Wikipedia:Dispute resolution noticeboard/Archive 122
This page is an archive. Do not edit the contents of this page. Please direct any additional comments to the current main page. |
Talk:Statewide opinion_polling_for_the_Democratic_Party_presidential_primaries,_2016#Fix_the_map
Have you discussed this on a talk page?
Yes, I have discussed this issue on a talk page already.
Location of dispute
- Talk:Statewide opinion polling for the Democratic Party presidential primaries, 2016#Fix the map
Users involved
Dispute overview
There were two polls conducted in Washington both within a day of each other. The PPP poll says Clinton has a 33% lead (margin of error: ± 5%) and the Gravis Marketing poll that was taken the next day says she has a 9% lead (margin of error: ± 6%). There is a map on the article that says Clinton's lead in Washington is less than 10%. But I wanted to stripe Washington with two colors to reflect that there is also another poll that says she has a 30-49% lead. The other user thinks that only the most recent poll should be used (even if it was taken the day after the second most recent poll), and striping should only be reserved for ties.
Have you tried to resolve this previously?
We've been discussing it at the article's talk page.
How do you think we can help?
I think we need you guys to weigh in on whether or not striping Washington is appropriate. Also, you guys might be able to find a way we can compromise.
Summary of dispute by Nitroxium
As explained in the talk page, the article for the democratic primaries has been following the same format as the republicans in using the most recent polls for coloring the map. Prcc27 proposes we utilize stripes on the maps to show what he considers to be "conflicting polls." However, stripes in the statewide opinion polls (Repub and democrat) are being used for virtual ties. Beyond this, saying that the two polls are conflicting would be
"First, you can't compare the results in PPP with the polls for Gravis including Elizabeth Warren. People could very easily switch from Clinton to Warren if she was an option in the primaries and there is absolutely no conflict there. Therefore, we must compare the results of PPP with the Gravis results WITHOUT Warren. In the PPP poll, she has 57% with a margin of error of 5%, meaning it could be a support of 52%. In the Gravis poll, she has a support of 45% with a margin of error of 6%, meaning it could be 51%. Likewise with Sanders, in PPP he has 24% which could be 29% and in Gravis he has 36% which could be 30%. Hence why if there was a change of 1% of support from Clinton to Sanders during the next few days (which is completely plausible), these two polls are not conflicting. I must add, the Gravis poll without Warren still includes De Blasio, which means there doesn't even have to be a 1% change of support in the next days. Clinton could have lost 2% to De Blasio in the Gravis poll. There's many possibilities."
Talk:Statewide opinion_polling_for_the_Democratic_Party_presidential_primaries,_2016#Fix_the_map discussion
@Prcc27 and Nitroxium: I'm seeing extensive talk page discussion and am happy to take this case. I'll do a little more brushing up on the issues before asking a few questions. Thanks much, North of Eden (talk) 18:09, 31 July 2015 (UTC)
- I am going to be working on concluding the English football dispute above, but once that's concluded I will give my full attention to this issue. I plan to put up some substantive comments tomorrow morning in this section. Thanks, North of Eden (talk) 22:03, 31 July 2015 (UTC)
- @Prcc27 and Nitroxium: Correct me if I'm wrong, but I think the basic issue is whether stripes should indicate conflicting polling results or ties between candidates. In this event, I think it's crucial to look at the Republican primary page and at precedent for this page. The same-sex marriage page may be helpful as a guide, but opinion polling on social issues and candidate polling are pretty different things. North of Eden (talk) 19:04, 1 August 2015 (UTC)
- They are different things, but the issue the same-sex marriage map had is the same one the Democratic map has. I believe one user even said is was "fraud" to leave out conflicting polls on that map. It does take longer for a state to trend a certain way on social issues than opinion of candidates, but it's near impossible for a candidate's lead to change drastically in one day just like it is near impossible for a state to change drastically on social issues (with a few exceptions like President Obama coming out in favor of same-sex marriage) in 1 month. Just because the poll that says Clinton has a 9% lead was taken 1 day after a poll said she has a 33% lead doesn't mean her lead is less than 10%. I don't know if I'd call it "fraud" to leave out stripes for conflicting results, but it's certainly misleading- whether it's on a map for opinion polling on social issues or a map for opinion polling on candidates. It definitely won't hurt the map to provide more information- it would actually help the map. Prcc27 (talk) 20:28, 1 August 2015 (UTC)
- Again, saying they are conflicting polls is ) 03:34, 2 August 2015 (UTC)
- Pardon me, but unfortunately I don't even know what your second and third sentences are referring to.. Nonetheless, I feel like what I said on the article's talk page applies to your statement on here too: "You already explained to me that both polls are accurate with each other within 1%, but that's if and only if Clinton has a 21% lead which the map does not have Washington colored as. And it's not the polls that are conflicting, it's that the map doesn't take into account the margin of error, and quite frankly a footnote isn't enough. If readers see Washington striped with two colors and they see the margin of error footnote- they will be more likely to scroll down to find out why Washington is striped and to find out what those margin of errors for Washington are. Then they will be able to come to the conclusion that Clinton has about a 21% lead (which is pretty much what you concluded from the polls yourself) or they could come to the conclusion that Clinton does in fact have a less than 10% lead because the poll that says so is more recent. Regardless, purposely leaving out information on the map adds ) 04:24, 2 August 2015 (UTC)
- I understand concerns about WP:OR and simiilar issues, of course). Thanks, North of Eden (talk) 15:12, 2 August 2015 (UTC)
