Jump to content

Talk:Opinion polling for the 2013 Italian general election

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

Coalitions

[edit]

Now that coalitions are known, polls for the year 2013 should be written with coalition partners side-by-side, and if there is enough place, add a "coalition sums" column, to see which winner is predicted by the pollster. Kahlores (talk) 21:51, 1 January 2013 (UTC)[reply]

That's what I did, but 4idaho disagrees (see below). --Checco (talk) 15:42, 9 January 2013 (UTC)[reply]

Format

[edit]

I strongly disagree with today's edits by 4idaho. They constitute a total rollback and contain errors and contradictions. In his edits he says that the format I implemented is a "bad format, internally inconsistent", including "incorrect information". "Do not attempt to revert again – he goes – or I will be forced report you for vandalism. Take it up on Talk if you want to reformat the article". He adds that his format "was extensively discussed" and concludes that "Wikipedia is not a dictatorship, and you do not have the right to unilaterally make changes. You may propose your changes on Talk for discussion".

First, it is his format (and his explanatory notes) to be inconsistent, incorrect and erroneous, other than bad looking. Secondly, his format was never extensively discussed and, in fact, the only comment in this talk is by Kahlores, who proposes "coalition sums", something that is part of my format and not of 4idaho's.

I have very few doubts on which format is better (take a look at mine and his), but I don't want to open an edit war. I just ask to the other contributors of to decide what is the best and go on. I have worked on this article since January 2012 (at that time the "opinion polls" section was part of the general article on the election). Wikipedia is not a "dictatorship" and I'm not a dictator: if 4idaho's version (with all its errors and inconsistencies) will be implemented as the standard for this article, it's not a big deal, I will simply stand aside and stop contributing to this article, despite being its legitimate father and main editor.

--Checco (talk) 15:42, 9 January 2013 (UTC)[reply]

