Talk:Opinion polling for the 2015 Spanish general election

"Others"

Why "Others" columns were removed?--MiguelMadeira (talk) 10:39, 4 November 2015 (UTC)

Because of lack of space and because it was not consistent (since the number of parties in the "Others" section varied depending on the poll). I prefered to use that space to show more useful info such as the MoE and Sample size instead. Impru20 (talk) 11:03, 4 November 2015 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified 3 external links on Opinion polling for the Spanish general election, 2015. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20141112233651/http://www.fundacionsistema.com/media/public/doc/dossierT241.pdf to http://www.fundacionsistema.com/media/public/doc/dossierT241.pdf
• Added archive https://web.archive.org/web/20131105143100/http://www.fundacionsistema.com/media/public/doc/Dossier%2023_T228.pdf to http://www.fundacionsistema.com/media/public/doc/Dossier%2023_T228.pdf
• Added archive https://web.archive.org/web/20130228040404/http://www.fundacionsistema.com/media/PDF/Dossier217.pdf to http://www.fundacionsistema.com/media/PDF/Dossier217.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 20:44, 20 November 2017 (UTC)

Trimming table content

Re Impru20's edit summaries, I boldly determined what was excessive. I used a reasonably conservative criteria where I removed parties and groups who only ever polled under 2%, hadn't won any seats in 2011 or 2015, and most of the time were completely omitted by polling organisations, many of whom were regional parties. Given the immense size of the table despite a tremendous amount of blank cells, I think this is a very reasonable action. Onetwothreeip (talk) 01:45, 22 January 2019 (UTC)

Could be reasonable, but not so reasonable considering how did you proceed, starting with the fact you did remove some parties that were shown in the chart (Compromís, Equo, Vox), or removed some parties in one table but not in the other. Then, the 2% threshold is pretty random, and besides, you did remove Equo despite it polling at 2% in some polls. You also removed Compromís, GBai and FAC despite them winning seats in both 2011 or 2015. I may concede we could live without some of the parties, but you removed parties based on selective criteria, then applied those criteria selectively as well. Nonetheless, as I said, I concede some parties could be spared. I'll check it out tomorrow and give some ideas, since it's late here right now. Impru20^talk 01:55, 22 January 2019 (UTC)

It's not required that we detail polling results as shown in a complementary chart, if anything the chart would have to fit with the table but I don't think that is required. It may be the case that a party was polling notable enough in one table but not in the other. The 2% threshold is arbitrary and it might as well be 3% or 4%. As for parties being removed that ended up winning seats, I believe this was because they ended up winning seats as part of a group ticket (and that ticket's polling was retained), or possibly because they only won a single seat. I'm sure you're aware as someone who has edited polling articles about different countries that the typical threshold would be something like a party consistently polling 5% or more. Onetwothreeip (talk) 02:11, 22 January 2019 (UTC)

That is no justification. Each country has its own dynamics, and most countries do not have relevant regional parties such as there are in Spain. 2% does not mean the same for a party running at the national level (where the party may have difficulties to win even a single seat) than for a regional party (i.e. the PNV typically scores about 25-30% in the Basque Country and is a very notable political force in Spain, yet it never ever reaches 2% nationally).

Further, opinion polling tables in Spain are typically not split; they were split here due to their size, but that does not mean that, somehow, there is an entitlement or justification to show a party in one of them but remove it from the other. I'm taking a look at the tables, though, since at the very least I guess that PA, PRC and others can be removed without much issue due to these not being reported by nearly any opinion poll; then, we may see from that. Impru20^talk 08:56, 22 January 2019 (UTC)

I was referring to irrelevant regional parties, not relevant ones. Such as the ones which would get maybe 5% in a particular region. We don't really need to have these small parties on the main table when we can have them in tables for their own regions like we do with United Kingdom polling for Scotland, Wales and especially Northern Ireland. Anyway, we're a long way from that.

Removing PA and PRC removes 10,000 bytes from the article, which is a start but much more is needed. While splitting the table wasn't done for the purposes of removing a party from one table but not the other, it that can definitely be an unintended possibility from that action. Onetwothreeip (talk) 09:45, 22 January 2019 (UTC)

Ok, I removed a further 34,000 bytes from the article by removing some other minor parties which weren't really polled that much and that, even when done, polled considerably at low levels. The parties which have been left are much more notable, were polled much more often and do polled greatly above 5% in their respective regions for most of the polling series, so in that sense I don't know what else could be done. On the table splitting issue, it was split precisely because of the size resulting from having so many parties in there, but with many less parties now the split could maybe make less sense.

