2013 (UTC)


 * Ok, I'm going to agree to add the UKIP column to 2010 and 2011, your argument on that point does make sense and I'm not going to be that one arrogant editor who refuses to accept obvious logic. I mean it does solve a number of problem! As far as my solutions for the lead columns go, (81.133.12.45 has got exactly what I meant and actually further convinced that this is a workable solution if it comes to it. And yes the odds of us needing to use these solutions are very low! If never it will only be occasional use! Also I've made my feelings plain about the 3rd party column in general, lets not get rid. I don't want to repeat myself or 81.133.12.45's arguments. We're at the point now where we have a workable solution that should be agreeable to all. I hope we don't end up going round in circles!Sheffno1gunner (talk) 13:42, 23 January 2013 (UTC)
 * I'm in agreement with 81.133.12.45's argument! Let's call it a day and set to work on it hey?130.88.52.99 (talk) 13:44, 23 January 2013 (UTC)


 * I've changed the text formatting of your replies in order to place it where they are intended to be, to avoid confusion and bring an order to the discussion. Hope you don't mind.


 * So, the UKIP issue seems to be settled too, since Sheff seems to be in agreement as well. One point less to discuss.


 * However, your concerned about consistency, well the lead columns are not consistent on other articles (i.e. pre 2010).


 * Not actually. My concern is not as much about consistency with other articles, but more about consistency with this article (the reference to other articles was to show how those were consistent with all of their data). Because polling is divided by years, each table works independently from the others when it comes to editing them. So, it has turned that now we have two different kind of tables: one for 2010 and 2011 (without the 3rd lead column and, until now, without UKIP) and other for 2012 and 2013 (with the 3rd lead column and with UKIP). That's what I mean by not being consistent: the organisation and interpretation of opinion polls works in a different way within the same article. I don't know which was the main reason behind diving polling by years, but I guess it wasn't to have a different formatting for each table. As I said several times, I'm not going to go against the 3rd lead column if the majority agree with it, so that would discard the option of removing it. So then, if consistency within the article is to be reached, the 3rd lead column should also be introduced to 2010 and 2011 (which will also be justified by the fact that now UKIP will be in those tables too, and it will be a way to compare the 3th-4th parties' distance back then and now).


 * As far as having the lead over ??? in the column, it kind of has already been established as a solution when needed.


 * Yeah, well, I don't have anything against it. I don't like it because it needlessly enlarges the width of the lead column, but I don't have major issues with it being in place where extremelly necessary (for example, in the 2010 article, or in the 2009 European election article, where it's obviously needed because the 2nd place is constantly changing hands at some points). Should the need arises for it to be placed here, I'll have nothing against it, so let's settle that issue here. Impru20 (talk) 15:19, 23 January 2013 (UTC)


 * Have started adding UKIP to 2011, too tired so am going to go to bed. Will carry on when I can but if anyone feels like helping then by all means! Carry on. Cheers81.133.12.45 (talk) 01:43, 23 January 2013 (UTC)
 * Have picked up where you have left off on the additional column. Shouldn't be too much longer!Sheffno1gunner (talk) 02:09, 24 January 2013 (UTC)
 * I have completed the changes that have been agreed. I was in agreement in the end because your logic did make sense, so the UKIP columns issue is done and dusted with.

As far as the potential use of "lead over" in the 1st party column goes, I don't see why it needs to widen the column, if we fix the column to be 20px, it will mean that the one particular row concerned deepens. Like we've said, we'll cross that bridge if and when we come to it!

Ok, our one remaining issue, the notorious (but necessary) 3rd party lead column:

I understand your argument of consistency which is what has persuaded me to agree with you on most of your suggested changes. I am a stickler for consistency but I am not wholly rigid when there is good reason for an exception. I see the lead column's as a readers aid, they're helpful! We've got the sight issue of UKIP not technically being the 4th party of the UK in some individual polls but yes in the vast majority they are, so I will be prepared to bend on that I guess. I am not whole heartedly against having a wall of yellow at the side of the 2010 and 2011 tables, that really doesn't bother me, I just don't think it's all that helpful. Also if we need to have a slightly different format for 2014 and 2015, something that has already been discussed, then it would be better to have 3 different takes on 3rd party rather than 2. Here's why:
 * If it becomes the case that 2nd and 3rd place are constantly changing (e.g. because of say another UKIP surge and Tory decline), while at the same time the gap between 3rd and 4th place becomes so large (e.g. the Lib Dem's being 10 points behind 3rd place), in that case the 3rd party lead column becomes redundant and it would then be better to have 1 column with "lead over in it".
 * Like I say the lead columns are a readers aid, to add 3rd party leads to a year when 3rd place comes nowhere near contention doesn't seem to help the reader to me. Also it's immediately obvious to see that UKIP's figures are always low single digits and the Lib Dem's are in double digits, that in a sense is doing the job of the 3rd party lead column because the contrast is so stark.
 * If we come to 2015 and the pigs aren't flying (no offence to UKIP and their supporters) then I'd be happy to add the 3rd party lead column to 2010 and 2011. For now at least, I don't see the point in adding it until we're sure that it won't need removing. You never know this pig might go on a diet and grow some wings, the situation is just so unpredictable at the moment. We're not here to predict it but we do need to accommodate any eventuality that effects our tables, so I'm very pleased that editors have engaged in that part of the discussion with an open mind.

I hope we've got this one settled now, thanks :-)Sheffno1gunner (talk) 13:20, 24 January 2013 (UTC)


 * Ok, your '3rd party lead' argument makes sense. Not that it made my change my mind about my liking of the 3rd party lead column, but it is convincing enough to explain that it doesn't do anything bad, and that it could be helpful at some degree. Anyway, I already convinced you of adding UKIP to the table in 2010 and 2011 (which also makes sense, for consistency), so I can concede on this issue, at least until (and if) the urge arises for it to be discussed again. ;) Impru20 (talk) 17:50, 24 January 2013 (UTC)
 * ✅ All changes complete. Thank's for a sensible discussion with a well discussed solution! Discussion closed
 * No, not closed. I still think the third party lead is a bad idea; my apologies that I haven't laid out a longer explanation of what I see as the problem yet. But, in brief, it still looks very much like WP:OR to me. Bondegezou (talk) 17:13, 30 January 2013 (UTC)

New Charts- to reflect new format or update/adapt existing? UKGeneralElection2015Polls.svg
Pleased to see that UKIP have been given their own column. This is a good decission because we have been careful to show that we're not simply reacting to recent events, we're giving it the "flash in the pan" test, by waiting a whole 9months of UKIP being within the margin of error of the Lib Dem's. On some occassions overtaking UKIP have over taken the Lib Dems in individual polls and this is now happening more frequesntly, we're looking about 50/50 now, we can now be sure we have waited long enough! Good decision guys, you have been completely fair and maintained wikipedia's partiallity! I have added a note on the page at the top of the 2012 block to briefly explain why this decision has been made! However for the time being I have removed the chart from the page because it doesn't reflect the data displayed properly. I suggest we do one of 2 things: 1. Have 2 tables; move the original table down to below the 2012 section, to reflect the 1st 2years of the Parliament. We can then have a new graph in the chart section, following the new format, including the 4main parties and "others" starting from the beginning of 2012 (or from about March/May 2012). I have taken the liberty of putting all the information for 2012 into Excell (quite Laborious, I tell you!) I can send it on to someone to form a new graph using wikipedia editor's favoured format. 2 We can just have 1 chart going back to the beginning of the Parliament, including all 4 main parties and "Others". In which case my offer still stands to send on the Excell spreadsheet.

Editors: what's your favourd option? Hope this helps. Many Thanks and keep up the good work! Nick  ' 01:01, 6 December 2012 (UTC)

Proposal 1 - Excel with trend line

 * I've done the entire results since 2010 in Excel, done a scatter graph and then put in a trendline, like on the Spanish Opinion Poll page. Spa-Franks (talk) 22:53, 7 December 2012 (UTC)

Oh, wow, your brilliant! And very patient! Can you upload it and give us a preview on this talk page please? Then we can discuss adding it. Nick  ' 08:12, 8 December 2012 (UTC)


 * It's not my graph, but I came across it while making one for the UK page. I am, and, for others, I did =100%-(SUM(B2:E2)) and so on and so forth. The problem is that when there wasn't a UKIP result in 2010, I took the previous result. I might have to go back over it. However, I've never uploaded a pic here before... Spa-Franks (talk) 18:34, 9 December 2012 (UTC)

Right ok, as for the UKIP results for say 2010, to be accurate what you might need to do is open the source for the polls and it will be in there somewhere under others. For the YouGov ones, I know it defo is but not sure about every single poll. Don't forget to subtract the UKIP % from the "Others" % if that is what you decide to do.

To upload an image it is quite simple: On the left hand side of the page (below the wiki logo) you will see a list of menus, go to toolbox and about half way down is Upload Fie. Since you've made it yourself, you click on "this is my own work" option, don't forget tgo label it clearly. If you want to upload what you've got so far and I'll take a look. Thanks again for this! Nick  ' 18:44, 9 December 2012 (UTC)

Right then. Correct up until 06 December 2012:. Thoughts? Spa-Franks (talk) 17:17, 10 December 2012 (UTC)

Yea good stuff, a few things though: Because there is so much data it doesn't seem worth plotting the points because you just a big smear! Is there any easy way of removing the dotts and jut showing the trend line instead? I mean we've got all the data, this is just a visual representation. Or one alternative might be to try and and make the plotted points smaller, is that possible? Woud it make any difference? If not then just the trend line will do in my view :-) Well done, again! Oh and can UKIP's line colour be changed to purple and "Others" to Grey, is that possible? Also can you get the graph scale to start from May 2012 and finish December2012 so that there are no gaps on the graph, this might help to avoid the blotchyness of the plotted points. Cheers Nick   ' 18:48, 11 December 2012 (UTC)

Proposal 2 - Proportional size box graph
I think that because the Lid Dem, UKIP and Other results are so close it makes it hard to read. I have made one up quickly: CH7i5 (talk) 23:38, 19 December 2012 (UTC)

Why on earth did I not think of that? It seems such an obvious solution, very well done CH7i5!, I vote to include it, you can always add to it again in May 2013, in the section below, we have agreed to add something else in 6months but since you've gone to the effort and it works really well, I say lets just upload it now. What do people think? — Preceding unsigned comment added by 217.41.32.3 (talk) 19:34, 20 December 2012 (UTC)