I'd just like to be clear with one thing, the notes are not mine, the table is not mine. And I'm not saying they can't be improved. I'm glad you came here and would like to work with you to implement most of your ideas. I'd like to work towards a consensus the way wikipedia is supposed to function. There is no record of past discussions here because this table was created on the original Italian General Election, 2013 article and this page was only created a matter of days ago. I have only added in polls and did not create either the 2012 table (which was created in the distant past, by who I don't know) or the 2013 table (created by RJFF.) I also did not write up the article (most of the write up comes from Angelo.romano IIRC)
Also, Kahlores suggested that coalition sums be added in "if there is enough space" which there clearly isn't. The current table is already spilling off the page and your table is so large it really wrecks wikipedia and will require off-the-page side scrolling on most laptops.
Additionally, the current table is already organized left to right, with the farthest left party on, well, the farthest left of the table and then progressing rightwards. Your table does not organize the parties left to right, seemingly running counter intuitively right-to-left (with the right on the left and the left on the right) but also with many clearly incorrect party placements, such as M5S to the left of RC, FD to the right of Monti's coalition, and the moderate radicals outside of the center-left coalition and to its left (by which of course I mean to its right.)
It was because of that incorrect party spectrum (when they were already organized left to right) while claiming you were organizing them by ideology, that I suspected you of vandalism (that, and the random and unexplained deletion of some of Angel.romano's text.)
Moving on, by internally inconsistent, I meant to point out that your table has no relation to the 2012 table. If you want to reformat the article you need to do it properly and consistently, and not have reformat one segment and leave other segments in a different format. That only serves to confuse causal readers. (Why get rid of party shading anyways, which is standard on most opinion polling articles?)
I understand your feelings as someone who helped form this article, but no one owns articles on wikipedia. Articles don't have fathers or "main editors" and wikipedia guidelines specifically discourage those attitudes. Articles are everyone's communal property; anyone can edit constructively, and articles are formed by consensus.
Like I said, I do like a lot of what you've done, and don't necessarily like the current table. Here's what I suggest:
  • English coalition names: I'll fix these right away. It's an english language article and should use the english names.
  • Coalition totals: I suggest you create a separate table, making two sub-sections under opinion polling called Opinion Poll Results By Party, and Opinion Poll Results By Coalition. One table gets too big, so lets make two. (Please do shade the leading coalition in the appropriate color though, as it makes it so much more simple to read.)
I hope this addresses your suggestions and you find this compromise reasonable, and that we can work together in the future with other editors on this article working towards a consensus format. --4idaho (talk) 20:46, 9 January 2013 (UTC)[reply]
Also, I'm a girl. --4idaho (talk) 22:01, 9 January 2013 (UTC)[reply]
I agree with 4idaho that it is more intuitive to have left-wing parties on the left and right-wing parties on the right, and not the other way round. I agree with 4idaho that it is disturbing to have a table that is too broad to be shown on one screen (at least on mine). I find 4idaho's version more easy to overlook and more simple to understand, especially if I imagine being a casual user without any previous knowledge. I cannot identify in which respect her version is supposed to be inconsistent, incorrect and erroneous, or bad looking. --RJFF (talk) 22:35, 9 January 2013 (UTC)[reply]
I have worked on these opinion polls since January 2012, so I'm very sorry that my knowledge and my help are not understood. I'll try to be cooperative, but with her total rollback 4idaho has cancelled lots of important edits.
Ok on the left-right scale, but why do we have some many inconsistencies in that scale? First, ALL parties should be ordered that way (just M5S is not easily classifiable). Secondly, why including irrelevant parties like AdC and not including the Radicals? Third, why using wrong names for the parties? The table is too big just because there are too many parties running not beacuse of the totals: 4idaho's table was also too broad, indeed. I'll correct the explanatory notes first, then I will propose two compromise versions for the table. --Checco (talk) 08:29, 10 January 2013 (UTC)[reply]
Ps: Coalition totals are essential in an election decided by coalition totals (and they are definitely more important small parties such as AdC and the "undecided" column). Anyway, as I'm interested in reaching a compromise, in my proposal I won't include coalition totals. As said, I will start with small edits with clear edit summaries. --Checco (talk) 08:40, 10 January 2013 (UTC)[reply]
Why do you keep editing the results for the 2008 election and replacing them with incorrect results? Not all of SA is in RC (part is in SEL), so it's inappropriate to show their results as that combined result -- and also the results for other parties (PD, PSI, PdL) are simply incorrect. And PD's colors are Green, White, and Red. Why is Green not appropriate as a color shading for when PD is in the lead?
Also, I agree with the deletion of AdC (like I said, I'm not the one who made the table.) The table now fits on the page, hooray! XD --4idaho (talk) 14:02, 10 January 2013 (UTC)[reply]
Why do you ignore my edit summaries? Everything was explained there.
The 2008 numbers I inserted are the official ones, taken from the Electoral Archive of Elections (official website run by the Ministry of the Interior).
SEL is a split in PRC: so what? Also the PdL and the PD had splits (FLI, ApI, etc.). We need to be consistent.
The PD has no official colors; it uses orange, green, read and white. So how do you deduce that green is its official color?
The table fits the page, but it won't because many lists are missing (notably Radicals).
--Checco (talk) 14:14, 10 January 2013 (UTC)[reply]
You're looking at the incorrect information, the entire center-left coalition scored 37+%, PD alone scored 34+%. Please click through to the 2008 wikipedia article for the information on those results (which were won by the PdL/Center-right silly! XD You should be able to add your results in your head and see they don't add up.)
Also, the difference between PD and RC is that PD is still PD, if they suffered a sectarian split they're still the same party with the same ideology, RC is a part of SA, combined with other new parties. Why does RC get to list SA's results but not SEL? Neither are SA, they're both a spin off themselves combined with new elements.
What you're doing is more like listing PD's '08 result as CD's '08 result. Also, I got Green, White, and Red from the wiki page on the party. I wasn't the one who chose Green (again, I didn't make the table) but I assume it was chosen because that color is most prominent on the party flag. --4idaho (talk) 14:25, 10 January 2013 (UTC)[reply]
Uhm, are sure? The data I implemented are exactly the official ones from the Ministry of the Interior... See Electoral Archive of Elections. On RC we clearly have two different opinions, let's see what RJFF thinks. --Checco (talk) 15:38, 10 January 2013 (UTC)[reply]
I changed my mind. I won't edit the table for now. First, we need to find a compromise on the order of parties and on which parties to include. Here are the current parties (ordered from left to right; I'm not certain on the order of SC, FLI and UdC):
  • M5S (neither left nor right, but it is better to have RC close to SEL)
  • RC (contiguous with SEL)
  • SEL (contiguous with RC; most leftist party of IBC)
  • PD (broad centre-left party)
  • PSI (more centrist than PD)
  • CD (most centrist party of IBC)
  • SC (broadest party of CMI, includes centre-left figures)
  • FLI
  • UdC
  • FD (centrist liberal party; in competition with SC)
  • LN (contiguous with FD; competition between the two parties in the North)
  • PdL
  • GS (continguous with PdL; centrist spin-off from PdL)
  • FdI (conservative spin-off from PdL; contiguous with LD)
  • LD (most conservative party among mainstream parties)
I would personally include the Radicals (historic party); I would also sacrifice small parties (PSI, CD and GS) and reintroduce coalition totals (or introduce an "others" column for each of the two big coalitions: there will be many smaller parties in both big coalition indeed). --Checco (talk) 09:36, 10 January 2013 (UTC)[reply]
Ps: Currently, three opinion polls are missing in the table: Piepoli 8-Jan, Lorien 9-Jan and Tecnè 9-Jan. I won't add them until we agree on the order of parties.
I can agree with this order of parties, but don't want to sacrifice PSI and CD as they hold a not insignificant number of seats. What was wrong with my proposal to create a new table for coalition totals? --4idaho (talk) 14:05, 10 January 2013 (UTC)[reply]
OK, many thanks for your opening, 4idaho. I will implement the new order.
I agree with your proposal on a separate table for coalition totals. It will be a little bit complicated to implement it and more work to do, but it's definitely a good idea. Would you like to work on that while I update the table with the new order and the new polls? --Checco (talk) 14:24, 10 January 2013 (UTC)[reply]
Sure, I can have the coalition table up in a little bit. :-) --4idaho (talk) 14:28, 10 January 2013 (UTC)[reply]
The table is implemented. Comments? (I'll slowly begin adding in coalition totals for all polls.) --4idaho (talk) 15:13, 10 January 2013 (UTC)[reply]
Good job! Ok for the colors. Style might be improved though. --Checco (talk) 15:38, 10 January 2013 (UTC)[reply]