- Well the current map isn't set up for ties because unlike the GOP map- the Democratic map goes into detail about what percentage a candidate is leading by. If let's say Sanders and Clinton were tied in Vermont it wouldn't make sense to stripe Vermont "Sanders ahead, <10%" and "Clinton ahead, <10%" because neither candidate is ahead- both are tied. Instead, we could simply add a new color for tied states like blue or red. Prcc27 (talk) 03:39, 3 August 2015 (UTC)
- That sounds like an interesting idea; what do you think, User:Nitroxium? One of the nice things is we have more leeway than we would with the Republican page, as there's only two serious contenders, and only one who is experiencing polling variations of the type we're discussing. I'm interested to hear Nitroxium's perspective on this, so I can get a better handle on where we should proceed with our discussion. Thanks, North of Eden (talk) 00:31, 4 August 2015 (UTC)
- The whole basis of this discussion has been that the polls are conflicting and that simply cannot be 100% concluded from the information given. Therefore, any conclusion on them being conflicting would be WP:OR. As shown in the talk page of the article, the two polls are not necessarily conflicting but rather could be within a margin of error (ie. Clinton could have lost two points because of De Blasio in Washington). Therefore, drawing an arbitrary line of mixing two poll results together from two different sources just because they are a few days close to each other does not make sense to me. The poll articles (Both democrat and republican) have been absolutely clear on the fact that the map represents the absolutely most recent poll, despite the color possibly being subject to change due to margin of error or not. There are many intricacies about polls, survey methodology and pollsters to mix them into the same bag arbitrarily. I propose an alternative which is to add a footnote leaving a very clear message about the map representing the absolutely most recent poll and an invitation to scroll down and read the section of each state for further information. Nitroxium (talk) 01:16, 4 August 2015 (UTC)
- Understood. You raise a very good point and I understand your concerns about OR/improper synthesis of information. Given that the polls are only a day apart, do you think that throws a hitch in things? I can see the argument in favor of "no, it doesn't," if we're operating under a strict "recent is the most valid" rule. It sounds like this may be precedent on other pages and, if so, we should probably give it some deference but, at the same time, we shouldn't cling too closely to rules that we neglect the obvious. In this case, I think the day-apart factor is worth considering, but not necessarily the linchpin of the conversation. North of Eden (talk) 01:31, 4 August 2015 (UTC)
- Another solution I may propose is one that WP:OR. Seeing more clearly WP:What SYNTH is not, I think it may work out and we can settle this conflict once and for all. We had proposed to create our own aggregate of the polls per state, defining our own rules based on what RealClearPolitics.com does, which is make an average of all polls within a timespan of a month. I believe this would be a synthesis that would not go against the rules of Wikipedia and would be closer to what User:Prcc27 considers which is a midpoint between two polls that he considers to be conflicing. In the case of Washington, it would literally be a midpoint between the two polls. Nitroxium (talk) 01:53, 4 August 2015 (UTC)
- I think this would be a good way to resolve the disagreement, and it likely meets WP:ORIGINALSYN, as some may have concerns about aggregating polls. I would encourage proposing this on the article talk page or perhaps beginning an RfC on this to determine other editors' feedback. North of Eden (talk) 02:40, 4 August 2015 (UTC)
- I have no problem averaging polls assuming it doesn't violate Wikipedia policy. Using an aggregate from a reliable source seems fine, but making our own aggregate does seem like it would violate WP:ORIGINALSYN. If we don't average the polls I'd say striping for Washington is our best option. Prcc27 (talk) 04:27, 4 August 2015 (UTC)
- If it were to break any Wikipedia rules, I'd go back to my previous alternative proposal of adding an extra footnote. Making them stripes on the map would be just as ) 00:00, 5 August 2015 (UTC)
- I have no problem averaging polls assuming it doesn't violate Wikipedia policy. Using an aggregate from a reliable source seems fine, but making our own aggregate does seem like it would violate
- I think this would be a good way to resolve the disagreement, and it likely meets
- Another solution I may propose is one that
- Understood. You raise a very good point and I understand your concerns about OR/improper synthesis of information. Given that the polls are only a day apart, do you think that throws a hitch in things? I can see the argument in favor of "no, it doesn't," if we're operating under a strict "recent is the most valid" rule. It sounds like this may be precedent on other pages and, if so, we should probably give it some deference but, at the same time, we shouldn't cling too closely to rules that we neglect the obvious. In this case, I think the day-apart factor is worth considering, but not necessarily the linchpin of the conversation. North of Eden (talk) 01:31, 4 August 2015 (UTC)
- The whole basis of this discussion has been that the polls are conflicting and that simply cannot be 100% concluded from the information given. Therefore, any conclusion on them being conflicting would be
- That sounds like an interesting idea; what do you think, User:Nitroxium? One of the nice things is we have more leeway than we would with the Republican page, as there's only two serious contenders, and only one who is experiencing polling variations of the type we're discussing. I'm interested to hear Nitroxium's perspective on this, so I can get a better handle on where we should proceed with our discussion. Thanks, North of Eden (talk) 00:31, 4 August 2015 (UTC)
- Well the current map isn't set up for ties because unlike the GOP map- the Democratic map goes into detail about what percentage a candidate is leading by. If let's say Sanders and Clinton were tied in Vermont it wouldn't make sense to stripe Vermont "Sanders ahead, <10%" and "Clinton ahead, <10%" because neither candidate is ahead- both are tied. Instead, we could simply add a new color for tied states like blue or red. Prcc27 (talk) 03:39, 3 August 2015 (UTC)
- I understand concerns about
- Again, saying they are conflicting polls is ) 03:34, 2 August 2015 (UTC)