May I echo the enthusiastic praise of the IP user! Good thinking Batman! Yes, lets include it, we can always update it at the agreed date. Hopefully CH7i5 will be prepared to hold on to the file untill May 2013/ end of 2013 to update the tabe again as agreed? I think given the poll ratings of Lib Dem, UKIP and Others being so close to one another it is impractical to use a line graph, I think for now, this is the only way. However times might change, so I suggest using this type of graph untill atleast Jan2014 and then we can review the situation. Once again, well done! Nick  ' 19:40, 20 December 2012 (UTC)

May I echo everything said thus far, this is a great chart! However, it could be slghtly better. I am in favour of ading it provided that it is t the end of the polling for 2012 and includes all of 2012's polls and that the x axis is modified to have mre mrker points e.g. 1st of every month. Also because I am a stikler for consistency, I think the name "UK Independence Party" should be shortened to "UKIP", this will tally with the table and evertything else. If you can do all of that, then please feel free to upload because it seems as if we have concensus. P.S. Thankyou for finding a solution to this problem. Sheffno1gunner (talk • contribs) 19:49, 20 December 2012 (UTC)

Thank you! Glad you like it, I will take the recommendations and upload the full file as soon as all the polls for 2012 are released. There are several polls that are not in the graph, these are the ones in which there was no individual figure for UKIP, however I can add these back in if needed. I did not do that as it looked like UKIP polled 0%, but if the majority of people want it I will do it. I will also add all the beginning of each month into the graph as well. CH7i5 (talk) 20:25, 20 December 2012 (UTC)

I think once again, you have made the right call! It would skew the data if those individual polls are included, so yea don't include them, it's only for a small handfull of polls so their omitionis not a big deal. However artificially giving any party zero would be a big deal and would cause serious undulacions in your lovely graph. It would also make the data less easy to understand but the bottom ine is that it would create an inaccurate graph. Stick to your guns, you made the right call to begin with! And yea, showing the 1st of each month on the axis would be great, easier for viewers to understand. Once again though, please change the name to UKIP, consistency is very important :-) Sheffno1gunner (talk • contribs) 22:32, 20 December 2012 (UTC)

Proposal 3 - New scatter plot & Line Graph
I like both the graphs offered above. The first one (by Spa-Franks) shows the vagaries of polling - different polls, different methodologies, margins of error - and the trend lines. The colours and time-scale needed changing, but that is a small thing. The trend lines were interesting, but misleading. It looks, for example, as if Labour's polling is about to plummet. It might, but we don't know yet, and Wikipedia is not a crystal ball. The second one (by CH7i5) shows the proportions of the four biggest parties and the others. It needs month markers (vertical lines), but even with this tweak to me it is not clear. That's because it is hard to see which party is polling the most at any given time and it is hard to discern a trend. I think this is a difficult problem to crack because we are trying to depict at least two different stories: the first is the battle between the Conservatives and Labour for who will form the next government; the second is the battle for third place between the Lib-Dems and UKIP. Here's my flawed attempt. The following graph is a reworking of the one by Spa-Franks. It is not a finished product: the time-scale is not accurate (you will find jumps of between 8 and 13 days) and it is just for 2012, to-date. Where no UKIP data existed, I imputed figures and adjusted the 'others' score. And the labelling would need to be clarified: in this case LabourM shows the moving average for Labour polling. I used a moving average of 7 data points, but another figure could equally be used.

Further comments and suggested improvements welcomed in this quarter. --Wavehunter (talk) 02:34, 21 December 2012 (UTC)

Now I'm confused, this is also a very good graph that you've produced! It tells the story of the battle for 1st place well and does a very reasonable job for 3rd place, whats more you've got the overall picture as well. I like it but one additional point, please correct the labeling, the 4 parties are: Cons, Lab, Lib Dem and UKIP, as well as Others; consistency is important. Tory is just a coloquial term and has not been a political party name for about 150 years. Another plus is that your graph would be more consistent with the ppearance of the 2010-2012 graph. Sheffno1gunner (talk • contribs) 02:58, 21 December 2012 (UTC)

I favour using both tables, the line graph for the whole of 2012 and the graph by CH7i5 for the whole of the Parliament, that way we're covering everything. I like Wavehunter's graph and agree with sheffno1gunner's comments. — Preceding unsigned comment added by 81.133.12.45 (talk) 12:13, 21 December 2012 (UTC)


 * If others agree, I'll happily upload an amended and updated version once the 2012 polling results are complete (probably early January). Perhaps someone will give me a nod when all the data is in. --Wavehunter (talk) 18:58, 21 December 2012 (UTC)

Sounds good to me! — Preceding unsigned comment added by 94.2.42.187 (talk) 19:35, 21 December 2012 (UTC)


 * Is the 2012 table now complete? If it is, I can upload an improved version of my chart. --Wavehunter (talk) 17:53, 5 January 2013 (UTC)

I believe it is, go ahead and upload, I look forward to seeing it :-) — Preceding unsigned comment added by 81.133.12.45 (talk) 14:30, 6 January 2013 (UTC)

Yea, go for it, we're all done with 2012 now, moved on to 2013. Please make sure you keep your graph and all its data safe, might ask you to add to it. Nick  ' 17:24, 6 December 2012 (UTC)

Yes please, could you upload it to the talk page first just so that we can look it over first, thank you. Sheffno1gunner (talk) 17:11, 7 January 2013 (UTC)

@Wavehunter is there anyway that you would be prepared to go through the data for 2010 and 2011 and add that to your graph, that way your graph would be the only graph we would need for the whole article. I know it's a lot to ask because you'd have to open each poll up to get UKIP's individual figure and then also subtract that figure from "Others". It would be ideal if we could have just one graph, I understand though if your not prepared to spend the time doing that. It's not ideal to have 2 separate graphs but it's workable! One thing we do know, is that your graph (with the alterations that we have mentioned would be the best for the article!Sheffno1gunner (talk) 20:38, 7 January 2013 (UTC)

Maybe put your key in a row underneath the graph as with other graphs, rather than in a column at the side. Also it does not need to be repeated. I would favour keeping the diamonds part of the key and deleting the line. — Preceding unsigned comment added by 217.41.32.3 (talk) 13:06, 8 January 2013 (UTC)

Solution
How is this? I won't have time to add polling data from previous years, but I will happily email my file (an Open Office file, compatible with Excel) to any volunteers. I don't know how you would get your email address to me, mind - I wouldn't recommend typing it in here. --Wavehunter (talk) 20:58, 8 January 2013 (UTC)

That is ideal! You have satisfied every request for amendments so I think we'll just upload it straight away! Well done! Personally I would not be prepared to give out my email on here, I suggest that we leave 2010 and 2011 as they are. That is unless you get really bored and want to merge the 2 graphs. I'm happy to keep it as it is. Could you update this at least once every 6months? The next opportune moment to update is after the final poll before the 2013 elections is published (around the 4th of May sort of time). Sheffno1gunner (talk) 05:25, 9 January 2013 (UTC)

✅ Uploaded your graph to the article. Please be sure to hold on to your data and graph.Sheffno1gunner (talk) 05:35, 9 January 2013 (UTC)

Looks really good, I approve! Is there any way that you could save the file online? I'm not great with technology but surely such a facility exists? A bit like Skydrive but more public, I guess is what I mean. Nick  ' 07:24, 9 December 2012 (UTC)

All sorted! Yes it would be helpful if that file could somehow be made public, that way any of us that felt like doing it could update it and be less of a burden on one person! Does Wikipedia not have it's own kind of version of Skydrive on Wikimedia? Does anyone know? — Preceding unsigned comment added by 81.133.12.45 (talk) 07:57, 9 January 2013 (UTC)

Size split?
 Split  - Article is over 200 kB, and should be split. Thoughts? Suggestions?--Jax 0677 (talk) 04:27, 1 January 2013 (UTC)

There's no point in splitting the article other than meeting the 200kb limit. This page is bound to be large because it contains a lot of data, it's the nature of the beast. I'd be against splitting it. — Preceding unsigned comment added by 86.145.3.115 (talk) 22:38, 1 January 2013 (UTC)

Splitting the article would be a bad idea. The bulk of the article is one thing: the table of opinion polls, plus some explanatory content. The only logical way to split the page would be to divide opinion polls by year, but it's more useful to readers to have all the opinion polls on one page, and not have to go clicking around to follow them from year to year. With the current table, readers can scan the changes and variations in the opinion poll over time, stretching from the last general election to the present day, concluding with the results of the next general election. Per WP:SIZESPLIT, size guidelines apply less strongly to lists and tables, particularly sortable tables. This is exactly the type of article to which those guidelines should not be applied and are not intended to be applied. Circumspect (talk) 11:03, 2 January 2013 (UTC)

This idea makes no sense! The only reason it goes over the limit is because it holds a lot of data! All the page is, is 4 tables, a graph, a bit of methodology and a wee bit of background, I really don't see why you'd want to split this, it would be of detriment to the article! — Preceding unsigned comment added by 217.41.32.3 (talk) 19:48, 4 January 2013 (UTC)

No, lets not split this article or make any changes along those lines. As others have said the only reason this article breaches the recommended character limit is because it holds a lot of data. In terms of written information there is hardly any! The vast majority of the characters are in the tables, which include the data and the formatting. There is no logical reason to split this article. Nick  ' 15:16, 5 January 2013 (UTC)

No, leave it as it is, I'm in agreement with what others have said. Other then a lot of data, there is not an awful lot on this page. I think we have consensus, no split required Sheffno1gunner (talk) 17:26, 5 January 2013 (UTC)

I have removed the splitting notice on the page since the issue has been resolved!Sheffno1gunner (talk) 17:32, 5 January 2013 (UTC)


 * It would be possible to reduce the size of this page by moving the polling company methodology to another article. Perhaps it could got to a site called "UK Opinion Pollsters". We could also make the page smaller by removing the "1st party lead" and "3rd party lead columns. Personally, I don't think they are necessary. --Wavehunter (talk) 17:53, 5 January 2013 (UTC)

Thanks for the suggestions Wavehunter but when you think that we're only half way through the Parliament and hence only half way through the polling your suggestions do not tackle the perceived problem. There are only 2 options here, split the article or do what we normally do and make an exception on the grounds that the article is only long because of the large amount of data! Removing columns and moving background to other articles will not make the slightest bit of difference to addressing the perceived problem. It's the nature of the beast, if we're to be accurate and include as many opinion polls as possible (not being overly reliant on one company) then it's always going to be a big article. It is pretty standard to have a column for 1st party lead but given the nature of where the polls are at, the third party column is also of benefit and makes the article more user friendly. The consensus seems to be leave the article as it is, in a few days the split message on the main page will be removed once it has served it's week! — Preceding unsigned comment added by 217.41.32.3 (talk) 20:10, 5 January 2013 (UTC)

✅ This matter is now resolved and the notice has been removed. Discussion concluded.