Scenari Politici

[edit]

Unlike the other listed pollsters, Scenari Politici is not acknowledged by the government's official site (www.sondaggipoliticoelettorali.it). Its survey is conducted only among internet users. Therefore I doubt that it is actually representative. Its result differs significantly from the findings of other pollsters. I think that we should not include the SP poll in the list. We have listed enough other polls from the same period of time. --RJFF (talk) 22:43, 9 January 2013 (UTC)[reply]

Pollsters are not mandated to present their results in the "official site". SP and Piepoli are both respected pollsters (the former contributes for Corriere della Sera, the latter with RAI, the national television) even though they don't present their data in the "official site". Both SP and Piepoli have to stay in the table. --Checco (talk) 08:29, 10 January 2013 (UTC)[reply]
If I could add to Checco's comment, many accurate pollsters worldwide (such as YouGov) conduct internet based surveys, so I don't see that as a basis for suspicion either. --4idaho (talk) 14:19, 10 January 2013 (UTC)[reply]

Graph

[edit]

We need a graph for this article! Lets discuss options. Obviously we can't include every individual party with it's own line in the graph because it would be impossible to interpret and too busy. I suggest that we have a line graph whereby we combine the polling for the political alliances and parties that have a consistently sufficiently high poll rating. I propose that we include: The Centre-right, Italy. Common Good, With Monti for Italy, Five Star Movement and all others should be included in an others column. Discuss! — Preceding unsigned comment added by 81.133.12.45 (talk) 14:15, 10 January 2013 (UTC)[reply]

