Talk:Opinion polling for the 2019 Portuguese legislative election

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Recent changes to table[edit]

Kahlores has tried to re-order parties in the table based on the left-to-right axis, as he has done on multiple articles (being reverted on various of them by other users as well). That this is a very bad idea is easily proven by this and this. These articles are for opinion polling, not for getting entangled in discussions about the parties' ideologies and on which one should be the actual column placement, which is something that the use of the left/right axis will foster as evidenced, to no particular gain in the polls' presentation. Ordering parties based on their last election result is the most objective and widely used criterion available, and unless sources themselves did widely use such a left/right axis (which is not the case here), it should be avoided completely. Another element which has been tried to be implemented is the sorting of the table, which is a major change that would also require a consensus seeing how it is contested as well. Impru20talk 15:56, 23 August 2019 (UTC)[reply]

  1. The position of communists towards liberal democracy is not entangling at all. It's quite clear cut, once you get it. That said, if Portuguese sources all see BE as further left than PCP-PEV, then they're of course better than an educated guess.
  2. Every Portuguese citizen knows that PS is center-left and PSD is center-right, unless, of course, they are themselves judging from a position in that spectrum (trotskystes would surely call both of them "right-wing"). Objective sources would confirm what everybody knows.
  3. On the other hand, very few non-Portuguese would be able to guess that PSD is a center-right party. The left-right arrangement makes it immediately clear.
  4. Should the rank in the previous election be the criterion? We have a good example here with Portugal. The right-wing coalition was ahead in 2015 (39%), but did not form a government because of an agreement between... left-wing parties. Coming 1st is legally meaningless, why give it an importance that it doesn't have?
  5. The left-right arrangement allows us to make sense of the mergers and splits, which are almost always between neighbors on the axis. Case in point here with Portugal Ahead, splitting in 2 and even 3.
  6. I don't see in what way how adding a 'sorting' function could be contentious?
I'm currently writing a draft essay to best explain my point to Wikipedians like you. For the biggest obstacle is the widespread misunderstanding of the left-right axis.
Kahlores (talk) 16:57, 23 August 2019 (UTC)[reply]
The very fact that we have to discuss the exact nature of communists against liberal democracy shows how pointless is making such a distinction in this article. This one is for opinion polling, not for discussing party ideologies. If you wish to learn more about a party's ideology, you already have links to that political party's article, where such ideologies would be much more precisely depicted, explained and sourced that what you could conceivable explain here by ordering parties based on their supposed ideologies. Yes, some parties may be very straightforward to clasify, but others will inevitably lead to discussion. Mixing such an issue here by ordering parties based on the left/right axis only brings confusion and fosters dispute among users (this is not an opinion of mine, but documented facts; it has happened on this very same article only hours after the change was made, and has happened in other articles elsewhere).
Every Portuguese citizen knows that PS is center-left and PSD is center-right This, again, is not the purpose of this article.
very few non-Portuguese would be able to guess that PSD is a center-right party You won't solve this just by ordering parties based on their left/right spectrum. This ordering will only tell a casual reader (at most!) that party A is more to the left/right than party B, but not how much to the left/right is the party placed (and this not counting those situations of parties which can't be properly described within the left/right axis). For instance, placing the PSD to the right of the PS won't necessarily tell readers that the PSD is a center-right party; it could be mistaken as a right-wing or a purely centrist party, depending on how many parties does it have to its left/right (which help nothing but further confusion). All in all, this is much more to explain than this article can handle, and indeed it is not relevant at all because this is not the point of the article. If someone wishes to learn more about a specific party's ideologies, this will be much more properly covered in the party's article.