- @Prcc27 and Nitroxium: Correct me if I'm wrong, but I think the basic issue is whether stripes should indicate conflicting polling results or ties between candidates. In this event, I think it's crucial to look at the Republican primary page and at precedent for this page. The same-sex marriage page may be helpful as a guide, but opinion polling on social issues and candidate polling are pretty different things. North of Eden (talk) 19:04, 1 August 2015 (UTC)
- I am going to be working on concluding the English football dispute above, but once that's concluded I will give my full attention to this issue. I plan to put up some substantive comments tomorrow morning in this section. Thanks, North of Eden (talk) 22:03, 31 July 2015 (UTC)
- Just as a question to User:Prcc27: is the footnote proposal at all acceptable to you? It's certainly fine if it wouldn't be, but just wondering. North of Eden (talk) 00:20, 5 August 2015 (UTC)
- No because even though technically all the information would be there- the map would be (and currently is) visually misleading! Prcc27 (talk) 01:03, 5 August 2015 (UTC)
- The map would not be misleading as all the information would be there. You are acting as if people look at the map and can't look down several pixels. It is the only solution that doesn't break ) 02:30, 5 August 2015 (UTC)
- I already explained it would be an WP:OR to say Clinton has a less than 10% lead without the source saying so explicitly using your logic! The map is so unreliable right now, especially since the map doesn't take into account for margin of errors and you only want to account for margin of error when it's close to a tie. I think all of our problems would be solved: Washington, New Hampshire (which has a statistical tie), etc. if we made the map like the GOP map. The percentages on the map aren't doing anyone good and the GOP map doesn't have them. They might have been necessary when Clinton lead in every single state in the country. But now we learned that Sanders had a lead in Vermont all this time, and New Hampshire has a statistical tie. So two states wouldn't be colored green if we got rid of the percentages. And we wouldn't have to worry about what to color Washington! Prcc27 (talk) 02:44, 5 August 2015 (UTC)
- I will repeat myself for the umpteenth time, no poll is being treated as more reliable than another. We are simply showing the absolutely most recent poll and that is as objective as we can get without breaking WP:OR. The map is reliable as the footnote states that it is not absolutely accurate. You are the only one who has been protesting over the methodology we've used, but many editors have edited the article before and followed through with it without complaint. I did not create the article and did not put the percentages, it was another editor and with good reason. The point is to show more information to the reader, since Clinton has been the commanding leader in the primaries it interests to know by how much. I have had no problem in following through with the methodology currently used unless the race becomes tighter. Nitroxium (talk) 02:58, 5 August 2015 (UTC)
- As North of Eden said, it's not WP:OR if it's an obvious truth. It's obvious that a poll isn't more reliable if it's only more recent by a day! You and I are the only ones who have been talking about changing the map since we colored Vermont for Sanders, thus the map is different when other people edited it because Clinton isn't the only candidate that leads in a state. So having a percentage map might have worked somewhat in the past, but now we have another candidate and it's becoming pointless. And there's nobody else around to weigh in on whether we should get rid of the percentages due to Clinton not leading in every state anymore. It's beyond stupid that when New Hampshire is within the margin of error for a tie you decided to stripe the map and threaten to report me for vandalism if I revert you again. Yet, we don't stripe South Carolina even though when you take the margin of error into account Clinton could either have a >69% lead or a 50%–69%. So how come South Carolina is colored for a >69% lead when the margin of error might suggest otherwise? Why is NH the only state that deserves to be striped to take the margin of error into account? It's inconsistent to only take the margin of error into account sometimes (for statistical ties). Did we use a percentage map for the 2008 democratic primary? Do we use a percentage map for the GOP primary? No. At what point is Clinton's lead no longer "commanding"? Because in April 2007 Clinton lead in every state except for 5 states. Was that a "commanding" lead? If so, then why didn't we use a percentage map in 2007? Prcc27 (talk) 03:33, 5 August 2015 (UTC)
- Also, why do you consider New Hampshire a statistical tie but not Washington..? Washington is within the margin of error of being tied using the most recent poll. Prcc27 (talk) 03:49, 5 August 2015 (UTC)
- As North of Eden said, it's not
- I will repeat myself for the umpteenth time, no poll is being treated as more reliable than another. We are simply showing the absolutely most recent poll and that is as objective as we can get without breaking
- I already explained it would be an
- The map would not be misleading as all the information would be there. You are acting as if people look at the map and can't look down several pixels. It is the only solution that doesn't break ) 02:30, 5 August 2015 (UTC)
- No because even though technically all the information would be there- the map would be (and currently is) visually misleading! Prcc27 (talk) 01:03, 5 August 2015 (UTC)
- @WP:AN/Edit warring. I have carefully read your postings from earlier today and appreciate the diligence you've put into them. Just as a thought experiment, which may help editors in defusing the current conflict as we proceed, what do each of you think are the other editor's strongest arguments? Essentially, if you were arguing the other editor's point of view, which of his or her arguments might you use or expand on? Sorry if this seems infantile, but I think it may help us make progress here. North of Eden (talk) 22:20, 5 August 2015 (UTC)