RC should be included, otherwise that sounds good. I wouldn't know how to make such a graph on the computer though. --4idaho (talk) 14:35, 10 January 2013 (UTC)[reply]

I beg your pardon, my mistake. Yes RC will need to be included as well. Setting up a graph is really quite simple if you've got Excell, I'll explain:

1. Just highlight the entire polling box from wikipedia, copy and paste it to Excell.

2. Then you need to merge the polling for the individual parties that are within an alliance. You do this by creating a new column for each alliance and doing an equation to get a cumulative figure for the alliance, you will need to do this for the 1st 3 polling figures for each alliance. After that you can simply drag and drop!

3. Then you delete the columns you don't need for the graph i.e. the individual parties that are now included in Alliance columns, the "institute" column and the "UA3" column.

4. Then you highlight all your data and press the graph tab and select line graph, you can the customize it and upload to Wiki.

Hope this helps — Preceding unsigned comment added by 81.133.12.45 (talk) 21:58, 10 January 2013 (UTC)[reply]

I have created this table >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

I am not 100% happy with it, it seems pointless including Civil Revolution with their own line because there is not much data for them. If you look hard and squint you can just about see a brown squiggle in the bottom right hand corner of the graph. I couldn't use Orange because it was practically invisible! I am in favour of removing Civil Revolution and including them within others for this reason. I will adapt and upload the new graph if others approve. Sheffno1gunner (talk) 23:31, 10 January 2013 (UTC)[reply]

I like it and yes that sounds a sensible idea, that squiggle is neither use nor ornament, put them in with the others. Thanks for doing this, it looks really good! — Preceding unsigned comment added by 217.41.32.3 (talk) 23:36, 10 January 2013 (UTC)[reply]

Thank you for the table, but RC should be included. It is included in the coalition totals table, and is a relevant part of the upcoming election. There will be more data on them by the time the election comes around. I understand the aesthetic concerns, but the first concern should be its relevance and usefulness. --4idaho (talk) 23:53, 10 January 2013 (UTC)[reply]

I'm all for including them within the coalitions table, they are polling high enough etc, that part makes sense. But if we're to be practical there is no sense in including them in the graph, at least not yet because you can hardly see them, the 2nd graph is so much better! If we had data going back to say half way through 2012 I would agree with you because they are polling high enough. A graph is simply a visual aid and I'm afraid their inclusion at this stage is unhelpful. Like I say though they have every right to be in the coalitions table! — Preceding unsigned comment added by 217.41.32.3 (talk) 00:22, 11 January 2013 (UTC)[reply]

Hi guys, thanks for what you're doing. It seems to me that the chart includes some big errors: if you read the results of opinion polls for 2012, you will see that the centre-right (PdL+LN+LD) was never above of 40%. There has to be a mistake. Please check. --Checco (talk) 10:02, 12 January 2013 (UTC)[reply]

No the graph is correct, it's just easy to confuse because of the 2 tables having different parties in them. In the 2012 table there are 4 parties in the Centre-Right block, not 3. The People of Freedom, Lega Nord, Italy of Values and The Right and for the early part of 2012 those totals did exceed 40% regularly. Thankyou for your concern but the graph is fine!217.41.32.3 (talk) 06:35, 14 January 2013 (UTC)[reply]

Uhm, so this is not just a mistake, it's a BIG mistake. Italy of Values has never been part of the centre-right: IdV is a strongly anti-Berlusconi party, indeed. Coalitions in the graph have to be the ones running in the 2013 election, so, for 2012 polls, PdL+LN+LD (=centre-right), PD+SEL (=Italy.CommonGood), UDC+FLI (=Monti), M5S and IdV+FdS (=RC). I'm not able to edit the graph by myself, so I will remove it from the article: as it is, it is totally misguided, mistaken and deceptive. Feel free to reintroduce a corrected graph. Also RC should be included. --Checco (talk) 14:54, 16 January 2013 (UTC)[reply]
Checco is correct. And rather than how RC was included in the previous graph, please do it as Checco suggested and show the long term trend lines with RC as FdS+IdV. This article really appreciate a graph, but it must be done correctly. --4idaho (talk) 15:12, 16 January 2013 (UTC)[reply]