New Graph to replace existing 2
I have spent many hours trawling through the 2011 and 2010 polling links to get individual figures for UKIP and then subtract that figure from the "Others" column! We are now able to have 1 graph for this entire article, spanning from the previous election to the present day!! Sheffno1gunner (talk) 04:40, 11 January 2013 (UTC)

Copy of data for 2010 and 2011
--- --- ---     Do Not Edit This Subsection! --- --- --- Below is a copy of the data for 2010 and 2011 which includes a figure for UKIP, I am sharing this so that it can be used for updating the chart as and when necessary, that way it doesn't have to be me. Not that I am washing my hands of this far from it!

To use the data, you will need to copy it via the edit page. As you can see it does not appear in table format on here but it does if you look on the edit page. You can simply copy this and attach it to the data in the existing tables.

Do Not Edit This Data!

21–22 Dec	40.00%	40.00%	9.00%	4.00%	7.00% 20–21 Dec	40.00%	40.00%	10.00%	4.00%	7.00% 20–21 Dec	37.00%	36.00%	15.00%	5.00%	8.00% 19–20 Dec	39.00%	40.00%	10.00%	4.00%	7.00% 18–19 Dec	38.00%	42.00%	9.00%	3.00%	7.00% 16–19 Dec	35.00%	38.00%	11.00%	3.00%	13.00% 16–18 Dec	35.00%	39.00%	12.00%	2.00%	13.00% 15–16 Dec	39.00%	42.00%	9.00%	4.00%	7.00% 14–15 Dec	41.00%	40.00%	10.00%	4.00%	6.00% 14–15 Dec	40.00%	34.00%	14.00%	3.00%	8.00% 13–14 Dec	40.00%	38.00%	10.00%	5.00%	7.00% 12–13 Dec	41.00%	39.00%	10.00%	3.00%	7.00% 11–12 Dec	39.00%	40.00%	10.00%	4.00%	8.00% 10–12 Dec	41.00%	39.00%	11.00%	4.00%	5.00% 9–11 Dec	38.00%	38.00%	12.00%	2.00%	10.00% 8–9 Dec	38.00%	39.00%	11.00%	5.00%	8.00% 6–7 Dec	35.00%	42.00%	9.00%	6.00%	8.00% 5–6 Dec	37.00%	41.00%	10.00%	5.00%	7.00% 4–5 Dec	36.00%	42.00%	11.00%	4.00%	7.00% 1–2 Dec	35.00%	43.00%	9.00%	6.00%	13.00% 30 Nov – 1 Dec	36.00%	41.00%	11.00%	4.00%	8.00% 29–30 Nov	37.00%	42.00%	9.00%	6.00%	7.00% 29–30 Nov	38.00%	36.00%	14.00%	2.00%	10.00% 28–29 Nov	38.00%	41.00%	9.00%	5.00%	7.00% 27–28 Nov	37.00%	39.00%	9.00%	6.00%	8.00% 25–27 Nov	37.00%	39.00%	10.00%	3.00%	11.00% 24–25 Nov	34.00%	43.00%	11.00%	5.00%	7.00% 23–24 Nov	33.00%	42.00%	8.00%	7.00%	10.00% 23–24 Nov	35.00%	40.00%	9.00%	8.00%	8.00% 22–23 Nov	35.00%	40.00%	11.00%	6.00%	8.00% 21–22 Nov	35.00%	42.00%	9.00%	6.00%	8.00% 20–21 Nov	36.00%	40.00%	9.00%	7.00%	8.00% 19–21 Nov	34.00%	41.00%	12.00%	3.00%	10.00% 18–21 Nov	36.00%	37.00%	9.00%	7.00%	11.00% 18–20 Nov	33.00%	41.00%	13.00%	4.00%	9.00% 18–20 Nov	36.00%	38.00%	14.00%	4.00%	8.00% 17–18 Nov	36.00%	40.00%	9.00%	7.00%	8.00% 16–17 Nov	34.00%	40.00%	11.00%	7.00%	8.00% 15–16 Nov	36.00%	41.00%	10.00%	6.00%	7.00% 14–15 Nov	36.00%	42.00%	7.00%	5.00%	8.00% 13–14 Nov	37.00%	40.00%	9.00%	6.00%	7.00% 10–11 Nov	36.00%	41.00%	9.00%	5.00%	9.00% 9–10 Nov	35.00%	42.00%	8.00%	7.00%	8.00% 8–9 Nov	36.00%	40.00%	10.00%	7.00%	7.00% 7–8 Nov	35.00%	40.00%	10.00%	6.00%	8.00% 6–7 Nov	36.00%	41.00%	9.00%	6.00%	7.00% 4–7 Nov	34.00%	38.00%	10.00%	6.00%	12.00% 3–4 Nov	35.00%	41.00%	9.00%	5.00%	10.00% 2–3 Nov	36.00%	41.00%	8.00%	7.00%	8.00% 1–2 Nov	37.00%	41.00%	8.00%	6.00%	7.00% 31 Oct – 1 Nov	35.00%	41.00%	9.00%	5.00%	10.00% 30–31 Oct	39.00%	41.00%	8.00%	4.00%	9.00% 27–31 Oct	36.00%	37.00%	11.00%	4.00%	12.00% 28–30 Oct	34.00%	38.00%	14.00%	4.00%	10.00% 27–28 Oct	36.00%	39.00%	8.00%	4.00%	12.00% 26–27 Oct	35.00%	42.00%	9.00%	4.00%	10.00% 25–26 Oct	35.00%	41.00%	10.00%	4.00%	11.00% 24–25 Oct	36.00%	40.00%	9.00%	4.00%	11.00% 23–24 Oct	36.00%	40.00%	9.00%	4.00%	10.00% 22–24 Oct	34.00%	38.00%	12.00%	4.00%	11.00% 21–24 Oct	33.00%	39.00%	9.00%	8.00%	11.00% 21–23 Oct	33.00%	41.00%	10.00%	7.00%	10.00% 21–23 Oct	35.00%	39.00%	13.00%	3.00%	10.00% 20–21 Oct	36.00%	38.00%	10.00%	5.00%	11.00% 19–20 Oct	36.00%	41.00%	10.00%	5.00%	8.00% 18–19 Oct	35.00%	41.00%	9.00%	5.00%	9.00% 17–18 Oct	38.00%	42.00%	9.00%	5.00%	6.00% 16–17 Oct	37.00%	40.00%	9.00%	5.00%	9.00% 14–16 Oct	33.00%	41.00%	8.00%	5.00%	7.00% 13–14 Oct	39.00%	42.00%	8.00%	6.00%	5.00% 12–13 Oct	37.00%	42.00%	9.00%	6.00%	6.00% 12–13 Oct	37.00%	39.00%	10.00%	6.00%	8.00% 11–12 Oct	36.00%	42.00%	9.00%	6.00%	8.00% 10–11 Oct	37.00%	41.00%	8.00%	7.00%	7.00% 9–10 Oct	36.00%	40.00%	11.00%	7.00%	6.00% 7–10 Oct	36.00%	37.00%	8.00%	7.00%	12.00% 6–7 Oct	38.00%	42.00%	9.00%	6.00%	6.00% 5–6 Oct	37.00%	41.00%	10.00%	6.00%	6.00% 4–5 Oct	37.00%	41.00%	9.00%	6.00%	7.00% 3–4 Oct	38.00%	42.00%	9.00%	6.00%	6.00% 2–3 Oct	37.00%	42.00%	9.00%	6.00%	6.00% 29–30 Sep	36.00%	42.00%	10.00%	6.00%	6.00% 27–30 Sep	33.00%	40.00%	9.00%	6.00%	12.00% 28–29 Sep	38.00%	41.00%	9.00%	4.00%	6.00% 27–28 Sep	37.00%	43.00%	9.00%	4.00%	6.00% 26–27 Sep	37.00%	43.00%	8.00%	4.00%	8.00% 25–26 Sep	39.00%	41.00%	8.00%	4.00%	8.00% 23–25 Sep	37.00%	36.00%	12.00%	4.00%	11.00% 22–23 Sep	36.00%	42.00%	11.00%	3.00%	8.00% 21–22 Sep	36.00%	42.00%	10.00%	3.00%	9.00% 20–21 Sep	35.00%	41.00%	9.00%	3.00%	11.00% 20–21 Sep	37.00%	38.00%	14.00%	3.00%	9.00% 19–20 Sep	36.00%	41.00%	10.00%	4.00%	8.00% 18–19 Sep	36.00%	42.00%	10.00%	4.00%	8.00% 15–16 Sep	36.00%	42.00%	9.00%	5.00%	8.00% 14–15 Sep	38.00%	41.00%	9.00%	5.00%	7.00% 13–15 Sep	33.00%	36.00%	9.00%	8.00%	14.00% 13–14 Sep	37.00%	41.00%	10.00%	5.00%	8.00% 12–13 Sep	35.00%	43.00%	10.00%	5.00%	7.00% 11–12 Sep	37.00%	41.00%	10.00%	4.00%	7.00% 10–12 Sep	35.00%	37.00%	13.00%	3.00%	12.00% 9–11 Sep	34.00%	38.00%	12.00%	3.00%	13.00% 8–9 Sep	38.00%	41.00%	9.00%	4.00%	7.00% 7–8 Sep	36.00%	42.00%	10.00%	4.00%	8.00% 6–7 Sep	