The left-right arrangement allows us to make sense of the mergers and splits, which are almost always between neighbors on the axis This can be already easily solved through notes and footnotes, where (btw) you will be able to give a much more precise explanation of such a situation than through a mere ordering of parties in the table. Also, you are making a rather wrong presumption that party alliances are understood only from the perspective of the left-right arrangement, when other factors may intervene than make such an arrangement pointless (i.e. government parties vs opposition parties (National Government (United Kingdom)), separatist parties vs unionist parties (Junts pel Sí), etc). Obviously this does not mention the fact that merging columns (as it was done for Portugal Ahead) makes editing the table much more complicated (specially when using the visual editor), and it may also prevent the sortability function from working properly (as for this one, some articles do include it, others don't. I don't see any specific benefit from it, specially considering that most of the table's columns must be left excluded from sortability and that the function poses some technical issues at times, but I will not make this a big issue of contention).
For the biggest obstacle is the widespread misunderstanding of the left-right axis. The biggest obstacle is to think that everything can (or should) be explained from the point of view of the left-right axis. It's not something that you need to explain your point to other Wikipedians so that they "see the truth" or something. The issue here is not a failure to understand the left-right axis, but precisely to understand that ideologies can go beyond such an axis and that we shouldn't be bringing the ideology issue to every article where parties are mentioned. Impru20talk 17:45, 23 August 2019 (UTC)[reply]
Okay, I get your points very well.
1. (on useless bickering) Not all parties' positions are contentious. Very few of them are. Most of the parties' positions are explained in the parties' infoboxes, with a reference. Discussions appear when (3) two parties overlap, like BE and PCP; (2) one is changing course; (1) someone did not understand the spectrum and thinks his preferred far-right party should not be put on the fringe (never happens on the left)
Ultimately, however, logic prevails.
2. (on positioning itself) You seem to have two points: 1) positioning is basically useless 2) it is not precise enough. For my part, I don't want even casual readers to lose their precious time. They want info about how a next election may unfold. What that means is they want to assess each possible outcome. They do not need to know "how much" PSD is on the right. That would be superfluous and useless. However, casual readers need to know quickly that PSD is not left-wing and is in fact, the major opposition to PS.
3. (on mergers and splits) You suggest to use footnotes instead. But both strategies can be complementary. You can see it for yourself here, and compare with this much stupider presentation.
Regionalism is indeed a special case of two-dimensional politics, this is why you'll never see me edit Catalonian or Basque articles. I hear you sigh :-)
4. (on sortable tables) I get it, you have been partially convinced :-)
5. (on the axis) some editors really do not understand the axis' logic, but others such as you seem to forget that party politics cannot be quickly explained in most countries without it! Its qualities are manyfold: (a) the left-right axis shows future coalitions at first glance; no nationwide coalition can be expected between BE and PSD; thus the parties are presented the way they present themselves to the Portuguese. (b) it shows the real ratio of power between parties, being understood that many minor parties' results never matter on their own, but only as part of one group (for instance, BE and CDS-PP). Many countries that seem to be multi-partite, are de facto bipartisan or follow a spectrum logic (for instance, Sweden's Riksdag, or Spain when there is no regionalist issue). We should thus follow this widely-accepted logic. (c) it allows the reader to see the swings between Left and Right halves, and the swings between moderates and extremes; trying to do that with the decreasing order is very tiring.
Kahlores (talk) 19:49, 23 August 2019 (UTC)[reply]
1. Yet again, this still doesn't deal with the fact that all of this is of absolute irrelevance here. Yes, not all parties' positions may be contentious, but this is not an article covering parties' positions nor should be, so we shouldn't even become worried about those parties whose positions may be contentious.
2. No, my points are 1) positioning is basically useless for opinion polling articles, which do not cover party positions; and 2) you can't get them precise enough in these kind of articles without distorting the nature and scope of the article itself. You know, opinion polling articles are usually created as split-offs from the main election articles, to cover just the opinion polling-bit of those. If you have to explain the parties' positions, ideologies and policies, that's what the main election article (as well as each political party articles) are for.
  • casual readers need to know quickly that PSD is not left-wing and is in fact, the major opposition to PS Here you have your concept wrong again. Is "main/major opposition" status dependant on whether the party is the ideological opposite of the ruling party? Because the last time I checked, this is usually determined by support obtained in the previous election. And that's what the current scheme, i.e. ordering by support in latest election, does fulfill. Ideologies have nothing to do here, for example in Poland you have that the main opposition to the ruling right-wing Law and Justice is the centre-right Civic Platform; in Germany the main opposition to the centre-right CDU's-led grand coalition is the far-right Alternative for Germany (and indeed, in Germany the national government is formed by two (supposedly) ideologically-opposite parties). Ideologies have nothing to do here.