- Their strongest argument is that we should follow the format of the GOP map. And for the most part I agree with them. I obviously still think that polls that were taken 1 day before the most recent one shouldn't be disregarded and should be reflected on the map. But if we switched to a map that only had three categories: "Clinton lead" (dark green), "Sanders lead" (dark blue), and "Clinton and Sanders tied" (dark gray or striped green/blue)- the dispute we are having with Washington would disappear since we would just color it as "Clinton lead". Since the GOP map doesn't differentiate by what percentage a candidate is leading by, nor did the 2008 Democratic primary map- even though Clinton had a commanding lead and was leading in every state except for 5- I feel like we should follow those maps' formats and not use percentages. The margin of errors are very problematic because a state could possibly qualify as different colors based on only one single poll when margin of error is accounted for. I still don't understand why NH is striped for margin of error, but SC isn't striped to account for margin of error. But a map without percentages would fix the margin of error problem we are having now. Prcc27 (talk) 00:05, 6 August 2015 (UTC)
- Removing percentages would certainly render this dispute moot. But I would imagine the rationale for the percentages is that Clinton is the clear frontrunner, so the issue is generally whether she has a commanding lead or not, as opposed to whether she or Sanders is polling higher in a given state. @Nitroxium: what are your thoughts? North of Eden (talk) 18:13, 6 August 2015 (UTC)
- I think that Prccc's point in that the map is not completely accurate is completely right and that is why we need to point that out in a footnote to the readers, as we have done so. I am even proposing going even further and will simply go ahead and add that footnote. However, Prccc is proposing to change completely the format that has been used by myself and many other editors since the creation of the article which had established said format because of the particularity of Clinton's commanding lead. To remove the percentages would be to make the map completely useless to the reader. I think we can work off the basis that the map is not completely accurate and work towards making the reader understand that through a simple footnote, that won't change the entire format that has been until now accepted unanimously, that will solve any issues of innacuracies and that has provided more information. Nitroxium (talk) 19:58, 6 August 2015 (UTC)
- "Prccc (sic) is proposing to change completely the format that has been used by myself and many other editors since the creation of the article which had established said format because of the particularity of Clinton's commanding lead." Umm.. the the map's format was not set up that way because of her commanding lead. The reason why the map was set up that way was because when it was first created, she had a lead in every single state in the nation (except Vermont, but it was inaccurately colored in her favor at the time). Now, there are two candidates that are leading in at least 1 state. In August 2007 Clinton had just as much of a commanding lead as she does now in 2015. She had a lead in 29 states compared to her current lead in 27 states. If a map without percentages can be used in 2007 when she had a commanding lead- a map like that can also be used now since Clinton's situation is similar then as it is now! "I think we can work off the basis that the map is not completely accurate and work towards making the reader understand that through a simple footnote". A footnote isn't going to fix the inaccuracies of a map; removing the percentages will. A footnote would pretty much be telling readers "this map is inaccurate! Please read article to find out about conflicting results and margin of errors". And if you truly did support footnotes, you would have left New Hampshire alone and let the margin of error footnote explain things instead of striping it without consensus. If South Carolina ins't striped for margin of error then neither should New Hampshire. Also, as you said yourself- all the information on the map is in the article. If a reader can find out about the percentages by themselves on the GOP article, they can do so on the Democratic article too. Prcc27 (talk) 23:18, 6 August 2015 (UTC)
- Including the percentages likely increases the overall accuracy and usefulness of the map, which is a plus. However, it sounds as if the precedent in similar Democratic Party articles, such as the one from 2007, is to use a map without percentages. On another note, I've put some thought into the rule that only the most recent poll is used when determining each state's coloration. Generally, I think this is a very good rule. There needs to be a way to determine the state's coloration without veering into WP:RS by including information which is easily impeachable, as it is contradicted by info from an equally reliable and recent source (i.e. the May 14-17 poll in Washington). So I guess the question now is, how do we get around these obstacles? How do we satisfy both OR and RS policies while retaining consistency and accuracy on the map? North of Eden (talk) 00:51, 7 August 2015 (UTC)
- Well a map with percentages may be useful but how the map is currently set up is not accurate. The map doesn't take the margin of error into account unless a state is in the margin of error of a tie because Nitroxium striped NH, but not SC and other states with conflicting results when margin of error is taken into account! I think we all agree that we should be consistent throughout the articles so since neither the GOP nor the Democratic map from 2007 used percentages I think that's a strong reason for getting rid of them. Should we possibly go to another noticeboard to find out how not to violate them i.e. Wikipedia:Reliable sources/Noticeboard & Wikipedia:No original research/Noticeboard, or should we only figure things out on this noticeboard? Prcc27 (talk) 01:32, 7 August 2015 (UTC)
- The RS noticeboard might be useful; I'm not sure how active the other one is. If you'd like to try that out to vet sources, you are more than welcome to do so. I am happy to keep this discussion open and continue working to resolve the dispute here, perhaps integrating the other processes as well. North of Eden (talk) 03:19, 7 August 2015 (UTC)
- Done. North of Eden, if you are confused about what I mean when I say the margin of error is only being taken into account when there is a statistical tie- let me know! Also, I think removing the percentages from the map is getting closer to having consensus; especially since another state was added to the Clinton/Sanders tied column. @Nitroxium:, do you have anything to say to try to refute the points I made above about removing the percentages? Prcc27 (talk) 04:03, 8 August 2015 (UTC)
- The RS noticeboard might be useful; I'm not sure how active the other one is. If you'd like to try that out to vet sources, you are more than welcome to do so. I am happy to keep this discussion open and continue working to resolve the dispute here, perhaps integrating the other processes as well. North of Eden (talk) 03:19, 7 August 2015 (UTC)