36.00%	42.00%	9.00%	5.00%	9.00% 5–6 Sep	38.00%	40.00%	9.00%	5.00%	8.00% 4–5 Sep	37.00%	43.00%	9.00%	6.00%	6.00% 2–5 Sep	36.00%	37.00%	9.00%	7.00%	11.00% 2–4 Sep	37.00%	38.00%	11.00%	2.00%	12.00% 1–2 Sep	33.00%	39.00%	11.00%	7.00%	11.00% 1–2 Sep	38.00%	39.00%	10.00%	6.00%	7.00% 31 Aug – 1 Sep	36.00%	42.00%	10.00%	5.00%	7.00% 30–31 Aug	37.00%	42.00%	10.00%	5.00%	7.00% 29–30 Aug	39.00%	40.00%	10.00%	5.00%	6.00% 25–26 Aug	38.00%	41.00%	9.00%	4.00%	9.00% 24–25 Aug	37.00%	42.00%	9.00%	4.00%	8.00% 23–24 Aug	36.00%	43.00%	9.00%	3.00%	10.00% 22–23 Aug	37.00%	44.00%	9.00%	3.00%	7.00% 21–22 Aug	35.00%	44.00%	9.00%	3.00%	9.00% 20–22 Aug	34.00%	40.00%	15.00%	2.00%	9.00% 19–21 Aug	37.00%	36.00%	17.00%	2.00%	8.00% 18–19 Aug	36.00%	40.00%	11.00%	3.00%	10.00% 16–19 Aug	37.00%	38.00%	9.00%	6.00%	11.00% 17–18 Aug	36.00%	44.00%	9.00%	5.00%	7.00% 17–18 Aug	38.00%	40.00%	11.00%	5.00%	6.00% 16–17 Aug	35.00%	44.00%	9.00%	5.00%	7.00% 15–16 Aug	36.00%	42.00%	10.00%	6.00%	6.00% 14–15 Aug	35.00%	43.00%	9.00%	6.00%	7.00% 11–12 Aug	36.00%	43.00%	9.00%	7.00%	5.00% 10–11 Aug	35.00%	43.00%	9.00%	7.00%	6.00% 9–11 Aug	34.00%	38.00%	9.00%	8.00%	11.00% 9–10 Aug	36.00%	43.00%	9.00%	7.00%	6.00% 8–9 Aug	36.00%	43.00%	9.00%	6.00%	7.00% 7–8 Aug	36.00%	44.00%	9.00%	6.00%	6.00% 4–5 Aug	35.00%	44.00%	9.00%	5.00%	7.00% 3–4 Aug	36.00%	42.00%	10.00%	5.00%	7.00% 2–3 Aug	35.00%	43.00%	10.00%	5.00%	6.00% 1–2 Aug	36.00%	45.00%	8.00%	5.00%	6.00% 31 Jul–1 Aug	35.00%	42.00%	11.00%	5.00%	7.00% 28–29 Jul	35.00%	44.00%	10.00%	4.00%	8.00% 27–28 Jul	36.00%	42.00%	11.00%	4.00%	7.00% 26–27 Jul	36.00%	43.00%	8.00%	4.00%	9.00% 25–26 Jul	35.00%	44.00%	9.00%	4.00%	8.00% 24–25 Jul	37.00%	41.00%	10.00%	4.00%	8.00% 22–24 Jul	34.00%	40.00%	13.00%	4.00%	9.00% 21–22 Jul	35.00%	43.00%	10.00%	5.00%	7.00% 20–21 Jul	36.00%	44.00%	9.00%	5.00%	7.00% 19–20 Jul	35.00%	43.00%	11.00%	6.00%	5.00% 19–20 Jul	34.00%	41.00%	10.00%	6.00%	10.00% 18–19 Jul	36.00%	43.00%	8.00%	5.00%	8.00% 17–18 Jul	37.00%	42.00%	9.00%	4.00%	8.00% 16–18 Jul	32.00%	39.00%	11.00%	4.00%	14.00% 15–17 Jul	37.00%	36.00%	16.00%	3.00%	8.00% 14–15 Jul	36.00%	42.00%	11.00%	3.00%	8.00% 13–14 Jul	36.00%	43.00%	9.00%	3.00%	8.00% 13–14 Jul	36.00%	40.00%	10.00%	3.00%	11.00% 12–13 Jul	35.00%	43.00%	10.00%	3.00%	9.00% 11–12 Jul	37.00%	42.00%	9.00%	3.00%	9.00% 10–11 Jul	35.00%	43.00%	10.00%	3.00%	9.00% 7–8 Jul	35.00%	44.00%	8.00%	3.00%	10.00% 6–7 Jul	37.00%	43.00%	8.00%	3.00%	9.00% 5–6 Jul	35.00%	43.00%	9.00%	3.00%	10.00% 4–5 Jul	35.00%	43.00%	10.00%	3.00%	9.00% 3–4 Jul	37.00%	43.00%	9.00%	3.00%	9.00% 30 Jun–1 Jul	36.00%	42.00%	9.00%	3.00%	10.00% 29–30 Jun	37.00%	42.00%	8.00%	3.00%	11.00% 28–29 Jun	37.00%	41.00%	10.00%	3.00%	9.00% 27–28 Jun	36.00%	43.00%	8.00%	3.00%	9.00% 26–27 Jun	37.00%	42.00%	10.00%	3.00%	8.00% 24–26 Jun	36.00%	40.00%	11.00%	3.00%	10.00% 23–24 Jun	36.00%	43.00%	9.00%	2.00%	10.00% 22–23 Jun	37.00%	42.00%	9.00%	2.00%	9.00% 21–22 Jun	36.00%	42.00%	9.00%	2.00%	11.00% 20–21 Jun	37.00%	42.00%	8.00%	2.00%	11.00% 19–20 Jun	37.00%	43.00%	9.00%	2.00%	9.00% 17–19 Jun	37.00%	39.00%	12.00%	2.00%	10.00% 16–17 Jun	37.00%	42.00%	10.00%	2.00%	9.00% 15–16 Jun	37.00%	43.00%	9.00%	2.00%	9.00% 15–16 Jun	37.00%	37.00%	11.00%	2.00%	13.00% 14–15 Jun	36.00%	42.00%	9.00%	3.00%	9.00% 13–14 Jun	37.00%	42.00%	10.00%	3.00%	8.00% 12–13 Jun	37.00%	42.00%	10.00%	3.00%	8.00% 10–11 Jun	39.00%	40.00%	9.00%	3.00%	8.00% 9–10 Jun	37.00%	42.00%	9.00%	3.00%	9.00% 8–9 Jun	37.00%	43.00%	8.00%	3.00%	9.00% 7–8 Jun	37.00%	42.00%	9.00%	2.00%	10.00% 6–7 Jun	36.00%	44.00%	8.00%	2.00%	10.00% 5–6 Jun	37.00%	43.00%	9.00%	2.00%	10.00% 2–3 Jun	37.00%	42.00%	9.00%	2.00%	10.00% 1–2 Jun	36.00%	42.00%	9.00%	2.00%	11.00% 31 May–1 Jun	39.00%	41.00%	9.00%	2.00%	10.00% 30–31 May	37.00%	42.00%	9.00%	2.00%	10.00% 27–29 May	37.00%	37.00%	12.00%	2.00%	12.00% 26–27 May	37.00%	43.00%	9.00%	2.00%	10.00% 25–26 May	37.00%	43.00%	8.00%	2.00%	10.00% 24–25 May	37.00%	41.00%	10.00%	2.00%	10.00% 23–24 May	38.00%	42.00%	9.00%	2.00%	9.00% 20–24 May	35.00%	42.00%	10.00%	2.00%	11.00% 22–23 May	38.00%	42.00%	10.00%	2.00%	8.00% 19–20 May	37.00%	42.00%	8.00%	2.00%	11.00% 18–19 May	38.00%	40.00%	10.00%	2.00%	10.00% 17–18 May	36.00%	42.00%	9.00%	2.00%	11.00% 16–17 May	39.00%	41.00%	9.00%	2.00%	10.00% 15–16 May	38.00%	41.00%	10.00%	2.00%	10.00% 12–13 May	36.00%	41.00%	9.00%	2.00%	11.00% 9–10 May	38.00%	40.00%	9.00%	2.00%	11.00% 8–9 May	38.00%	42.00%	8.00%	2.00%	10.00% 6–9 May	35.00%	38.00%	9.00%	7.00%	10.00% 6–8 May	37.00%	39.00%	11.00%	2.00%	11.00% 5–6 May	36.00%	41.00%	10.00%	2.00%	10.00% 4–5 May	37.00%	39.00%	10.00%	3.00%	10.00% 3–4 May	36.00%	40.00%	11.00%	3.00%	11.00% 2–3 May	37.00%	42.00%	10.00%	3.00%	8.00% 27–28 Apr	36.00%	41.00%	12.00%	3.00%	9.00% 26–27 Apr	36.00%	42.00%	10.00%	3.00%	9.00% 25–26 Apr	36.00%	41.00%	10.00%	3.00%	10.00% 20–21 Apr	36.00%	42.00%	10.00%	3.00%	9.00% 19–20 Apr	36.00%	43.00%	9.00%	3.00%	9.00% 18–19 Apr	36.00%	43.00%	9.00%	3.00%	9.00% 17–18 Apr	36.00%	42.00%	9.00%	3.00%	9.00% 15–17 Apr	40.00%	40.00%	9.00%	3.00%	8.00% 14–15 Apr	37.00%	41.00%	9.00%	3.00%	9.00% 13–14 Apr	36.00%	42.00%	10.00%	3.00%	9.00% 12–13 Apr	35.00%	44.00%	10.00%	4.00%	8.00% 11–12 Apr	37.00%	42.00%	9.00%	4.00%	7.00% 10–11 Apr	36.00%	42.00%	10.00%	5.00%	7.00% 8–11 Apr	31.00%	42.00%	11.00%	6.00%	11.00% 8–10 Apr	36.00%	40.00%	11.00%	4.00%	10.00% 7–8 Apr	36.00%	43.00%	9.00%	2.00%	10.00% 6–7 Apr	35.00%	44.00%	10.00%	2.00%	9.00% 5–6 Apr	36.00%	42.00%	10.00%	2.00%	10.00% 4–5 Apr	37.00%	42.00%	