Talk:Statewide opinion polling for the 2016 Democratic Party presidential primaries/Archive 2

Archive 1

Archive 2

Different colors for states that have already had their primaries / caucuses

I think there should be additional colors for states that have already had their primaries and caucuses. For the three variants we have right now, there could be equivalents in states that have already voted. For example, states that are really close (within a few percents) could have one variant, and harder Hillary / Bernie states could have variants. That's just how I think I'd go about changing the map. Does anyone else have thoughts about this? Dustin (talk) 17:03, 7 February 2016 (UTC)

Hi, if you go on commons you will see a note on Iowa that says the caucuses in Iowa have already occurred. Keep in mind that this is an article that deals with opinion polling rather than actual results. Also, for states where the popular votes aren't released i.e. Iowa it would be very hard to have an accurate coloring. Prcc27 (talk) 22:04, 7 February 2016 (UTC)

I believe that states, territories, and the District of Columbia, where the primary has already occurred, or the first round of the caucus has occurred, need to be shown with the colors differentiated from jurisdictions where the primary or caucus has not commenced. I am not talking about showing any perceptions of the outcome, but displaying the standings in the latest poll prior to the commencement of voting or caucusing in a different set of colors. This simply makes sense to me. Not visually differentiating between future primaries and caucus starts and past primaries and caucus starts does not make sense to me. Info por favor (talk) 06:39, 9 February 2016 (UTC)

@Prcc27, can you provide a link to the talk page on Wikimedia Commons where that consensus was reached, because I can not find it on the talk page of the map file on Commons. Info por favor (talk) 06:50, 9 February 2016 (UTC)

Unless Dustin or someone else opposes your idea I'm okay with it. However my main concern is consistency with the GOP map. Can you address that please? Prcc27 (talk) 17:36, 9 February 2016 (UTC)

Suggestion: Reformat tables to allow user to order by their preference

I don't understand why the tables are split up by state and don't have the following format allowing the user to either sort by state, date, or polling source. Am I missing something?

Sortable table

State	Date(s) Conducted	Date Published	Polling Source	MOE	Sample Size	First	Second	Third	Other
Alabama	2016-02-02		Overtime Politics	5.1	385	Hillary Clinton 54%	Bernie Sanders 43%	Martin O'Malley 1%	undecided 2%
Alabama	2015-08-11		News-5/Strategy Research	2	3,500	Hillary Clinton 78%	Bernie Sanders 10%
Alaska	2016-01-23		Alaska Dispatch News/Ivan Moore Research	3.8	651	Bernie Sanders 48%	Hillary Clinton 34%	Martin O'Malley 6%	undecided 14%
Arizona	2015-11-05		Behavior Research Center	7.3	186	Hillary Clinton 47%	Bernie Sanders 19%	Martin O'Malley 2%	uncommitted 32%
Arkansas	2015-12-23		Overtime Politics	5.2	209	Hillary Clinton 47%	Bernie Sanders 42%	Martin O'Malley 4%	undecided 7%
Colorado	2015-12-17		Overtime Politics		319	Hillary Clinton 49%	Bernie Sanders 36%	Martin O'Malley 4%	undecided 11%

@173.52.18.146: That is a very good point, we definitely don't need all of these unnecessary state sections! I support your proposal. By the way, it would be very helpful if you signed your posts on talk pages by typing 4 tildes (~). Prcc27 (talk) 06:53, 12 February 2016 (UTC)

• Oppose Sorting these poll results across states gives users no additional value and is highly misleading, as the results represent completely different populations. I'm actually not even sure if these polls should be covered in a single article at all, or if they should be moved to the individual articles on the respective state's primary. --PanchoS (talk) 07:14, 12 February 2016 (UTC)

• That didn't make any sense whatsoever. Of course the results represent completely different populations. This is a national race! 173.52.18.146's proposal gives the readers the ability to sort by state if they want to find a particular state's polls or they can sort by date if they want to find the newest polls. Your proposal to move polls to individual articles is not what we are discussing here. In fact, doing that almost make this article moot. Let's stick with one proposal at a time. Prcc27 (talk) 07:23, 12 February 2016 (UTC)

• No, with 1 out 7 delegates being superdelegates not determined by any of these primaries, this article's scope is not (much) more than the sum of multiple subnational races, with any WP:SYNTHESIS being highly speculative. You're saying a single table would help the reader finding a particular state or just the newest polls. Now, while the former is already possible without having to sort a table, the latter isn't of any encyclopedic value. Actually, the Latest polls section isn't encyclopedic either and could be removed on that basis. --PanchoS (talk) 10:37, 12 February 2016 (UTC)

Please elaborate on your WP:SYNTH concerns. Consolidating the tables doesn't synthesis anything. Prcc27 (talk) 19:49, 12 February 2016 (UTC)