- Well a map with percentages may be useful but how the map is currently set up is not accurate. The map doesn't take the margin of error into account unless a state is in the margin of error of a tie because Nitroxium striped NH, but not SC and other states with conflicting results when margin of error is taken into account! I think we all agree that we should be consistent throughout the articles so since neither the GOP nor the Democratic map from 2007 used percentages I think that's a strong reason for getting rid of them. Should we possibly go to another noticeboard to find out how not to violate them i.e. Wikipedia:Reliable sources/Noticeboard & Wikipedia:No original research/Noticeboard, or should we only figure things out on this noticeboard? Prcc27 (talk) 01:32, 7 August 2015 (UTC)
- Including the percentages likely increases the overall accuracy and usefulness of the map, which is a plus. However, it sounds as if the precedent in similar Democratic Party articles, such as the one from 2007, is to use a map without percentages. On another note, I've put some thought into the rule that only the most recent poll is used when determining each state's coloration. Generally, I think this is a very good rule. There needs to be a way to determine the state's coloration without veering into
- "Prccc (sic) is proposing to change completely the format that has been used by myself and many other editors since the creation of the article which had established said format because of the particularity of Clinton's commanding lead." Umm.. the the map's format was not set up that way because of her commanding lead. The reason why the map was set up that way was because when it was first created, she had a lead in every single state in the nation (except Vermont, but it was inaccurately colored in her favor at the time). Now, there are two candidates that are leading in at least 1 state. In August 2007 Clinton had just as much of a commanding lead as she does now in 2015. She had a lead in 29 states compared to her current lead in 27 states. If a map without percentages can be used in 2007 when she had a commanding lead- a map like that can also be used now since Clinton's situation is similar then as it is now! "I think we can work off the basis that the map is not completely accurate and work towards making the reader understand that through a simple footnote". A footnote isn't going to fix the inaccuracies of a map; removing the percentages will. A footnote would pretty much be telling readers "this map is inaccurate! Please read article to find out about conflicting results and margin of errors". And if you truly did support footnotes, you would have left New Hampshire alone and let the margin of error footnote explain things instead of striping it without consensus. If South Carolina ins't striped for margin of error then neither should New Hampshire. Also, as you said yourself- all the information on the map is in the article. If a reader can find out about the percentages by themselves on the GOP article, they can do so on the Democratic article too. Prcc27 (talk) 23:18, 6 August 2015 (UTC)
- I think that Prccc's point in that the map is not completely accurate is completely right and that is why we need to point that out in a footnote to the readers, as we have done so. I am even proposing going even further and will simply go ahead and add that footnote. However, Prccc is proposing to change completely the format that has been used by myself and many other editors since the creation of the article which had established said format because of the particularity of Clinton's commanding lead. To remove the percentages would be to make the map completely useless to the reader. I think we can work off the basis that the map is not completely accurate and work towards making the reader understand that through a simple footnote, that won't change the entire format that has been until now accepted unanimously, that will solve any issues of innacuracies and that has provided more information. Nitroxium (talk) 19:58, 6 August 2015 (UTC)
- Removing percentages would certainly render this dispute moot. But I would imagine the rationale for the percentages is that Clinton is the clear frontrunner, so the issue is generally whether she has a commanding lead or not, as opposed to whether she or Sanders is polling higher in a given state. @Nitroxium: what are your thoughts? North of Eden (talk) 18:13, 6 August 2015 (UTC)
- Their strongest argument is that we should follow the format of the GOP map. And for the most part I agree with them. I obviously still think that polls that were taken 1 day before the most recent one shouldn't be disregarded and should be reflected on the map. But if we switched to a map that only had three categories: "Clinton lead" (dark green), "Sanders lead" (dark blue), and "Clinton and Sanders tied" (dark gray or striped green/blue)- the dispute we are having with Washington would disappear since we would just color it as "Clinton lead". Since the GOP map doesn't differentiate by what percentage a candidate is leading by, nor did the 2008 Democratic primary map- even though Clinton had a commanding lead and was leading in every state except for 5- I feel like we should follow those maps' formats and not use percentages. The margin of errors are very problematic because a state could possibly qualify as different colors based on only one single poll when margin of error is accounted for. I still don't understand why NH is striped for margin of error, but SC isn't striped to account for margin of error. But a map without percentages would fix the margin of error problem we are having now. Prcc27 (talk) 00:05, 6 August 2015 (UTC)
- @Prcc27: I think I get what you're saying (by all means correct me if I'm getting it wrong!). Your view, as I understand it, is that margins of error are used when determining which states should be striped as ties. They are not, however, used when determining the state's coloration if it's a Clinton-percentage state? North of Eden (talk) 14:56, 8 August 2015 (UTC)
- I have also given this matter more thought and think that it wouldn't break WP:OR if we put stripes based on the margins of error (ex. Hillary above 50-69 and 69+ at the same time) but only based on the most recent polls. Is that okay with Prcc27? Nitroxium (talk) 22:13, 8 August 2015 (UTC)
- I assume, if accepted by User:Prcc27, this would resolve the dispute at hand? North of Eden (talk) 15:10, 9 August 2015 (UTC)