Nonetheless, I see your main issue here is to reduce the article's size; I guess we could port some of the content into templates and then transclude them here (for example, the party polling table, which is also somewhat big). However, any splitting from this article will require careful micro-management as most of the other tables are dependant on referencing from the main polling table. Impru20^talk 10:32, 22 January 2019 (UTC)

EDIT: Ok, I see merging the two tables helps reducing a further 2,000 bytes. Removing the "group=p" tag from references also helps reducing the article by a further 8,500 bytes. Possibly not a great deal by itself, but all in all the combined edits have helped reduce the article's size by about 68,000 bytes. Impru20^talk 10:47, 22 January 2019 (UTC)

@Onetwothreeip: Transclusion was a success. As the relevant content is meant to be transcluded here, there isn't any need to work out the referencing into the transcluded template, as the reference tagging will detect the referencing in this article once transcluded even if the links themselves aren't present in the Template. Further, editing is very simple, and the transclusion itself goes smoothly to the point it can't be spotted at first glance, yet helps saving a further 50,000 bytes in the article. Impru20^talk 11:08, 22 January 2019 (UTC)

Another transclusion conducted for the Prime Minister polling section, worth about 10.5 kB, plus another one for the graphical summary worth 1.5 kB. All in all, the article has been reduced by 133 kB counting all the tweaks performed, and is now the 207th longest Wikipedia article, down from 5th as it was earlier. Impru20^talk 11:19, 22 January 2019 (UTC)

@Onetwothreeip: You're now removing parties just for the sake of it without even caring to discuss it further. these might be relevant in a table for regional polling Seriously? So you propose a solution which takes even more article size and needless duplicities just because you can't seemingly live with these being in the main table? Further, such "regional parties" table is nothing that has been done anywhere else in Wikipedia. This is openly disruptive.

Further, you are helpessly focusing your attention in a pointless issue such as the number of regional parties. A vast majority of the article's oversize is taken up by sourcing, not by the amount of regional parties in the table, and you can't just remove sources without breaking WP:VERIFY which is a core WP policy. Impru20^talk 21:02, 22 January 2019 (UTC)

I carried on with removing parties from the table in the same manner that you were doing. They had no more than 2 seats at the 2011 election, and no more than 1 seat at the 2015 election. They never polled significantly during the parliamentary term by any reasonable benchmark, if any party "polled" at five seats which is purely speculative, it was not at all consistent and still only five seats out of the 350. Which party that I removed do you want to keep? None of what I did was random and was for reducing the width of the table. These are only the small regional parties, not the big regional parties which I have kept, and if they are relevant on a regional scale at any time they can be shown in regional polling. Onetwothreeip (talk) 21:09, 22 January 2019 (UTC)

In the same manner I was doing? You basically removed those parties you did already remove when I contested the removal on the first place. I removed some of those that were more expendable, but it seems that you wish to carry on with your own particular vision anyway.

They had no more than 2 seats at the 2011 election, and no more than 1 seat at the 2015 election And? Compromís did poll at 2% and 5 seats or more at some points, which is a whole lot. This is an article for opinion polling, not for election results. Unlike some of the removed parties, all of the regional parties in the current table are polled by a large number of pollsters and are notable enough in their own to be kept. I really do not understand what you are aiming to do here. Reducing the table width, indeed, but at the cost of mutilating it. Impru20^talk 21:14, 22 January 2019 (UTC)

That's right, at their peak they only polled 2%, so it's likely an outlier anyway. We don't normally give a column to every party that ever gets 2% for the reason that it includes too many. They simply aren't important on the national level, which is what this article is about. Is Compromis the only party you disagree with removing? Onetwothreeip (talk) 21:23, 22 January 2019 (UTC)

Compromís got 18% of votes in the 2015 Valencian regional election, which would have entitled it to as many as 7 seats in the Congress. I'm not sure how do you reach the conclusion that these polls are "likely outliers". Every country has its own dynamics, so what may be useful for other countries may not be for Spain, where regional parties play an important part in the government formation processes because they typically get a considerable chunk of the seats up for election. Further, polls at the regional level in Spain are pretty scarce in comparison to national polls, so merely re-directing to those to justify mutilating the table and hide some parties under random criteria is not acceptable, as that could be seen as POVish.

Nonetheless, looking at other articles, it looks like the trend is in favour of adding data for regional parties to the main tables when available. Taking the UK example you mentioned earlier, it depicts Plaid Cymru despite it polling consistently at 0-1%. On the other hand, the Canada article has recently seen the inclusion of a party that polls consistently below 2% (such a result would possibly mean such a party winning 1 or no seat in the next election, yet it is included anyway). I do not understand from where are you getting the argument that "they simply aren't important on the national level". Impru20^talk 21:59, 22 January 2019 (UTC)

Suggesting that I am mutilating the table, or that I want to, is simply assuming bad faith, and I don't think you want to do that. Obviously the table isn't "mutilated" just because it may have ten parties instead of twenty.

Plaid Cymru is not featured in the polling table alone, it is only there to accompany the SNP which polls at the threshold. The reason for this is that some organisations report figures separately for PC and SNP, while others combine them into "nationalists". If these were separate we would certainly not have Plaid Cymru in a column alone, and we haven't had them as a column in the past even though they were about as popular as they are now, but when the SNP wasn't as popular. As for the People's Party in Canada, this has been included because they've had higher polling than 2% and a member of parliament in a non-proportional electoral system. If you are thinking I said that all small or regional parties are not important on the national level then you are completely mistaken, I was referring only to the ones which are small and regional.