3. Your complain on the ("stupider"?) representation in Opinion polling for the September 2019 Israeli legislative election seems more based on the multi-table design rather than on the actual left-right arrangement. Nonetheless, you are re-directing to a section of the page covering alternative scenario possibilities, so I don't think this is comparable: obviously those would require different tables, as those are different scenario hypotheses. I don't think that is stupid.
4. Nope, but I don't think it merits the effort on a major discussion if there's no major opposition on it. Further, seeing how it is freely used in some articles and that the current article is not too large in size to sustain it, it doesn't pose much of an issue for me.
5. But this is going back to point 1. Indeed, some countries' politics may be explained with the left-right axis (while others do not), but I should remind again that this is an opinion polling article. Dealing with the country's political scheme should be within the scope of the main election article, and dealing with each party's specific ideologies should pertain to each political party's article. That's what those articles are meant for and that's what their purpose is. The purpose of opinion polling articles, on the other hand, is dealing with opinion polls, and trying to mix up other election elements here would just mean the discussion would inevitably deviate from the main point of issue.
The "Central Bloc" totally fits the spectrum paradigm;
The Conte Cabinet follows a hung parliament, from which awkward coalitions can be expected;
The Torra Government is due to a second dimension; the left-right axis alone does not explain Catalonian politics, something I admitted;
Crystal-balling is what uninformed readers are doing right now, as they're not provided with the factual, referenced spectrum when there is one.
  • it shows the real ratio of power between parties, being understood that many minor parties' results never matter on their own, but only as part of one group. Ok, so should we obviate that this poses yet another exercise of crystal-balling (which we aren't supposed to do in Wikipedia, so this of yours can't be an argument for the proposed change), then what happens if one of these minor parties refuses to support the largest party within their bloc? That would obviously make that minor party's result to matter on its own, since it would not be joining any group (Portuguese's neighbours know it well). Again, everything of what you're explaining here is all matter pertaining to the main election article, which is where you'd explain the parties' positions ahead of the election, their ideological affinities and any alliances or coalitions that they choose to form. This article is for opinion polls.
I don't see your point. It's obvious that parties that do not stand together can have tough discussions, even within their bloc. But having discussions means you're closer than having no discussion. Therefore, we ought to show CDU-PP near PSD, as they will never support a PS government that PSD doesn't.
  • Spain when there is no regionalist issue Has there any time when regionalism wasn't an issue in the 17 autonomous community-Spain? xD Precisely, the examples you bring demonstrate that your logic shouldn't be applied. In Sweden the blocs have been broken after the Liberal and Centre parties choose to break the centre-right Alliance, whereas in Spain you can't establish a logic based just on the left-right axis, as other factors are usually present as well. Your logic is based on the assumption that the left-right axis is the ever-present dominant logic, that it will always be and that it should be of relevancy when reporting opinion polls, yet none of these are true, actually.
Yes, there was, for I have read many of your pages about elections in Spanish municipalities. Where there is no regionalist party (as often the case in Central Spain), it is rare to see discrete spectrum coalitions, most are a continuous spectrum, or an arc of the hemicycle if you will. Ever seen Podemos-PP coalitions? Vox-PSOE? Yes there can be PSOE-PP without C's, but there's a reason, either the math of majorities, or it's a special case, an exception that proves the rule.
By the way, your seating diagrams always perfectly follow the left-right paradigm, so what is all this fuss about?
  • it allows the reader to see the swings between Left and Right halves, and the swings between moderates and extremes; trying to do that with the decreasing order is very tiring This assumes that swings are always between ideologically-close parties and/or between parties within the same ideological "group", which has been demonstrated in many occasions throughout many countries to not be the case (or at least not explanatory of all inter-party swings). This is also an attempt to try to manually depict ourselves what are the "moderate" and the "extreme" parties, which again is not within the scope of an opinion polling article to deal with (unless pollsters did make such a left-right arrangement in their polls, but this happens to not be the case). Further, we shouldn't be giving readers indications as to how they should interpret sources, as that would constitute WP:SYNTH unless such an interpretation was backed by the sources themselves. Impru20talk 21:37, 23 August 2019 (UTC)[reply]