- If they support striping Washington for both polls then yeah. Not sure if that's what they meant from their comment. While I do prefer striping states to account for margin of error I feel like with all these different shades of green and stripes the map will be too cluttered. Getting rid of the percentages altogether would make the map cleaner and more consistent with other maps. But I will probably propose this on the article's talk page since if Nixtroxium concedes that Washington should be striped- the dispute is resolved. Prcc27 (talk) 12:16, 10 August 2015 (UTC)
- That sounds excellent, but I'd like the striping resolution to be confirmed by Nitroxium before this is closed. Thanks, North of Eden (talk) 00:18, 11 August 2015 (UTC)
- Just to be clear, Washington would be triple striped to account for the margin of error for both polls (which isn't very desirable for me), but at least both polls would be included. Prcc27 (talk) 00:55, 11 August 2015 (UTC)
- As we have discussed at length before, using both polls would be WP:ORIGINALSYN. I am saying to use the most recent polls with margin of error included and to add a footnote clarifying that they are the utmost recent poll. Nitroxium (talk) 01:31, 11 August 2015 (UTC)
- Well then the dispute is not settled. so far there is no response from the reliable sources noticeboard, but I just commented about this dispute on the original research noticeboard. Prcc27 (talk) 04:52, 11 August 2015 (UTC)
- A user on the reliable sources noticeboard said "this doesn't appear to be a reliability issue so much as it could be a ) 23:02, 11 August 2015 (UTC)
- As we have discussed at length before, using both polls would be
- Just to be clear, Washington would be triple striped to account for the margin of error for both polls (which isn't very desirable for me), but at least both polls would be included. Prcc27 (talk) 00:55, 11 August 2015 (UTC)
- That sounds excellent, but I'd like the striping resolution to be confirmed by Nitroxium before this is closed. Thanks, North of Eden (talk) 00:18, 11 August 2015 (UTC)
- If they support striping Washington for both polls then yeah. Not sure if that's what they meant from their comment. While I do prefer striping states to account for margin of error I feel like with all these different shades of green and stripes the map will be too cluttered. Getting rid of the percentages altogether would make the map cleaner and more consistent with other maps. But I will probably propose this on the article's talk page since if Nixtroxium concedes that Washington should be striped- the dispute is resolved. Prcc27 (talk) 12:16, 10 August 2015 (UTC)
- I assume, if accepted by User:Prcc27, this would resolve the dispute at hand? North of Eden (talk) 15:10, 9 August 2015 (UTC)
- I have also given this matter more thought and think that it wouldn't break
- @original synthesis. Ultimately, these are the options I think we've determined:
- 1. Keep the map in its current state.
- 2. Remove percentages from the map altogether; keep using striping to denote margin-of-error issues between candidates.
- 3. Use striping to denote margin-of-error issues for a single candidate, but only using the most recent poll (Nitroxium's recent proposal)
- 4. Use striping to denote conflicting polls for a single candidate (presumably with consideration given to margin-of-error issues)
- Obviously, if folks have alternative ideas, or if you want to clarify these, let me know. Thanks, North of Eden (talk) 01:04, 12 August 2015 (UTC)
- I support #2 and proposed it on the article's talk page since it's separate from the original dispute. Prcc27 (talk) 02:13, 12 August 2015 (UTC)
- Number 3, there's no reason we should sacrifice an opportunity to give more information to the reader just because Prcc does not agree with more information being given to the reader. #3 while giving more information also fixes the issue. Meanwhile, option #2 does not fix the issue and I find it strange for Prcc to support it, since in the case of two polls coming out on the same day (One showing a statistical tie, another showing a clear winner), the dispute wouldn't be resolved at all. I am sensing bias here. Nitroxium (talk) 04:57, 12 August 2015 (UTC)
- @Nitroxium: I am not bias. My support for #2 is entirely separate from what this dispute is about: what we do if two conflicting polls come out on the same day or one or two days within each other. If #2 becomes the new map I would still be in favor of reflecting both polls that come out within a day of each other. But obviously we'd have to change the coloring. If New Hampshire had two polls come out a day within each other- one saying Sanders leads the other saying there's a statistical tie, I would stripe the state dark gray (tie) and blue (Sanders lead). North of Eden, since my proposal to get rid of percentages is separate from the original dispute of what to do when there are conflicting polls- do we have to discuss that proposal on the article's talk page first before bringing the dispute here or are we allowed to have that exclusively discussed here? But I will note that if a user goes on the article's talk page and supports my proposal to get rid of percentages that would most likely mean it has consensus and it wouldn't need to be discussed here. Prcc27 (talk) 05:19, 12 August 2015 (UTC)
- I don't have time to provide full comments right now, but I will later today. In the meantime, it is perfectly fine to discuss the percentages proposal here. Thanks, North of Eden (talk) 13:08, 12 August 2015 (UTC)
- There are now 4 states that aren't solid Clinton support. That's 4 times as many states as when Nitroxium initially said that Clinton's lead was commanding. Now, Clinton's lead can no longer be called "commanding" because there are several states where she is not leading. As they said themselves, all the information is already in the wikipedia article. Since the GOP map doesn't provide all the information i.e. percentages I think the readers on the Democratic article can manage. Prcc27 (talk) 22:45, 12 August 2015 (UTC)
- 4 out of 50 is completely commanding, and proposal #2 shouldn't even be considered since this dispute is about "conflicting polls". #2 does nothing to fix that.Nitroxium (talk) 23:18, 12 August 2015 (UTC)
- @Nitroxium: I think you mean 4 out of 30 since 20 states were not even polled! For the last time, in 2007 Clinton's lead was just as remarkable and there were no percentages on that map. When exactly is Clinton's lead no longer commanding? She is tied in a vital early state and I hardly find that "commanding". North of Eden said we can discuss both issues about the map here. Otherwise, proposal #2 would have consensus since I'm the only one who commented on the article's talk page about it! Prcc27 (talk) 23:28, 12 August 2015 (UTC)