Consistent layout needed

Per WP:MOS "Style and formatting should be consistent within an article", now how can we apply this to the tables? Should all of the tables stretch out 100%, or should they all be one size? Do we need <br> between things such as February 17,<br>2016 or shouldn't we use it? Please for those who are reading this, choose one format or the other. - Knowledgekid87 (talk) 17:24, 17 February 2016 (UTC)

Most recent Utah poll

Dan Jones & Associates [1]
Margin of error: ± ?
Sample size: 625
February 10-15, 2016
Hillary Clinton
51%
Bernie Sanders
44%

The poll was conducted among a sample of 625 Utahns, all of whom were were asked about both the Democratic and Republican races. This is already the way that the polls in this section from DJ&A are listed; as among the Democrats interviewed, not all Utahns. RealClearPolitics and FiveThirtyEight also cite the poll with Hillary Clinton 51%-44% Bernie Sanders. (From an earlier editor)

That's not true: the previous Dan Jones & Associates poll from November shows the result for all Utahns. Perhaps because there are five times as many independents in Utah as Democrats ... see http://www.sltrib.com/home/3227864-155/utah-dems-will-be-able-to (From 172.96.33.214)

Okay so my question is what is the deal with this poll, and why is it causing so much confusion? - Knowledgekid87 (talk) 15:15, 23 February 2016 (UTC)

Fixed for further clarification. The 31/30 poll from September is listed correctly, as it shows the result only among Democratic voters; the November poll I just corrected, which was citing all Utahns; and the newest poll I also corrected to show only among Democratic voters for the same rationale as I've previously mentioned. This is already how other DJ&A polls were cited on this page and by pollsters, regardless of the inclusion of Republican voters who were asked these questions as well. Oversteek (talk) 15:58, 23 February 2016 (UTC)

DJA polls

Please don't revert the Utah poll again to show Sanders 41/Clinton 19. The poll was conducted among a sample of 625 Utahns, and those surveyed were asked about candidates in both the Republican and Democratic races. It doesn't make sense to use the 41/19 number unless all Utahns caucus for both the Democrats and Republicans; it's the same sample that's surveyed. (See the image and article below for clarification.)

• https://i.imgur.com/ZNvXuEA.png
• http://utahpolicy.com/index.php/features/today-at-utah-policy/8602-poll-cruz-sanders-tops-among-utah-voters

Oversteek (talk) 05:41, 23 February 2016 (UTC)

Exactly! Just look at how the two major pollsters RealClearPolitics and FiveThirtyEight cite the poll:
http://projects.fivethirtyeight.com/election-2016/primary-forecast/utah-democratic/#polls-only
http://www.realclearpolitics.com/epolls/2016/president/ut/utah_democratic_presidential_caucus-5766.html Ryopus (talk) 23:01, 23 February 2016 (UTC)

How do you edit the map?

I'm trying to edit the map to show that Ohio is not lead by Clinton, but is instead split between both, due to the latest poll which has Sanders ahead by 1. How do you easily edit the map? — Preceding unsigned comment added by Jose18641 (talk • contribs) 18:55, 24 February 2016 (UTC)

Sanders isn't ahead by one as the most recent poll was conducted from more recent dates. - Knowledgekid87 (talk)

Inclusion of Overtime Politics polls

Input to establish a consensus on the topic would be appreciated; see the discussion here. 108.2.58.56 (talk) 22:14, 6 January 2016 (UTC)

I believe someone removed the poll from the article. Should we restore it until there is consensus over at the GOP talk page with regards to the reliability of the poll..? Prcc27 (talk) 21:46, 11 January 2016 (UTC)

Doesn't hurt to put it back up. At least until we can see if they are accurate.

Bumping this; RfC initiated at Talk:Statewide opinion polling for the Republican Party presidential primaries, 2016#RfC on inclusion of Overtime Politics polls. 108.2.58.56 (talk) 18:09, 25 January 2016 (UTC)

Tremoe777 (talk) 06:28, 25 February 2016 (UTC)Overtime should stay on. They have proved the most accurate so far in the democratic race. I am curious why weighted demographics don't account for this accuracy, and what actually does? Maybe a note could be made of this next to the polls to clarify to visiting readers, and therefore educate.Tremoe777 (talk) 06:28, 25 February 2016 (UTC)

RfC was closed. No map changes on the Republican article, and only change for the Democratic article is that Missouri should be green. 108.2.69.172 (talk) 00:51, 26 February 2016 (UTC)

3 months

Should we exclude polls on the map that are older than 3 months? Polls from more than 3 months ago aren't really that relevant. Prcc27💋 (talk) 04:56, 24 February 2016 (UTC)

I disagree for now, wait for when super Tuesday is over. - Knowledgekid87 (talk) 14:11, 24 February 2016 (UTC)

What does Super Tuesday have to do with anything..? Prcc27💋 (talk) 07:59, 25 February 2016 (UTC)

Update: The Republican article now excludes polls older than 3 months from the map and I feel like we should do the same for consistency. Plus, waiting for an arbitrary event like Super Tuesday is a merit-less reason to include polls that are 6 months old and likely to be inaccurate. Prcc27💋 (talk) 06:22, 26 February 2016 (UTC)