I can't be bothered anymore, you've got Excell don't you, do it yourself! It's not hard, it's just fiddly, what with moving tables around! I'm not going to spend my time making tables for an election I am not massively interested in only to be bossed around and insulted. Do it yourself, it's not that much effort for someone who's interested. I'm not particularly interested, I just thought I'd lend a hand, a hand that I thought was needed! Goodbye!Sheffno1gunner (talk) 18:20, 17 January 2013 (UTC)[reply]

Who are you? Are you the same person as the IP we were talking with? If yes, I thought you were the one who wanted to have a graph in the article. No-one asked you to craft it in the first place. Anyway, yes, you were right: a graph would make the article better. I'm sorry if you felt bossed and insulted, but it has been a little bit difficult to talk with you due to the login problems and we surely need to have a graph which is correct and consistent with the article. I hope you will come back here and do it. It's not for us, but for the encyclopedia. Thank you in advance for that. --Checco (talk) 12:26, 19 January 2013 (UTC)[reply]

Proposal to condense table: remove FiD and Rd

[edit]

1). Checco, I thought it'd already been agreed not to add Rd since the table is already spilling off the table, but 2). both FiD and Rd are polling far bellow the 4% threshold for parties and unitary electoral lists, so it would be a good idea to remove them. They are effectively irrelevant, and wikipedia's policy is not to insert unnecessary information. If we remove them the table will be more readable, relevant, and will almost fit on my browser page.

The page provide links to the full polls, so if some people are interested in the totals for minor parties, they can easily find them with a single click of their mouse without us having to destroy the article format.

It might also be a good idea to discuss removing the minor left/right parties (PSI, CD, GS, FDI, LD) and replace them with two simple "Other Italy. Common Good" and "Other Center-right" columns. Thoughts? --4idaho (talk) 15:03, 16 January 2013 (UTC)[reply]