- I'm tired, I'll let North of Eden respond and see his opinion. The fact that it was done in the past doesn't mean we can't improve. I don't see why you see the need to sacrifice being able to offer more information to the reader. #2 doesn't fix the issue of "conflicting polls" at all but it does stop showing as much information to the reader, while #3 solves both issues and doesn't sacrifice anything. Nitroxium (talk) 23:35, 12 August 2015 (UTC)
- @Nitroxium: Well I think we should have the discussion pertaining to #2 on the article's talk page since there are other users involved in the discussion there. The current !vote is 3-2 in my favor and since the discussion isn't exclusively between you and me- we can have that particular discussion on the article's talk page instead of here! Prcc27 (talk) 23:43, 12 August 2015 (UTC)
- This is dispute resolution, I don't know why you're saying to go back to the talk page if the whole reason we came here is to resolve dsiputes in the talk page? Please just allow North of Eden to respond before reverting my edits for the umpteenth time. Nitroxium (talk) 23:46, 12 August 2015 (UTC)
- @Nitroxium: Well I think we should have the discussion pertaining to #2 on the article's talk page since there are other users involved in the discussion there. The current !vote is 3-2 in my favor and since the discussion isn't exclusively between you and me- we can have that particular discussion on the article's talk page instead of here! Prcc27 (talk) 23:43, 12 August 2015 (UTC)
- I'm tired, I'll let North of Eden respond and see his opinion. The fact that it was done in the past doesn't mean we can't improve. I don't see why you see the need to sacrifice being able to offer more information to the reader. #2 doesn't fix the issue of "conflicting polls" at all but it does stop showing as much information to the reader, while #3 solves both issues and doesn't sacrifice anything. Nitroxium (talk) 23:35, 12 August 2015 (UTC)
- @Nitroxium: I think you mean 4 out of 30 since 20 states were not even polled! For the last time, in 2007 Clinton's lead was just as remarkable and there were no percentages on that map. When exactly is Clinton's lead no longer commanding? She is tied in a vital early state and I hardly find that "commanding". North of Eden said we can discuss both issues about the map here. Otherwise, proposal #2 would have consensus since I'm the only one who commented on the article's talk page about it! Prcc27 (talk) 23:28, 12 August 2015 (UTC)
- 4 out of 50 is completely commanding, and proposal #2 shouldn't even be considered since this dispute is about "conflicting polls". #2 does nothing to fix that.Nitroxium (talk) 23:18, 12 August 2015 (UTC)
- There are now 4 states that aren't solid Clinton support. That's 4 times as many states as when Nitroxium initially said that Clinton's lead was commanding. Now, Clinton's lead can no longer be called "commanding" because there are several states where she is not leading. As they said themselves, all the information is already in the wikipedia article. Since the GOP map doesn't provide all the information i.e. percentages I think the readers on the Democratic article can manage. Prcc27 (talk) 22:45, 12 August 2015 (UTC)
- I don't have time to provide full comments right now, but I will later today. In the meantime, it is perfectly fine to discuss the percentages proposal here. Thanks, North of Eden (talk) 13:08, 12 August 2015 (UTC)
- @Nitroxium: I am not bias. My support for #2 is entirely separate from what this dispute is about: what we do if two conflicting polls come out on the same day or one or two days within each other. If #2 becomes the new map I would still be in favor of reflecting both polls that come out within a day of each other. But obviously we'd have to change the coloring. If New Hampshire had two polls come out a day within each other- one saying Sanders leads the other saying there's a statistical tie, I would stripe the state dark gray (tie) and blue (Sanders lead). North of Eden, since my proposal to get rid of percentages is separate from the original dispute of what to do when there are conflicting polls- do we have to discuss that proposal on the article's talk page first before bringing the dispute here or are we allowed to have that exclusively discussed here? But I will note that if a user goes on the article's talk page and supports my proposal to get rid of percentages that would most likely mean it has consensus and it wouldn't need to be discussed here. Prcc27 (talk) 05:19, 12 August 2015 (UTC)
- Number 3, there's no reason we should sacrifice an opportunity to give more information to the reader just because Prcc does not agree with more information being given to the reader. #3 while giving more information also fixes the issue. Meanwhile, option #2 does not fix the issue and I find it strange for Prcc to support it, since in the case of two polls coming out on the same day (One showing a statistical tie, another showing a clear winner), the dispute wouldn't be resolved at all. I am sensing bias here. Nitroxium (talk) 04:57, 12 August 2015 (UTC)
- I support #2 and proposed it on the article's talk page since it's separate from the original dispute. Prcc27 (talk) 02:13, 12 August 2015 (UTC)
@Nitroxium: Fine, but we will have to add the other users that commented on the article's talk page and the Commons talk page about proposal #2 to this dispute resolution. The original dispute is between you and me, but the #2 proposal dispute is between you, me, and 3 other users. I will wait for North of Eden to comment. But on another note, what I reverted you for has nothing to do with either dispute. That particular dispute is about what is considered a statistical tie and what is not. Do we need to have that discussion here too and ping the user to discuss here what they discussed on that matter on the Commons talk page? Prcc27 (talk) 23:53, 12 August 2015 (UTC)
- Note I am traveling today (August 14) and tomorrow (August 15), but I will be able to give my full attention to the discussion tomorrow evening. Sorry for the delay. North of Eden (talk) 11:54, 14 August 2015 (UTC)
- I originally started this dispute resolution because Nitroxium and I were the only ones discussing the issue and neither one of us could agree on what to do. However, since a user just weighed in on the article's talk page in favor of Nitroxium I'm willing to yield and only use the most recent poll when coloring a state (unless someone at the talk page backs me- then I will try to get consensus there). Prcc27 (talk) 05:33, 15 August 2015 (UTC)
