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

RfC: Should the Green Party, along with other parties be included in the table of polling results

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

As from the several sections above there has been some debate and discussion over whether the Green party along with others should be added to the table of polling results as a 'main party' (i.e. not listed under the 'others' section). Both sides feel as though Wikipedia policies and guidelines support their side and I feel as though we need some input from an outside, impartial source in order to come to a valid conclusion as neither side wishes to budge and there seems to be no in-between option for us to come to a compromise we are both happy with. CH7i5 (talk) 06:24, 9 July 2014 (UTC)

Just to point out that only one of those who voted in favour actually quoted a policy/made some form of policy based argument. One was purely just a non policy based "opinion" and the other largely opinion with no real point of policy. The other referred to policy but could not mount any challenge to the policy based points put against his argument. Owl In The House (talk) 22:18, 9 July 2014 (UTC)

Just wanted to note that I wrote "all parties" in the "consensus vote", which would also indicate including respect/bnp. (Though, respect are rarely in the tables, and BNP does not have seats, so there would be a clear cutoff there if that was the point.) I will admit I wasn't very clear on this though. Øln (talk) 17:59, 10 July 2014 (UTC)

Okay, so after reading WP:PRIMARY again it seems quite clear that data from primary sources should not be synthesized, which I would interpret as meaning that either all parties have to be included , or we have to use the secondary sources and include only what's reported there, so I agree that only including the greens, which some has suggested, may not follow the guidelines (unless they happen to be reported in the secondary source). Whether lumping greens and others together if they are reported would violate WP:SYN seems a bit unclear to me though. Otherwise is a question of weight, where there seems to be a disagreement. WP:PRIMARY also seems to suggest that if the article should only list the parties reported in news articles(secondary sources), those have to be cited, something which is not done here. Øln (talk) 21:05, 11 July 2014 (UTC)

1. Respect and BNP are always in the YouGov, Opinium tables and do also appear in other tables. Lets not forget that over 50% of this data is from YouGov as they do the daily polls. The BNP are always included in the Populus, ICM and Ipsos Mori tables (as well as Lord Ascroft - though he is not BPC registered) and Survation put the Greens with the BNP and TUSC. So no, there is not a "clear cut off point" that you suggest as the BNP are always included in the data and are always reflected in the same way as the Greens, SNP and Plaid. Also the majority of the time Respect are represented too. Your "cut off point" is selective and violates the policy point you have tried to make.

2. You falsely make this point about synthesising Primary Sources. We are not the one's doing the synthesising here, the pollsters are doing it themselves by officially publishing a headline figure. They also publish full data tables but do we publish everything in those data tables? (i.e. excluding/including don't knows, excluding/including refuseds, adjustments for likelihood to vote etc etc) No, we don't. The only reason we know what data to use out of all of the data published in the tables is because the pollsters publish a headline figure. It is not for us to then delve into the data and be selective as to what should be included in that headline figure. As far as Primary sources go, surely the pollster is the primary source and they provide 2 sources, the headline/summary and then they have a link to the full data for those who want to delve further in. I don't see why we should deliberately seek to cover our reliable sources in a way they seem to state is not appropriate. By Doctoring those headline figures and being selective about the data we do and don't include in our summary table is a blatant demonstration of original research and that is a clear breach of Wiki policy. To directly answer your question on WP:SYN, it is blatantly clear that it is not a violation of this policy because it is a format that the primary sources have published.

This is getting tedious. Owl In The House (talk) 08:51, 12 July 2014 (UTC)

Firstly, I already removed the part about cutoff long before you replied, so I don't know why you are complaining about that. I am not suggesting selectively doctoring the headline figures. I don't see how listing the full list of parties from the tables would constitute OR, rather, the question would be about whether it would be undue weight to include everything. As for the second part, I didn't see that yougov published headline figures before now, as the yougov page kept redirecting me to the page for my country rather than UK, but I managed to access the UK page now and you are correct there. Ashcroft and opinium does list more than the big 4 in their poll articles though, opinium even has an "others" table which list UKIP, greens and more(Ashcroft does have a summary table that lists greens as a note under "others", couldn't find an "others" aggregation excluding UKIP on opinium though.). Though seing as the headlines are used, shouldn't the article link to the them then as opposed to the full tables, since that is what's being used here?Øln (talk) 10:32, 12 July 2014 (UTC) (Also, I made a mistake by writing "secondary sources", when I should have been referring to "headlines", so sorry for the confusion there.)Øln (talk) 12:05, 12 July 2014 (UTC)

In either case, the point of this section was to get an outside opinion, and my reply was just to clarify what I meant, as I felt I was being misquoted, the issues have already been discussed in detail. Øln (talk) 11:07, 12 July 2014 (UTC)

I'd ask how the Polls generally do this?Serialjoepsycho (talk) 01:20, 12 July 2014 (UTC)

Generally polls report their 'headline figures' (Cons, Lab, LibDems and UKIP) on the main page and then you have to open separately and go into the polling results and look through the information to find the results for each individual party. It is the headline figure that we normally use. CH7i5 (talk) 07:37, 12 July 2014 (UTC)