Premise: computers have different resolutions, in my pc the table doesn't spill off the table. This said, I have no problems with removing PSI, CD and FLI (PSI won't even run lists for the Chamber and will run in just three regions for the Chamber, I have more problems with removing GS, FdI and LD as they will probably take around 2-4% on 24-25 February. Also the Radicals and FiD will probably score around 2-4%. I may live without GS, FdI, LD, Rd. and FiD, while I fairly disagree with "Other Italy. Common Good" and "Other centre-right": they are already included in coalitions' totals. --Checco (talk) 15:14, 16 January 2013 (UTC)[reply]
Good idea, lets not create new columns and simply remove them from the table then (once we come to an agreement on which parties should go.) I feel that the first parties on the chopping block should be Fid and the Radicals, since, your expectations notwithstanding, they have been included in only about 1/3 of the polls (compared to much more often for the minor coalition parties) and almost always poll below 1% when polled.
Also, the parties in coalition only need 2% if their coalitions collectively get over 8% (which the major coalitions obviously will) which makes them much more likely to win seats, as opposed to Fid and Rd, which need to reach the 4% threshold.
Overall, I agree the minor center-right parties (GS, FdI, and LD) should stay, think FiD and Rd should definitely go. I'm also OK with PSI and CD being excluded since although they only need to cross a 2% threshold, they rarely even poll 1%. And, while its true different computers are of different size and quality, we should work to accommodate as many people as possible and not just those with nice computers. --4idaho (talk) 15:29, 16 January 2013 (UTC)[reply]
We need to be prudent and clever. It is true that FiD and the Radicals/AGL are not polling very well (although there are some polls giving FiD more than 4%: the last SP and, regularly, SpinCon): the two lists are not known enough and pollsters tend to underestimate them; I expect them rising in polls over the next two weeks. Same for FdI, GS and, possibly, CD. I know Wikipedia is not a crystal ball, but let's be prudent and not put into the trash a lot of work done (we might have to rescue those data later on). For now I would simply shorthen the table by removing PSI (it won't even take part to the election with its own list!) and the three notes after RC, SC and U/A (we can leave the explanations at the bottom of the table). --Checco (talk) 15:46, 16 January 2013 (UTC)[reply]
We must be prudent and clever, but unbiased and objective. CD has not polled any higher than PSI, so we can not use a flimsy rationale like what we expect to happen to remove one but not the other. I accept your rationale for keeping FiD, since while SpinCon is an outlier, outliers can be correct (I honestly had forgotten about SpinCon's results since they haven't released a new poll in a while.) There is no objective rational for Rd, so Rd should go. No poll has shown them coming even remotely close to the 4% threshold, and the election is only a little more than a month away. We're also not including other tiny parties polling at a similar level, so the article is showing unfair bias towards Rd.
So I think we should cut Rd back out immediately to condense the table, and either cut or keep both CD and PSI. I don't have a strong opinion, so I'm fine with you deciding whether PSI and CD should go, but I since they're polling similarly, and both have to meet the same 2% threshold, they should either stay or go together. --4idaho (talk) 19:38, 16 January 2013 (UTC)[reply]
I think there is a big difference between a party (PSI) which is not running for the Chamber (all the opinion polls refer to the Chamber election) and any other party which is actually running. I don't see why we should remove the Radicals, who are polling better than other parties. Having a table that fits into your browser page is not a sufficient reason for removing parties from it. However, I condensed the table by removing the numbers after RC, SC and U/A. What do you think? And what about this? [I took the idea from Opinion polling for the next United Kingdom general election, the standard article template for articles like ours.] Anyway, I still think that PSI is totally unuseful in the table as it won't take part to the election. --Checco (talk) 12:26, 19 January 2013 (UTC)[reply]
Ps: My computer is a very standard (and low quality) pc: I bought it a couple of years ago at Best Buy for just $300!
Fine, PSI goes and CD stays, I was only objecting to removing PSI and not CD on the ground you thought CD would rise in the polls, that's not a good reason. The reason why Rd should go despite out polling CD and GS is that it's a different threshold. As part of a coalition polling over 8% collectively, they only need to poll 2%, and they're just under that threshold. Despite polling marginally higher, Rd is nowhere near the 4% threshold for a party outside a coalition.
To put it another way CD and GS are only about 1% away from entering parliament, while Rd is about 3% away!
(P.S. I like the new tables aesthetically, but I'm worried the text is too small... I find it a bit hard to read. Is there anyway to just increase the font size slightly?) --4idaho (talk) 16:43, 19 January 2013 (UTC)[reply]
As long as you don't remove any party other than PSI, I won't complain. In my view, the tables are OK as they are now (I have no problems with reading them and my PC screen is standard size) and we can even leave PSI where it is. I'm not a template specialist, but I'm sure you can increase the font size slightly to improve your view. Your argument on threholds is very smart, but I still prefer not to make differences between parties polling similarly.
--Checco (talk) 10:38, 26 January 2013 (UTC)[reply]
OK, I'll probably remove PSI when I find the time.
By the way, I also prefer coloring PD red instead of green, but you need to do both tables to make it consistent. It will be too confusing to have '12 shaded Green, and '13 shaded red; people will think there was some sort of change in PD that warranted the change.
--4idaho (talk) 13:41, 26 January 2013 (UTC)[reply]

Poll

[edit]

Last poll 8 feb?? — Preceding unsigned comment added by 81.58.144.30 (talk) 22:20, 17 February 2013 (UTC)[reply]

Polling tables

[edit]

Hi @Impru20:, I copied in my sandbox you great job for the opinion polling tables about 2013 election and I made some modifications. Could you insert the tables in the article? If you don't I can do it; it's still better to have an almost complete list of polls that the current one. Thank you and have a good day! -- Nick.mon (talk) 07:45, 8 April 2016 (UTC)[reply]