9.00%	2.00%	10.00% 3–4 Apr	37.00%	42.00%	9.00%	2.00%	9.00% 31 Mar–1 Apr	36.00%	42.00%	11.00%	3.00%	8.00% 30–31 Mar	35.00%	42.00%	10.00%	2.00%	11.00% 29–30 Mar	35.00%	45.00%	9.00%	2.00%	9.00% 28–29 Mar	36.00%	42.00%	10.00%	2.00%	9.00% 27–28 Mar	36.00%	44.00%	9.00%	2.00%	9.00% 25–27 Mar	35.00%	41.00%	13.00%	2.00%	9.00% 24–25 Mar	38.00%	41.00%	11.00%	2.00%	8.00% 23–24 Mar	37.00%	41.00%	11.00%	2.00%	9.00% 23–24 Mar	37.00%	36.00%	16.00%	2.00%	9.00% 22–23 Mar	36.00%	42.00%	10.00%	2.00%	9.00% 21–22 Mar	35.00%	42.00%	9.00%	2.00%	12.00% 18–21 Mar	32.00%	41.00%	10.00%	8.00%	10.00% 17–18 Mar	37.00%	43.00%	9.00%	2.00%	9.00% 16–17 Mar	35.00%	43.00%	10.00%	2.00%	10.00% 15–16 Mar	35.00%	43.00%	10.00%	2.00%	10.00% 14–15 Mar	35.00%	45.00%	9.00%	2.00%	9.00% 13–14 Mar	35.00%	44.00%	9.00%	3.00%	10.00% 11–13 Mar	37.00%	41.00%	10.00%	3.00%	10.00% 10–11 Mar	33.00%	44.00%	10.00%	3.00%	10.00% 9–10 Mar	34.00%	45.00%	9.00%	3.00%	9.00% 8–9 Mar	36.00%	42.00%	9.00%	3.00%	11.00% 7–8 Mar	36.00%	44.00%	10.00%	3.00%	8.00% 6–7 Mar	36.00%	42.00%	9.00%	3.00%	10.00% 4–6 Mar	35.00%	41.00%	11.00%	5.00%	11.00% 3–4 Mar	35.00%	43.00%	10.00%	7.00%	5.00% 3–4 Mar	33.00%	41.00%	10.00%	5.00%	10.00% 1–2 Mar	36.00%	41.00%	11.00%	3.00%	9.00% 28 Feb – 1 Mar	34.00%	43.00%	11.00%	3.00%	9.00% 27–28 Feb	36.00%	43.00%	10.00%	3.00%	8.00% 25–27 Feb	35.00%	39.00%	12.00%	3.00%	11.00% 24–25 Feb	36.00%	44.00%	10.00%	3.00%	7.00% 23–24 Feb	38.00%	42.00%	10.00%	3.00%	7.00% 22–23 Feb	36.00%	44.00%	11.00%	3.00%	6.00% 21–22 Feb	37.00%	43.00%	9.00%	3.00%	8.00% 20–21 Feb	36.00%	42.00%	11.00%	3.00%	8.00% 18–20 Feb	35.00%	38.00%	18.00%	3.00%	6.00% 18–20 Feb	33.00%	43.00%	13.00%	3.00%	8.00% 17–18 Feb	37.00%	41.00%	10.00%	3.00%	8.00% 16–17 Feb	36.00%	42.00%	10.00%	3.00%	8.00% 15–16 Feb	35.00%	45.00%	10.00%	3.00%	7.00% 14–15 Feb	37.00%	44.00%	10.00%	3.00%	7.00% 13–14 Feb	36.00%	44.00%	10.00%	3.00%	7.00% 10–11 Feb	35.00%	45.00%	9.00%	3.00%	8.00% 9–10 Feb	35.00%	44.00%	10.00%	3.00%	9.00% 9–10 Feb	36.00%	42.00%	11.00%	5.00%	8.00% 8–10 Feb	34.00%	40.00%	11.00%	6.00%	12.00% 8–9 Feb	36.00%	43.00%	10.00%	3.00%	8.00% 7–8 Feb	35.00%	43.00%	10.00%	3.00%	7.00% 6–7 Feb	37.00%	43.00%	9.00%	3.00%	7.00% 4–6 Feb	36.00%	39.00%	11.00%	3.00%	11.00% 3–4 Feb	36.00%	42.00%	10.00%	3.00%	8.00% 1–2 Feb	36.00%	44.00%	9.00%	3.00%	7.00% 31 Jan–1 Feb	39.00%	44.00%	8.00%	3.00%	7.00% 30–31 Jan	40.00%	42.00%	8.00%	3.00%	8.00% 28–30 Jan	34.00%	43.00%	10.00%	4.00%	9.00% 27–28 Jan	32.00%	43.00%	11.00%	4.00%	10.00% 27–28 Jan	39.00%	43.00%	8.00%	5.00%	5.00% 26–27 Jan	38.00%	44.00%	8.00%	3.00%	7.00% 25–26 Jan	39.00%	41.00%	10.00%	3.00%	8.00% 25–26 Jan	33.00%	41.00%	12.00%	6.00%	9.00% 24–25 Jan	37.00%	43.00%	10.00%	4.00%	6.00% 23–24 Jan	37.00%	42.00%	11.00%	4.00%	7.00% 21–24 Jan	33.00%	43.00%	13.00%	4.00%	7.00% 21–23 Jan	35.00%	39.00%	15.00%	2.00%	8.00% 20–21 Jan	39.00%	43.00%	9.00%	3.00%	7.00% 19–20 Jan	36.00%	43.00%	10.00%	3.00%	8.00% 18–19 Jan	37.00%	42.00%	9.00%	2.00%	8.00% 17–18 Jan	39.00%	44.00%	8.00%	2.00%	7.00% 16–17 Jan	37.00%	42.00%	9.00%	2.00%	9.00% 13–14 Jan	37.00%	43.00%	9.00%	2.00%	9.00% 12–13 Jan	36.00%	40.00%	10.00%	2.00%	12.00% 12–13 Jan	41.00%	41.00%	8.00%	2.00%	9.00% 11–12 Jan	36.00%	43.00%	9.00%	2.00%	9.00% 10–11 Jan	40.00%	41.00%	7.00%	2.00%	10.00% 8–10 Jan	40.00%	43.00%	8.00%	2.00%	7.00% 7–9 Jan	34.00%	42.00%	12.00%	2.00%	10.00% 6–7 Jan	35.00%	40.00%	12.00%	5.00%	8.00% 6–7 Jan	38.00%	41.00%	10.00%	5.00%	6.00% 5–6 Jan	39.00%	43.00%	7.00%	2.00%	8.00% 4–5 Jan	40.00%	41.00%	10.00%	2.00%	7.00% 3–4 Jan	40.00%	42.00%	8.00%	2.00%	8.00% 22–23 Dec	39.00%	41.00%	9.00%	2.00%	9.00% 21–22 Dec	41.00%	42.00%	8.00%	2.00%	8.00% 20–21 Dec	40.00%	42.00%	9.00%	2.00%	7.00% 19–20 Dec	40.00%	43.00%	8.00%	2.00%	7.00% 17–20 Dec	35.00%	41.00%	9.00%	5.00%	10.00% 16–19 Dec	37.00%	39.00%	13.00%	2.00%	9.00% 16–17 Dec	39.00%	42.00%	9.00%	2.00%	8.00% 15–16 Dec	37.00%	39.00%	11.00%	2.00%	11.00% 15–16 Dec	41.00%	41.00%	9.00%	2.00%	8.00% 14–15 Dec	42.00%	40.00%	8.00%	2.00%	7.00% 13–14 Dec	39.00%	42.00%	9.00%	2.00%	7.00% 12–13 Dec	41.00%	42.00%	9.00%	2.00%	6.00% 10–12 Dec	38.00%	39.00%	11.00%	4.00%	10.00% 9–10 Dec	40.00%	42.00%	9.00%	2.00%	8.00% 8–9 Dec	41.00%	39.00%	11.00%	5.00%	7.00% 7–8 Dec	41.00%	41.00%	8.00%	2.00%	8.00% 6–7 Dec	42.00%	39.00%	9.00%	2.00%	8.00% 5–6 Dec	42.00%	39.00%	10.00%	2.00%	7.00% 2–3 Dec	41.00%	39.00%	10.00%	2.00%	8.00% 1–2 Dec	40.00%	40.00%	11.00%	2.00%	7.00% 30 Nov–1 Dec	41.00%	38.00%	11.00%	2.00%	8.00% 29–30 Nov	35.00%	40.00%	13.00%	2.00%	10.00% 29–30 Nov	40.00%	40.00%	10.00%	4.00%	8.00% 28–29 Nov	40.00%	40.00%	10.00%	2.00%	8.00% 26–29 Nov	36.00%	40.00%	12.00%	2.00%	10.00% 25–26 Nov	40.00%	40.00%	9.00%	2.00%	9.00% 24–25 Nov	42.00%	39.00%	10.00%	2.00%	7.00% 23–24 Nov	40.00%	40.00%	9.00%	2.00%	9.00% 22–23 Nov	42.00%	40.00%	10.00%	2.00%	7.00% 21–22 Nov	41.00%	38.00%	11.00%	2.00%	8.00% 19–21 Nov	36.00%	38.00%	14.00%	3.00%	9.00% 18–19 Nov	40.00%	38.00%	11.00%	1.00%	10.00% 17–19 Nov	37.00%	38.00%	13.00%	2.00%	10.00% 17–18 Nov	40.00%	40.00%	11.00%	2.00%	7.00% 16–17 Nov	40.00%	40.00%	11.00%	2.00%	8.00% 15–16 Nov	37.00%	42.00%	10.00%	2.00%	8.00% 14–15 Nov	40.00%	42.00%	10.00%	2.00%	6.00% 12–14 Nov	36.00%	39.00%	14.00%	2.00%	9.00% 11–12 Nov	39.00%	41.00%	10.00%	2.00%	7.00% 10–11 Nov	40.00%	40.00%	10.00%	2.00%	8.00% 9–10 Nov	42.00%	37.00%	11.00%	1.00%	9.00% 8–9 Nov	40.00%	39.00%	13.00%	1.00%	7.00% 7–8 Nov	42.00%	39.00%	11.00%	1.00%	8.00% 4–5 Nov	40.00%	39.00%	12.00%	1.00%	9.00% 3–4 Nov	40.00%	39.00%	