That's right, apart from one thing. It's always the headline figure that gets used. There is not a single piece of data in those data tables that is not a headline figure. Owl In The House (talk) 08:59, 12 July 2014 (UTC)

Yougov doesn't actually list "others" in the headline, so that data isn't explicitly stated in the headline at least. As per WP:CALC doing simple subtraction to get an others number may be acceptable, though could be misleading due to rounding errors. In this case you are actually using data that is not from the headlines to be able to state that the percentage left is in fact "other parties", rather than e.g "don't know".Øln (talk) 11:52, 12 July 2014 (UTC) Forget that, they actually have a list which includes others.Øln (talk) 12:19, 12 July 2014 (UTC)

Honestly this seems very much to be a political dispute. From what I have displayed before me it seems listing the Green Party as other is standard not only on wikipedia but in the polls. I have to question if this is an attempt to further legitimize the Green Party. I feel they can do this without help from wikipedia. I can not recommend changing the status quo at the moment.Serialjoepsycho (talk) 20:53, 12 July 2014 (UTC)

Thanks for taking a look at this. Normally things like this can be nipped in the bud pretty easily but the Green activism on Wikipedia has been relentless over the last few months. I'm pleased to say it seems to have calmed down now. Owl In The House (talk) 11:48, 19 July 2014 (UTC)

• Comment Looking at the individual poll types and what they separate out in their first table:

Polling agent	Conservatives	Labour	Lib Dems	SNP	Plaid	UKIP	Greens	BNP	Respect	Notes
Populus	Y	Y	Y	N	N	Y	N	N	N
YouGov	Y	Y	Y	Combined		Y	Y	Y	Y	Main three listed first
Opinium	Y	Y	Y	Y	Y	Y	Y	Y	N	Main three listed first
Ipsos Mori	Y	Y	Y	Y	Y	Y	Y	Y	N	Main three listed first
ICM	Y	Y	Y	Y	Y	Y	Y	Y	N	Main three listed first
Ashcroft	Y	Y	Y	N	N	Y	N	N	N
Comres	Y	Y	Y	N	N	Y	N	N	N
TNS	Y	Y	Y	Combined		Y	N	N	N

• Personally, as this is supposed to be an article about opinion polling, and therefore more detailed than a table on the election article, I think the Greens should be included, alongside the SNP, BNP and Plaid Cymru. The table would easily be able to accommodate it, and I don't see the problem with including more information. Number 57 22:30, 18 August 2014 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

I'm pleased to say that this issue was resolved long ago but since you mention it Number your table is inaccurate as it is not a table of how polls are represented, you've done what everyone else has done and picked the part of the data you want to, not the part the pollsters themselves highlight (in their headline poll). But lets just say it was an accurate table reflective of reality (which it isn't), it would still point to the same conclusion, the majority of the time and in the overwhelming majority of all the polls included in 4 and a half years of data there are 4 parties highlighted, not 5, not 6, not 7, not 8 but 4. The format we have now is the best of all worlds because it is reflective of reality and how reliable sources publish things and it also enables the reader to get as much information as they could possibly want as all the polls have clickable links. Any other format would require us to deliberately ignore wiki policies...all this has already been gone over in the discussion above. Owl In The House (talk) 22:21, 26 August 2014 (UTC)

I did not pick the part of the data I wanted to, I used the first page of every poll document (I don't really care about this subject (I don't believe I have ever edited this article), and was only offering an opinion as someone started an RFC - your assumption that I have some kind of motive is as unseemly as your patronising tone). It is quite clear that there are four main parties included in the front page results for every poll, but I really don't understand the problem of including as much information as possible in the table, as it can clearly be expanded to cover it. As the closer pointed out, a table showing the other parties' results as well would be perfectly acceptable. Number 57 22:28, 26 August 2014 (UTC)

I make no assumptions about who you are etc and if my tone offends you then sorry but that is not anything against you it is a combination of tiredness and being fed up by this discussion still continuing, I have no preconceptions about yourself. I don't care to go over details that have already been gone over in detail above but the reason this format has been settled on is all outlined above and specific wikipedia policies cited and referenced. While I accept that you have picked a piece of the data consistently, it is not the piece of data the pollsters themselves highlight and it is not the way the pollsters themselves highlight it. The article's current format does that. There is nothing to say that the first table in each data set is the right one to use, we know what to use through the pollsters summary. Anyway short of endlessly repeating what myself and others have said, I think that answers your point with brevity. I don't want to come across as rude or anything, I don't know you, never even come across you on here, I have no preconceptions about your view on this or who you are as a person, if it came across the wrong way, then I'm sorry, I'm just tired and fed up of having to continue to return to issues that had previously been settled. Anyway wish you well, goodnight :) Owl In The House (talk) 22:40, 26 August 2014 (UTC)

In democracy all parties should have similar amounts of publicity, otherwise it is a self-fullfilling profecy and enforced barrier to election. With exception of those parties getting 100 votes in constituency for example, lets regard every national party which has candidates in most constituencies, an MP or highish vote share. I do not think just potential to win or form a large part of the house is the only criteria, the potential for them to be voted for (reasonable amount of constituents) should be.