I do not "assume" in any way that swings are "always" between halves, where have you read this "always"?
Indeed, many swings can be chaotic, and some parties do take voters from everywhere.
But now put yourself in the shoes of the poll analyst, or the casual reader. What does he or she want to know?
  1. If one party has declined, where have its voters gone?
  2. If one party has surged, whom did it take voters from?
To be able to say that a party did not lose to neighbors on the axis, or took from them, the reader has to check first... those neighbors on the axis.
And to do so quickly, the best way is to put those neighbors nearby!
As for the other remark. The term "moderate", although not the best term, is the usual term that applies for parties near the center. So we're doing nothing subjective here, as this is already referenced in each parties' infoboxes. Just like the color and name.
Kahlores (talk) 19:25, 24 August 2019 (UTC)[reply]
You are making assumptions about swings based on the data given by opinion polls, which goes beyond WP:SYNTH. The existing problem (and the one you seemingly don't seem or don't want to understand) is that your analysis is based on the assumption that if one party declines, its support goes to another one which is close to it in the left-right spectrum, and you wish to re-arrange the tables that way to seemingly "help" readers to understand your point of view. You must realize that this is your own POV that you seemingly wish to enforce into all polling tables so as to "guide" casual readers about party swings, and that this is in clear violation of both SYNTH and WP:NPOV.
your analysis is based on the assumption that if one party declines, its support goes to another one which is close to it. I said the contrary: "many swings can be chaotic, and some parties do take voters from everywhere".
To be able to say that a party did not lose to neighbors on the axis, or took from them, the reader has to check first... those neighbors on the axis. This is your own POV, not something constituting an undisputed truth or an unavoidable requirement. This is your opinion, but maybe some other readers don't care at all about "neighbors on the axis". We, as a Wikipedia, shouldn't be giving "hints" to readers as to the direction of party swings. Wikipedia is not a mean to publish opinion pieces, a crystal ball or a newspaper. Poll results must be presented just as sources report them, without additional hints, assumptions or data interpretations of our own creation. Casual readers should be able to reach conclusions themselves with the presented data without them being purposedly directed in any specific way, yet the later is exactly what you are aiming for. If sources do not arrange opinion poll data for parties within the left-right axis, we shouldn't do it ourselves. It's fairly easy to understand. Impru20talk 19:57, 24 August 2019 (UTC)[reply]
Get off your high horses. I am arranging columns according to a factual info which is found in 95% of the parties' infoboxes. If being left-wing is NPOV, CRYSTAL, SOAP, or whatever HIGHHORSE you can find, then you must remove 100% of your own unsourced diagrams, for the sake of consistency. If you do this I will gladly un-do all of my left-to-right rearrangements, for I'm just following your steps.
Kahlores (talk) 22:16, 24 August 2019 (UTC)[reply]
The issue is that you're combining different information from various sources in order to reach your own conclusion and translate your own interpretation of these into a table arrangement for "informing casual readers" of what you wish them to be informed. And you are not even trying to hide this, so there is little else to discuss on it. Please, I'm linking WP:SYNTH gain to you because this is exactly what you're doing.
Do not mix up different issues here. Parliamentary diagrams using a left-wing arrangement is the standard in Wikipedia (not mine). Parliamentary composition and seat distribution in a diagram is a whole different thing that what you are pretending with opinion polling tables; this indeed may deserve a separate discussion, but under that reasoning, should "Results" tables in election articles also be re-arranged according to the left-to-right axis as per your arguments? Absurd, isn't it? That left-right arrangements are used in some situations under some conditions does not mean they must be used every time. Nonetheless, responding your direct remark on myself, I should note that all diagrams I upload merely stick to established practice in WP; here it is you the one going off WP standards in order to promote a self-declared particular purpose. So no, you are not following "my steps".
Finally, as far as I know, Wikipedia is not yours, so no one needs your permission to undo any edits in a given article. Impru20talk 22:44, 24 August 2019 (UTC)[reply]

A Commons file used on this page or its Wikidata item has been nominated for deletion[edit]

The following Wikimedia Commons file used on this page or its Wikidata item has been nominated for deletion:

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 20:26, 4 January 2021 (UTC)[reply]

Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:Opinion_polling_for_the_2019_Portuguese_legislative_election&oldid=1204443959"