11.00%	1.00%	9.00% 2–3 Nov	40.00%	40.00%	9.00%	1.00%	9.00% 1–2 Nov	41.00%	40.00%	11.00%	1.00%	7.00% 31 Oct–1 Nov	41.00%	39.00%	11.00%	1.00%	8.00% 29–30 Oct	35.00%	37.00%	16.00%	3.00%	9.00% 28–29 Oct	42.00%	37.00%	13.00%	1.00%	7.00% 27–28 Oct	35.00%	37.00%	15.00%	3.00%	10.00% 27–28 Oct	41.00%	39.00%	11.00%	1.00%	8.00% 26–27 Oct	41.00%	39.00%	12.00%	1.00%	7.00% 25–26 Oct	40.00%	38.00%	12.00%	1.00%	8.00% 24–25 Oct	40.00%	40.00%	11.00%	1.00%	8.00% 22–24 Oct	39.00%	36.00%	16.00%	1.00%	8.00% 22–24 Oct	37.00%	38.00%	15.00%	3.00%	7.00% 21–22 Oct	41.00%	40.00%	10.00%	2.00%	7.00% 21–22 Oct	40.00%	36.00%	16.00%	2.00%	7.00% 21–22 Oct	35.00%	37.00%	10.00%	8.00%	10.00% 20–21 Oct	41.00%	40.00%	10.00%	2.00%	7.00% 19–20 Oct	41.00%	39.00%	11.00%	2.00%	7.00% 18–19 Oct	42.00%	39.00%	11.00%	1.00%	7.00% 17–18 Oct	41.00%	39.00%	12.00%	1.00%	6.00% 15–17 Oct	39.00%	36.00%	14.00%	3.00%	5.00% 14–15 Oct	41.00%	39.00%	11.00%	2.00%	7.00% 13–15 Oct	40.00%	34.00%	14.00%	4.00%	8.00% 13–14 Oct	42.00%	38.00%	12.00%	1.00%	7.00% 12–13 Oct	41.00%	40.00%	11.00%	1.00%	7.00% 11–12 Oct	43.00%	36.00%	12.00%	1.00%	7.00% 10–11 Oct	42.00%	38.00%	12.00%	1.00%	7.00% 7–8 Oct	42.00%	38.00%	12.00%	1.00%	7.00% 6–7 Oct	42.00%	38.00%	12.00%	1.00%	7.00% 5–6 Oct	42.00%	40.00%	11.00%	1.00%	6.00% 4–5 Oct	43.00%	39.00%	11.00%	1.00%	6.00% 3–4 Oct	41.00%	39.00%	12.00%	1.00%	7.00% 30 Sep–1 Oct	39.00%	41.00%	11.00%	1.00%	8.00% 30 Sep–1 Oct	35.00%	38.00%	16.00%	4.00%	7.00% 30 Sep–1 Oct	41.00%	37.00%	13.00%	2.00%	7.00% 29 Sep–1 Oct	39.00%	36.00%	15.00%	3.00%	7.00% 29–30 Sep	41.00%	39.00%	12.00%	1.00%	7.00% 28–29 Sep	41.00%	39.00%	12.00%	2.00%	7.00% 28–29 Sep	35.00%	37.00%	18.00%	3.00%	7.00% 27–28 Sep	41.00%	40.00%	12.00%	1.00%	7.00% 26–27 Sep	39.00%	40.00%	12.00%	1.00%	8.00% 23–24 Sep	39.00%	38.00%	15.00%	1.00%	8.00% 22–23 Sep	41.00%	37.00%	13.00%	1.00%	7.00% 21–22 Sep	43.00%	36.00%	14.00%	1.00%	6.00% 20–21 Sep	39.00%	39.00%	13.00%	1.00%	8.00% 19–20 Sep	42.00%	38.00%	11.00%	1.00%	8.00% 16–17 Sep	41.00%	39.00%	13.00%	1.00%	7.00% 14–16 Sep	37.00%	35.00%	15.00%	5.00%	8.00% 15–16 Sep	41.00%	38.00%	12.00%	1.00%	8.00% 14–15 Sep	42.00%	39.00%	12.00%	1.00%	6.00% 13–14 Sep	40.00%	39.00%	12.00%	1.00%	8.00% 12–13 Sep	41.00%	38.00%	12.00%	1.00%	8.00% 10–12 Sep	39.00%	37.00%	14.00%	2.00%	8.00% 10–12 Sep	37.00%	37.00%	15.00%	2.00%	9.00% 9–10 Sep	42.00%	38.00%	14.00%	1.00%	5.00% 8–9 Sep	42.00%	37.00%	14.00%	1.00%	7.00% 7–8 Sep	43.00%	38.00%	12.00%	1.00%	6.00% 6–7 Sep	42.00%	38.00%	13.00%	1.00%	7.00% 5–6 Sep	42.00%	37.00%	13.00%	1.00%	7.00% 3–5 Sep	38.00%	34.00%	18.00%	1.00%	9.00% 2–3 Sep	42.00%	37.00%	12.00%	1.00%	8.00% 31 Aug–1 Sep	43.00%	37.00%	12.00%	1.00%	7.00% 30–31 Aug	43.00%	38.00%	11.00%	1.00%	7.00% 26–27 Aug	41.00%	37.00%	13.00%	1.00%	7.00% 25–26 Aug	42.00%	37.00%	12.00%	1.00%	8.00% 24–25 Aug	42.00%	37.00%	12.00%	1.00%	7.00% 23–24 Aug	41.00%	38.00%	13.00%	1.00%	7.00% 22–23 Aug	41.00%	39.00%	12.00%	1.00%	8.00% 19–20 Aug	41.00%	38.00%	12.00%	1.00%	7.00% 18–19 Aug	41.00%	37.00%	14.00%	1.00%	7.00% 17–18 Aug	44.00%	36.00%	12.00%	1.00%	6.00% 16–17 Aug	42.00%	37.00%	14.00%	1.00%	7.00% 15–16 Aug	41.00%	37.00%	15.00%	1.00%	7.00% 13–15 Aug	37.00%	37.00%	18.00%	1.00%	7.00% 13–15 Aug	39.00%	33.00%	15.00%	5.00%	8.00% 12–13 Aug	42.00%	37.00%	13.00%	1.00%	6.00% 11–12 Aug	42.00%	37.00%	14.00%	1.00%	7.00% 10–11 Aug	41.00%	37.00%	15.00%	1.00%	7.00% 9–10 Aug	42.00%	38.00%	14.00%	1.00%	6.00% 8–9 Aug	40.00%	36.00%	15.00%	1.00%	8.00% 6–8 Aug	39.00%	33.00%	16.00%	5.00%	8.00% 5–6 Aug	42.00%	36.00%	13.00%	1.00%	7.00% 3–4 Aug	42.00%	36.00%	13.00%	1.00%	8.00% 2–3 Aug	41.00%	36.00%	13.00%	1.00%	8.00% 1–2 Aug	42.00%	38.00%	12.00%	1.00%	7.00% 29–30 July	42.00%	38.00%	12.00%	1.00%	6.00% 27–28 July	42.00%	36.00%	14.00%	1.00%	7.00% 26–27 July	42.00%	37.00%	14.00%	1.00%	6.00% 25–26 July	42.00%	35.00%	15.00%	1.00%	7.00% 23–25 July	38.00%	34.00%	19.00%	1.00%	7.00% 23–25 July	40.00%	38.00%	14.00%	1.00%	7.00% 22–23 July	41.00%	36.00%	14.00%	1.00%	8.00% 21–22 July	43.00%	35.00%	15.00%	1.00%	6.00% 20–21 July	44.00%	35.00%	13.00%	1.00%	6.00% 19–20 July	43.00%	35.00%	14.00%	1.00%	7.00% 18–19 July	42.00%	35.00%	15.00%	1.00%	7.00% 15–16 July	40.00%	37.00%	15.00%	1.00%	6.00% 14–15 July	43.00%	34.00%	15.00%	1.00%	7.00% 13–14 July	43.00%	34.00%	15.00%	1.00%	7.00% 12–13 July	42.00%	35.00%	15.00%	1.00%	7.00% 11–12 July	42.00%	35.00%	15.00%	1.00%	8.00% 8–9 July	42.00%	34.00%	17.00%	1.00%	6.00% 7–8 July	42.00%	35.00%	16.00%	1.00%	6.00% 6–7 July	40.00%	36.00%	17.00%	1.00%	6.00% 5–6 July	41.00%	35.00%	16.00%	1.00%	8.00% 5–6 July	41.00%	36.00%	15.00%	1.00%	8.00% 4–5 July	40.00%	36.00%	16.00%	1.00%	7.00% 25–27 June	40.00%	31.00%	18.00%	4.00%	7.00% 24–25 June	43.00%	36.00%	16.00%	1.00%	5.00% 23–24 June	41.00%	35.00%	16.00%	2.00%	6.00% 22–23 June	42.00%	34.00%	17.00%	1.00%	6.00% 22–23 June	39.00%	33.00%	18.00%	3.00%	7.00% 21–22 June	41.00%	37.00%	15.00%	1.00%	6.00% 20–21 June	41.00%	33.00%	18.00%	1.00%	6.00% 18–20 June	39.00%	31.00%	19.00%	2.00%	10.00% 18–20 June	39.00%	31.00%	21.00%	1.00%	7.00% 17–18 June	39.00%	34.00%	19.00%	1.00%	7.00% 16–17 June	36.00%	30.00%	23.00%	4.00%	7.00% 10–11 June	40.00%	32.00%	18.00%	1.00%	8.00% 10–11 June	39.00%	32.00%	19.00%	3.00%	7.00% 1–9 June	36.00%	30.00%	25.00%	1.00%	8.00% 28–31 May	37.00%	33.00%	21.00%	1.00%	8.00% 21–23 May	39.00%	32.00%	21.00%	2.00%	6.00% 20–21 May	39.00%	32.00%	21.00%	1.00%	6.00% 13–14 May	37.00%	34.00%	21.00%	1.00%	7.00% 12–13 May	38.00%	34.00%	21.00%	1.00%	6.00%