Debating the Neutrality of the Article

It seems as though the article is slanted pro-Hillary Clinton, as seen by:

-Map not updating to reflect current trends (already discussed in different topic) -Removal of various poll results (Overtime Politics poll, Quinnipiac poll, etc.) that show Bernie Sanders ahead of, statistically tied with, or narrowing the gap on Hillary Clinton's lead (thereby making Hillary Clinton appear to be leading by more than she actually is) — Preceding unsigned comment added by 96.27.140.6 (talk) 18:05, 26 February 2016 (UTC)

Please assume good faith. By the way, Overtime Politics's reliability has been questioned by many users and I personally don't think the poll is reliable. Prcc27💋 (talk) 18:18, 26 February 2016 (UTC)

Overtime Politics has absolutely demonstrated their accuracy so far. Check out their predictions for Iowa, New Hampshire, and Nevada. It was just about perfect, and totally impressive compared to every other pollster on the democrat side. Removing them from this page looks like a mistake. Why don't you personally think the poll is reliable? How could you disregard their outstanding record so far? — Preceding unsigned comment added by 66.92.48.168 (talk) 04:18, 27 February 2016 (UTC)

Considering Iowa and Nevada didn't even release their popular vote totals, we may never know how close Overtime Politics came to the actual results. But honestly, that's besides the point, because correlation does not imply causation. The current consensus is to not include the poll, so it should not be included until they are mentioned in more reliable sources. Prcc27💋 (talk) 04:29, 27 February 2016 (UTC)

They predict delegates, not popular vote. We do know "how close they came to the actual results." You mention the republican consensus, but you cannot deny that their record is near perfect so far in the democrat race. Correlation does not imply causation, but do you think three states in a row is a coincidence? Is there any other pollster this year with such a fine record? You argue that the IA and NV popular vote is not released, but they do not "predict" popular vote the day before the vote. They did not attempt to predict the popular vote in NH. They predict the delegate totals the day before the vote, based on their polling, in order to demonstrate their methodology best as they can to us skeptics. Check out their website for yourself. They are not mentioned in reliable sources because they are new and unknown. That does not mean they are useless. If you want reliability, look at their record so far, which is quite a bit more important than waiting for a reliable source. The purpose of Wikipedia is to document useful information. (I agree with the person above -- the fact that these now-deleted polls are uniquely showing Sanders ahead is mysterious.) — Preceding unsigned comment added by 75.144.246.234 (talk) 21:34, 29 February 2016 (UTC)

There was no delegate prediction in SC from Overtime Politics, but their popular vote polling three days earlier was right around the average of everyone else's SC numbers, so that is beside the point with regard to their "predictions." — Preceding unsigned comment added by 75.144.246.234 (talk) 21:37, 29 February 2016 (UTC)

This map is only concerned with popular vote polling, not delegates. Prcc27💋 (talk) 20:18, 2 March 2016 (UTC)

Level up for the states?

The page looks properly on mobile phones, however it is awful to scroll long to get to state X. I'd like to level up all state sections on the content page. Do you agree? --HarisKoeln (talk) 13:53, 9 March 2016 (UTC)

New map

@Chessrat: I am mixed about the usage of the new map. On one hand I agree that it is better that the map reflect on the current states in play, but in the end the map will turn into a useless all shaded grey color as the primaries finish. For historical purposes I feel that File:Polling map for the 2016 Democratic primaries.svg should continue to be updated, and eventually replace the eventually to be all grey map. - Knowledgekid87 (talk) 17:17, 10 March 2016 (UTC)

Good point. However, I think that, even for historical purposes, it makes little sense to use File:Polling map for the 2016 Democratic primaries.svg. That map only displays the results of the most recent poll to be held prior to the primary in question, and does not display any information about the polls held prior to that. This is fine for a map that's regularly being updated, as it ensures the latest available data is being shown, but for a historical article such a map is meaningless. For example, on that map Massachusetts is shown in striped Clinton/Sanders color, because the most recent poll was within the margin of error, despite the fact that most polls and the actual primary results showed a narrow victory for Clinton. Is this really the most accurate way to present the data?
Instead, after the primaries have finished, I would support using either File:Democratic Party presidential primaries results, 2016.svg (the actual primary results), or a new map that shows aggregate polling scores, or no map at all. Chessrat ^{(talk,contributions)} 17:44, 10 March 2016 (UTC)

I can't see people wanting a map of old polling data. They'd want to see timeline graphs and data, not a stitched up pile of arbitrary polling results from whenever. As a tool for historians, it's silly. The premise that this map could be helpful to gauge current or future outcomes is flawed as well. It's actually quite misleading in that regard. In short, the map is absolutely useless. Yet, if you look up statewide polls, google directs here first. — Preceding unsigned comment added by 137.54.125.244 (talk) 18:48, 10 March 2016 (UTC)