Anyway, I take it from this you only object to the removal of Compromis? With regards to them, the results of a regional election aren't relevant here, and what I meant by outliers was that if they have only polled at more than 2% once then this is almost certainly an outlier. Onetwothreeip (talk) 22:41, 22 January 2019 (UTC)

Sure, I don't want to assume bad faith. But it is true as well that, so far, I'm the only one that has made effective concessions here in order to reach a compromise.

Plaid Cymru is not featured in the polling table alone, it is only there to accompany the SNP which polls at the threshold. That's not correct. Looking at the discussion, Plaid's inclusion was ultimately decided based on inclusiveness criteria. Indeed, a suggestion was done so as to show Plaid in a similar way as done for 2017 (with a brief footnote next to SNP figure, showing that the column was not needed at all), but ultimately the whole column was included. we haven't had them as a column in the past. But when practice and consensus evolve to the point they end up being included based on inclusiveness, that probably means that, maybe (just maybe) what was done in the past was not the correct thing to do. Further, it is also true that pollsters in the UK have been polling regional parties only recently, and that in the past they simply gave no numbers for these so you couldn't obviously add some data that did not exist. This isn't the same for Spain.

The PPC in Canada has polled higher than 2% in only 4 polls out of 32 conducted since its formation. According to your own criteria, these four polls would be "outliers", so these shouldn't be counted and thus the party should be excluded from the table. Further, the party commands just one single seat as of currently and could win none according to some polls numbers; this, opposed to twice, thrice, fourfold or fivefold of some of the regional parties you are intent on removing here. I'm sorry, probably you're here with the better of faiths or with the best intents, but you are being inconsistent; you're being clearly more inclusive when it comes to other articles than when it comes to this one. I guess this is part of your attempt at reducing the article's size, which is nice, but that just can't come by attempting to cut down tables at any cost.

Anyway, I take it from this you only object to the removal of Compromis? No. I was putting Compromís as the most striking example, but I'm objecting any further removal for now, specially considering your own criteria are dependant on very subjective arguments based on your own impression of what constitutes a "relevant" party in Spain. Impru20^talk 23:20, 22 January 2019 (UTC)

EDIT: If there is really a very strong and insurmountable need to reduce the table size, I think both FAC and GBai columns could be removed at most, as they only typically get 1 seat in opinion polls. But Compromís, CC and BNG are unnegotiable here on my part, as these poll near 1% and about 3/5 seats at some points of the table (which is much for regional parties in Spain). That would leave the parties in the table at 14. Impru20^talk 23:32, 22 January 2019 (UTC)

There was a discussion about including Plaid Cymru in the polling table, and it was revealed that they are there in accompanying SNP, per [[1]]. The amount of seats a party can win have never been relevant as criteria for including them. It's not as if the People's Party is being reported because they can win a seat, and then therefore all the regional parties of Spain should be included. If a party could win 20% of the vote but none of the seats obviously we would still include them, but it's also a disingenuous comparison since Canada does not have proportional representation. Being inclusive of PPC means only having six parties in the table, while this article has much more than that. You've made argument for retaining Compromis in the table, but are there any other parties to keep as well? I think we should include 11 parties on this article, and that doesn't make me noninclusive. Onetwothreeip (talk) 23:36, 22 January 2019 (UTC)

This was written prior to your post-edit being published. Onetwothreeip (talk) 23:36, 22 January 2019 (UTC)

Ok then, I've already replied to the Plaid discussion issue and I stand by my previous edit in that sense, so I'll avoid repeating myself. I would keep CC and BNG as well because, under the Spanish electoral system (which is inspired on proportionality but does not work as a proportional system) they are more relevant and notable than some parties under a FPTP system. I've already removed FAC and GBai due to them being the less significant of the remaining regional parties. Seriously, can you explain me what is your aim? I do not understand why now you bring 11 as the number of parties that should be in the table. If you want to reduce the table's width, there are other ways to accomplish it. Impru20^talk 00:13, 23 January 2019 (UTC)

I'm content with how the table is currently. I didn't choose the number 11, I simply counted the parties I wanted to keep and it was 11. I'm fine with 14 and while I think it could be better, I don't intend on making those changes. Thank you for your co-operation and patience. Onetwothreeip (talk) 00:28, 23 January 2019 (UTC)

Ok, that's fine. Nonetheless, it's true that the table was pretty wide, but the 2011–2015 legislature was particularly fragmentary in terms of opinion polling (normally, there are many less parties than those shown here before). I hope it does not cause any troubles now. I'll be cutting down some very minor parties from other Spanish opinion polling articles based on the precedents here. Regards. Impru20^talk 00:42, 23 January 2019 (UTC)

A Commons file used on this page has been nominated for deletion

The following Wikimedia Commons file used on this page has been nominated for deletion:

• Logotipo EAJ-PNV.svg

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 06:09, 15 November 2019 (UTC)