Do Not Edit This Data!



Hope this helps. Please do not edit this subsection, Thank you.Sheffno1gunner (talk) 06:31, 11 January 2013 (UTC)

Resume Discussion
Thank you. I think that's a definite help in seeing the trend line over the course of time between the last election and the present day. Circumspect (talk) 13:14, 11 January 2013 (UTC)

The dates labeling the x-axis of the chart appear to be a bit misaligned. Is there anything that can be done to make the x-axis labeling clearer? Circumspect (talk) 13:18, 11 January 2013 (UTC)

Thank you, like I say it was a considerable effort! I had noticed the misaligned labeling of the x-axis. Unfortunately there is nothing I can do to change this in the program I have used. However I have asked User:Wavehunter (the creator of the graph for 2012) to use the data I have copied and pasted to this page to extend their graph. Sheffno1gunner (talk) 13:54, 11 January 2013 (UTC)

It is great that you have done this! You have solved the graphing/data problem. We now have the ability to create any kind of graph we decide upon using ALL the data! I hope that user:Wavehunter does decide to add your data to his/her chart because we did fairly unanimously decide that their graph was the best! — Preceding unsigned comment added by 217.41.32.3 (talk) 14:11, 11 January 2013 (UTC)

Wow, someone has been busy, great work! Personally I would favour Wavehunter's graph to be adapted to include 2010 and 2011, we did agree it was the best. However if not/in the mean time this graph is the best we have and is good enough. Yes I am also concerned about the labeling of the x-axis but this is a relatively small issue in the grand scheme of things. Nick  ' 14:55, 11 January 2013 (UTC)

Sorry. I've been ridiculously busy. I had never expected to do more than 2012, but I will try to backdate graph today. Certainly Sheffno1gunner has put a lot of work in. I can't promise to be able to continue this until the next general election, or even until February. I tried to upload the graph on Wikimedia but it wouldn't take. If anyone can suggest a way for me to transfer the file (ODS - Open Office Spreadsheet) to other volunteers, please suggest away! --Wavehunter (talk) 12:50, 12 January 2013 (UTC)

Great Stuff Wavehunter that would be fantastic! Could you make sure you include the few polls we have for 2013 as well, we might as well include as much data as possible at the point of every update, don't know if you want to wait til tomorrow for the 1st major stream of Sunday Newspaper polls (I expect there will be 4 or 5), if you can't wait then that's also fine. It's up to you, just make sure you include what's there. Also could you change the names on the key to "Conservative", "Labour", "Lib Dem", "UKIP" and "Other". I think it would also be better with a title, similar to my graph, how about "Opinion Polling for the 2015 United Kingdom General Election". Many Thanks Sheffno1gunner (talk) 13:48, 12 January 2013 (UTC)

I believe party colors in the table should reflect those shown in the article, as Sheffno1gunner's graph does, but about the style, Wavehunter's graph looks nicer. However, once 2010 and 2011 data is included within Wavehunter's graph, dots will look mixed with the trend line because of the increse of data and because they use a similar color. Maybe dots could use a lighter shade of the color (this can be achieved by giving them more transparency). Maybe something like this:



All data from 2010 to 2013 has been included, in order to show how would the table look in its entirety. Impru20 (talk) 16:40, 12 January 2013 (UTC)

I don't know that mine's any better than Impru20's, but I've stuck it in anyway. It may be worth noting that this covers GB only: Northern Ireland parties are excluded. I went up to 7 January as a seven-point moving average (the way I've calculated it) requires three polls ahead and three polls behind. --Wavehunter (talk) 17:05, 12 January 2013 (UTC)

Great stuff guys! I think we should keep Wavehunter's chart for now because his dots are clearer. However there is going to come a point whereby we can't extend the width of the chart any further and as it gets condensed the dots will join together. When that point comes, I suggest we revert to Impru20's chart (obviously with the dots removed. It's clear that that time is fast approaching! I suggest that we not update this chart again until after the May Local Election results have come in. It seems a logical time to update! I think at that point we will have to use Impru20's graph on a permanent basis.Sheffno1gunner (talk) 17:31, 12 January 2013 (UTC)

Yea, the 3 of you have now made this article complete! Well done ✅

I agree with Sheffno1gunner, keep Wavehunter's graph for now. Ask both of them to do another update in May and then we pick the best (dependent on visibility of the plotted dots!). If it's Impru20's graph then the usefulness of Wavehunter's graph will have run it's course.217.41.32.3 (talk) 17:45, 12 January 2013 (UTC)

Graph to show tie for 3rd Place
I have chopped Wavehunter's old graph in half to produce:

I have added a description with detail, it is well sourced and written neutrally. The purpose of it is to give a more detailed explanation as to what is actually happening with regard to third place in the polls.Sheffno1gunner (talk) 23:18, 13 January 2013 (UTC)

✅I took the liberty of putting in all your sources for you!217.41.32.3 (talk) 12:38, 14 January 2013 (UTC)

Will editors please refrain from removing unsourced material! I have removed one sentence and one repeated word "throughout" that another editor disagreed with and thought was disproportionate. I agree with his reasoning. This discussion took place on Talk:UK Independence Party. It seems to me that the removal of sourced material is vandelism!Sheffno1gunner (talk) 19:38, 16 January 2013 (UTC)


 * Editors will always remove unsourced material because things should be sourced. However, Bondegezou made edits because he felt the sources did not support the claims, and were therefore original research (take it up with him, not me). However, please don't mislabel edits as vandalism. His edits certainly were not vandalism because they were well-intentioned. Your claim that "Simply deleting sourced material without discussion is vandelism [sic]! is untrue. He had good reason to remove them. The discussion should take place now. –  Richard  BB  11:41, 17 January 2013 (UTC)

Ok, fair enough, sloppy use of terminology, I accept that as fault on my part and apologise unreservedly! Very Sorry. While I accept that Bondegezou's deletions might well have been made with the best of intentions, I do not feel it is right to remove well sourced material from a page without even consulting the talk page. Their have been full comments on this talk page since before the edits were made. Another user who has being assisting me has added sources for me as well as you will see from the above commentary! There is no evidence to suggest that Bondegezou has consulted with this page before reverting edits. There is certainly no comment to indicate that Bondegezou has done so. I would have thought it standard practice to look at the talk page before reverting anyone's edits. I always look at the talk page before making an edit, never mind deleting someonelse's comments.

Bondegezou if you have a problem with the content please raise your concerns on this page please. These edits have also been made on another page there is also in depth discussion as to what is and is not apropriate to include on that talk page. What I have done in my last edits is to remove a sentence thaqt others disaproved of because it was a bit of a stretch of the use of a source to say "this implies a statistical tie", it was also an over elaboration! the word throughout has also been removed by concensus on that page. This information and graph are relevnt to both pages for some quite obvious reasons, it is noty unusual to repeat the same graph on more page than one. It does however mean that the text needs to be a bit different (as well as the links). I have seen to this. I think I have made every effort to ensure that the wording of these edits are impartial and neutral and reflect the factual content of the sources. I hope you agree, if not please raise each specific concern individually.

Please note, I am not able to sign into wikipedia as often as I'd like because of my work pressures, so please be patient when it comes to waiting for a reply!Sheffno1gunner (talk) 17:50, 17 January 2013 (UTC)
 * It is absolutely central to Wikipedia policy that material needs to be verifiable, which means it can be directly supported by reliable sources. One consequence of that is that original research is not allowed. The text I have removed consists of your interpretation of the polling evidence. I'm not saying your interpretation is right or wrong: I'm saying it is your interpretation. You have given no cites to support that interpretation. The cites you give explain basic terms (like "margin of error") or provide raw data. The WP:OR policy is very clear and I urge you to read it.
 * There has clearly been interest in UKIP's poll surge that has led to several pages being edited. It is appropriate that Wikipedia reflects that poll surge, but we have to follow policy, which means we have to follow reliable sources. Find reliable sources interpreting the polling data and discussing which is the 3rd party of UK politics today and we can include them. I've added some citations on the topic, although they're more general and don't specifically discuss the polls much.
 * However, editors must refrain from interpreting primary evidence themselves. Similarly, and as I've said, the "third party lead" stuff above seems to me inappropriate and reflecting editors' interpretations rather than reliable sources.
 * Nor does it matter how many editors support a particular piece of text if it is against policy. We all have to follow policy. Three of you together not wanting to follow policy is no different from one editor not wanting to follow policy.
 * I also removed a sentence citing a Guardian article because I read the Guardian article and it simply does not support the text to which it is attached.
 * Sheffno, you've made many great contributions to Wikipedia, so I really do urge you to review core Wikipedia policy and understand the point I'm making. Bondegezou (talk) 20:49, 17 January 2013 (UTC)

I understand the point you are making but i disagree with your interpretation of original research and interpretation of policy. All I have done is stated facts that are verrified by sources! Why are verifiable facts not as note worthy as an opinion piece written by a journalist. I shall go through my piece point by point to show you that this is not my own analysis in any way, it is simply a description of what is shown in sourced material.

1. You are right to say that those 2 sources (The Economist and Nick Moon) only outline basic terms. I am only using them to support that 3% is the standard margin of error, a number or leading accademics, including Nick Moon say this. So that's fact number one.

2. the sources I'm using for showing that UKIP have been within this standard margin of error of the Lib Dems is another Wikipedia page which illustrates everything. However, I realise that this is not enough, so I added a BBC link to illustrate this. I will gladly copy this reference in so that it is at the end of both sentences of which it is relevent for.