- @Prcc27 and Nitroxium: That sounds good. Do we still have an agreement on striping based on margins of error per Nitroxium's earlier suggestion? North of Eden (talk) 23:41, 16 August 2015 (UTC)
- In the past few days, the disputes have been resolved, though the striping for margins of error I now have the opinion of not including them, since some reputable sources indicate that. Nitroxium (talk) 00:09, 17 August 2015 (UTC)
- @Nitroxium: I'm a little confused; what exactly do these sources indicate? That the margins of error shouldn't be noted? North of Eden (talk) 00:59, 17 August 2015 (UTC)
- I fell like the margin of errors should be noted because why would they be given in the first place if they weren't meant to be considered..? Prcc27 (talk) 04:28, 17 August 2015 (UTC)
- There's no risk of WP:OR by striping based on margins of error; in fact, all it would do is make the map more accurate. That said, we can certainly discuss this further. I'm interested to see Nitroxium's views on this. North of Eden (talk) 12:31, 17 August 2015 (UTC)
- There's no risk of
- I fell like the margin of errors should be noted because why would they be given in the first place if they weren't meant to be considered..? Prcc27 (talk) 04:28, 17 August 2015 (UTC)
- @Prcc27 and Nitroxium: That sounds good. Do we still have an agreement on striping based on margins of error per Nitroxium's earlier suggestion? North of Eden (talk) 23:41, 16 August 2015 (UTC)
- I originally started this dispute resolution because Nitroxium and I were the only ones discussing the issue and neither one of us could agree on what to do. However, since a user just weighed in on the article's talk page in favor of Nitroxium I'm willing to yield and only use the most recent poll when coloring a state (unless someone at the talk page backs me- then I will try to get consensus there). Prcc27 (talk) 05:33, 15 August 2015 (UTC)
User talk:Patalexander
This is not a content dispute, just clearly someone that want's to know why an article was speedy deleted. No reason for this to be here, just contact the editors who deleted it. If it was deleted without reason, ask the editors who deleted it why. RMS52 Talk to me 15:14, 31 August 2015 (UTC) |
Closed discussion |
---|
2015 Thalys attack
Futile, a primary party has chosen, as his/her right, to not participate at this time. See discussion below. I will note that there are no "channels" or prerequisites to filing here except for extensive discussion having occurred on the article talk page and no similar proceeding being under way. But necessity can also mean that a party feels, as would appear to be the case, that the matter may still be worked out through discussion and that's a perfectly acceptable reason, though no reason is needed, to decline participation. This may be refiled yet again, but the filer might want to take a straw poll to see if everyone is willing to participate before doing so. — TransporterMan (TALK) 13:57, 31 August 2015 (UTC) |
Closed discussion |
---|
Talk:2015 ICC_World_Twenty20_Qualifier
The volunteer handling this case has asked me as the current DRN coordinator to make a closing note. This case has devolved to an unproductive debate between one tenacious editor and the volunteer. The editor has suggested conducting a straw poll. He is, of course, free to attempt that but the proper place for it would be at the article talk page where it will gain the attention of other interested editors, not just those who have been involved here but should remember that Wikipedia is not a scoreboard or newspaper, so there's no need to keep articles such as this current on a day by day basis. — TransporterMan (TALK ) 13:58, 2 September 2015 (UTC) |
Closed discussion |
---|
Talk:Alfred de_Grazia
This case has already been declined once. Refiling this same case from an IP address doesn't change the fact that it has already been declined as explained. dispute resolution can do is to try to help parties come to consensus. — TransporterMan (TALK ) 17:06, 2 September 2015 (UTC) |
Closed discussion |
---|
Talk:Popular Republican_Union_(2007)
Going by the dispute summary of the filer, there seems to be a conduct issue. And if so, conduct issues are not resolved here and there are noticeboards like Talk 18:38, 2 September 2015 (UTC) |
Closed discussion |
---|
Talk:Direct Action_Everywhere#Edit_warring_on_this_page
Premature. Like all other moderated content dispute resolution venues at Wikipedia, DRN requires extensive talk page discussion before seeking assistance. While there's been a bit of discussion on the talk page, it's been almost all about conduct, not content. This noticeboard does not handle conduct matters (one of which is whether someone is editing in good faith). I will say a word or two at the article talk page since one issue here is pretty simple. — TransporterMan (TALK ) 00:56, 3 September 2015 (UTC) |
Closed discussion |
---|
Chris Bell_(politician)
Pending in other forum. DRN will not accept cases which are pending in other dispute resolution processes such as RFC. Please wait for the RFC to be closed, typically 30 days. If no resolution is achieved, then you may consider refiling here. — TransporterMan (TALK) 14:07, 3 September 2015 (UTC) |
Closed discussion |
---|
Talk:British Pakistanis
Resolved, as all the involved editors agreed to use both the words (British Mirpuri and British Kashmiri) in the article.
What has been agreed upon? To use both the terms in the article with the self definition of the term disclosed in the article's lead (perTalk 14:12, 3 September 2015 (UTC) |
Closed discussion |
---|
Talk:Awans of Pakistan
This dispute doesn't seem to have been extensively discussed in the article's talk page. DRN requires that the dispute be discussed extensively on a talk page before it can be moderated here. Please consider the recommendations made Talk 15:28, 7 September 2015 (UTC) |
Closed discussion |
---|
Flitfire
As stated, this is a conduct dispute, in effect an allegation of ANI for conduct disputes. Even if this had been focused on the content matters in question, however, there's not been sufficient talk page discussion to satisfy DRN's requirement of extensive talk page discussion. And let me remind everyone that discussion about an article should be on the talk page of that article so as to leave a record of it for future editors. — TransporterMan (TALK ) 14:16, 8 September 2015 (UTC) |
Closed discussion |
---|