There is no way you can dispute this to be anything other than fact! Or atleast it is no less of a fact then anything written in The Guardian or the Daily Mail! 1 it is backed up by research by leading accademics, 2 it is the simple reporting of cold hard facts, there is no way of arguing that they are not within that 3% figure, it is demonstrated in the BBC's table and Wikipedia's table which is also shown in graph form! to say this is not individual research, I'm simply saying what is already reported by the sources!Sheffno1gunner (talk) 15:08, 18 January 2013 (UTC)
 * Wikipedia policy is very clear: we only include material that is true and verifiable. So, simply claiming something is "fact" is never enough. Something can be 100% factually true, but if reliable sources don't say it, it doesn't go in Wikipedia. You may feel this isn't the best way for Wikipedia to work, but that is current policy and we work within it.
 * You have given sources explaining certain terms (e.g. "margin of error"). You have then given sources to some raw data (polling figures). You have then done an analysis of the data in which you have drawn conclusions. That is original research; I really don't see any way around that.
 * You seem to be arguing, in part, that your conclusion is so obvious from the table that it doesn't constitute OR. If it's that obvious, then it doesn't need to be said. The reader can see the table, they can see how the two parties under discussion relate to each other. They can draw their own conclusions. Or, if it's that obvious, then it shouldn't be difficult to find a reliable source saying the same thing, thus solving the problem. But we need to avoid interpreting data ourselves.
 * I don't think it is appropriate for us to be drawing conclusions without specific citations supporting those conclusions. And I see from Talk:UK_Independence_Party that other editors agree with me. I will also put a note on WP:NORN requesting further input.
 * In addition to the above, I also question whether your interpretation is fact. I discussed this on a previous Talk page discussion: see my final comment at Talk:United_Kingdom_local_elections,_2013. Your use of "margin of error" is too simple. If we have multiple polls in a reasonably short period of time, we can combine them and reduce the margin of error. So, if you have a situation where lots of polls put party A 2% ahead of party B, it is wrong to say that party A and party B are within the margin of error. Given so many polls, you could reliably conclude that party A is above party B. Bondegezou (talk) 16:32, 18 January 2013 (UTC)

Have reverted this part of the edit as it is sourced. However i think you may have a point about what The Guardian source does and doesn't say with regard to UKIP's claim of being 3rd place. What do you suggest instead, personally I support that bottom paragraph that you have written, do you feel that anything more needs adding to it?Sheffno1gunner (talk) 15:29, 18 January 2013 (UTC)

I mean, yea it is blatantly obvious. If you look at the graph, you can quite clearly see that the 2 parties are within 3% of each other since April. You say that this is so obvious that it doesn't need stating but really I don't see what harm it is doing, I don't see how it can be constituted as original research given all my sourcing and given the graph. If you look at the graph it is blatantly obvious that they are within 3%. Answer me this though, would most readers know what the standard margin of error of an opinion poll was? If you asked 10members of the public in the street, how many would get the right answer? You'd be lucky to find one I bet! This is what makes the information useful to readers, so that they are able to understand the significance of it. I get what your saying that we shouldn't have to state the blinking obvious to people, i.e. as you put it "when something is so obvious it doesn't need stating" but at the same time Wikipedia is an encyclopedia and is meant to inform it's readers! By not including this we would be leaving our readers totally in the dark about what this means, they would have less information as to what this means when in reality we could provide well sourced information but choose not to! Notice that LD were ahead most of the time before December but since December UKIP have been ahead more often than LD, this could lead readers to think that UKIP have overtaken the Liberal Democrats. The graph also reflects this. Instead of leaving the reader to assume that LD were 3rd but now UKIP have overtaken them is not the right thing to do. We need to provide as much sourced info as possible and that is what I have attempted to do! By all means lets add more sources, I'm never against that!

On a seperate note: As for your argument about the Lib Dem's being ahead in most of the polls: Because of the margin of error it is irrelevent who is ahead! That's what the margin of error is, you seem to misunderstand the concept, this is why I put a link in (as well as the sources) for readers and editors. If in every single poll since April 2012 UKIP were 3% behind the Lib Dems, that would still mean that they have been within the margin of error for 9months! The reality is it's been a little bit more neck and neck then that but never the less all we can explicitly say in polling terms without any analysis is that they have been within the margin of error. I accept that we can not say that they are in a statistical polling tie, that would of course be analysis (an interpretation of what the data means/might mean). What I have done is simply stated 2 sourced facts and made a graph: 1)here are your polling numbers, 2)here's your definition of standard margin of error and 3)here's your graph showing UKIP/LD within 3% of one another.

However, as with any Wikipedia page, extra sources are always a bonus, so I suggest that we set to work with looking for more sources but as far as I am concerned I see no violation, I have just clearly stated what is already there, I haven't drawn any conclusions from it!Sheffno1gunner (talk) 19:09, 18 January 2013 (UTC)

You have to admit he has a point there! I didn't know that 3% was a standard margin of error until I'd looked it up, if I had to guess I'd have probably said 1% and been wrong! Also that is a point, in not stating the margin of error we are in effect showing that UKIP are the 3rd most popular party. I'm happy to go with that and let readers conclude that UKIP have overtaken the Lib Dems but it seems that if we're to paint a true picture we need to mention the margin of error and state that both parties are within margin of each other! It's all sourced for, I know because I helped with the sourcing!81.133.12.45 (talk) 22:20, 18 January 2013 (UTC)


 * I humbly suggest both of you have misunderstood margin of error. I said earlier that my objection to Sheffno's interpretation was that is constitutes original research. I will now explain why it is not merely inappropriate for inclusion because it is OR, but because it is wrong as well.
 * Margin of error is a familiar term in opinion polling, but often a misunderstood one. Statisticians usually talk about a related concept, the confidence interval. A poll is based on a sample from a population: we use the proportion in the poll supporting party X (or whatever) to estimate the proportion doing so in the population. But that estimate won't be spot on because of sampling error. Assuming we have sampled well (meaning randomly), the sample proportion is an estimate of the population proportion, with the confidence interval/margin of error describing the precision of that estimate. A 95% confidence interval is defined such that the confidence interval for the sample estimate will include the true (population) value 95% of the time. (We can also calculate 99% confidence intervals, or 80% confidence intervals, or whatever, but we usually just use 95%.) The margin of error is then half the confidence interval, so a value may be reported as 40% ± 3% (margin of error), meaning the confidence interval is 37%-43%. This means the true value could be outside that range, but usually (95% of the time), the interval will capture the true value. The only problem is we can't tell from a single poll whether it's one where we were unlucky and the confidence interval doesn't capture the true result.
 * The precise calculation of confidence intervals is a large topic. However, if we presume a proper sampling technique, for large samples, of large populations, the margin of error for a proportion can be easily calculated: see margin of error for details.
 * The margin of error and confidence interval vary with the sample size. The details depend on what you are doing, but generally the margin of error is inversely proportional to the square root of the sample size. (Thus, to get half the margin of error, you need four times the sample size.) Most political opinion polls are done on a similar number of people (so as to achieve a suitably small margin of error), but some are bigger or smaller, so you need to see what margin of error is reported for each poll, or do the maths yourself.
 * The margin of error for a proportion, like vote share, also depends on the vote share. We get better estimates of proportions near 0% and 100% than in the middle. The margin of error is smaller for a party with a vote share of 10% (or 90%) than one on 50%. The margin of error quoted is, thus, usually the maximum margin of error at 50%.
 * So, consider the following examples. There is a poll reported in the article by Survation done on 5 Jan 2013 with a sample of 790, and another by YouGov done on 5-6 Oct 2011 of 2723 (link broken). We can calculate the maximum margin of error for each of these: for the Survation poll, it is 3.49%, but just 1.88% for the YouGov poll. Remember: not all polls have the same margin of error.
 * Now let's consider the 21-2 Dec 2011 YouGov poll of 1721 people . This puts the Conservatives on 40% and the Liberal Democrats on 9%. I note the YouGov pdf doesn't even give the margin of error, but we can work it out. Thus, the maximum margin of error is 2.36%, but the margin of error for the Conservative vote is 2.31% and for the Liberal Democrat vote, a mere 1.42%.
 * With me so far? Good. Because it gets more complicated. Often, we're interested in the difference between two parties, e.g. the lead Labour may have over the Conservatives. We could look at the margin of error for each party, but the margin of error for the difference requires a separate calculation: see Margin of error. So, let's take the 18-9 Dec 2011 YouGov poll, again on 1721 people, so we know the maximum margin of error is 2.36%. The Conservatives are on 38% and Labour are on 42%, so the difference between them is 4%, larger than the maximum margin of error. But the margin of error of the difference between the two parties is 4.22%, and our 4% difference is smaller than that.
 * Often, we have more than one poll, at which point we can consider combining multiple polls. When we do that, we're effectively making the sample size bigger, so the margin of error falls. If we have, say, 5 polls within a few days and they all show two parties within the margin of error, the combination of all 5, by reducing the margin of error, may show the two parties as being significantly different. See and  for details.
 * This all, of course, presumes the polling was done well. The margin of error only considers the sampling error under certain assumptions. In practice, polls are usually less accurate because of methodological challenges in doing them right: see for discussion. Some polling strategies appear to be generally more reliable than others. Thus, in the UK, phone polls appear to perform better than online polls.
 * So, unless you have done the maths, or a reliable source has done it for you, for each poll, taking into account sample size, the UKIP vote share and the LibDem vote share, you don't know whether their vote shares are within the margin of error or not. Which is why your interpretation of the polling data is wrong.
 * But even if you were right, it's your analysis, so it clearly falls foul of WP:OR anyway. Bondegezou (talk) 17:58, 19 January 2013 (UTC)

Cool, that's fine by me. I'm happy to leave it for the readers to assume that UKIP have already overtaken the Lib Dem's. I guess that's what the actual polling numbers point at, so let's leave it at that!81.133.12.45 (talk) 18:59, 19 January 2013 (UTC)

I'm still not happy with this but it looks like I've got to accept it! I feel that we are leading people into a false and premature conclusion. The evidence we have presented makes it look to the untrained eye that UKIP have taken the lead over the Lib Dem's already! Oh well, I tried to give as much info as possible. I guess it's for readers to make the obvious conclusion we have left! I reluctantly accept this final decision and consider the matter closed!Sheffno1gunner (talk) 06:49, 20 January 2013 (UTC)

Compile all Charts (old & current)under one Thumbnail
Basically because of the recent changes it has meant that we have had a lot of upoaded images/files displaying opinion polling all of which (except arguably 2012) replace their preceeding chart. It would be good if we could compile them all under one file: i.e. the as thumbnails within the file history in chronological order and not for them to be as separate pages. Chronological orderis as follows:
 * Opinion Polling Chart for the 2015 UK General Election.png (or deleted- this was error, line was too thick), replaced by:
 * Opinion Polling Chart for the 2015 UK general election.png
 * UK opinion polling 2010-2015.png
 * GB Polling May 2010 to Jan 2013.jpg (current)

I have no idea how to do this. I have asked in the Teahouse and I just got a load of waffle that answered a different question to the one I was actually asking! Something to do with grouping that was not the same as this!

Can someone please do this, I would be very much obliged!Sheffno1gunner (talk) 00:02, 19 January 2013 (UTC)