I'm going to say this once as a Clinton supporter: the sponsored PPP poll is bullshit

[Godhumor]

Ignore it completely. And I am now speaking as a statistical analyst. It was nothing more than advertising disguised as a poll.

That said please don't start posting the internet widget instant votes as if they mean anything either. Who "won" in public perception won't be known for days, if at all.

[willvotesdem]

1. All will be revealed in the VOTING primary states like NH and SC.

[elleng]

2. Thanks. I think I've said that before to you,

and at least I pay virtually no attention to polls at this stage. Unfortunately, during our times, 'nothing' is more than nothing.

[beam me up scottie]

3. Thanks for being honest.

I don't pay attention to any of the polls, it's a matter of opinion and not worth fighting over.

K & R!

[Godhumor]

4. First and foremost I am an statistician

And that poll was designed to trick people into thinking it was sound. Actually makes me angry.

[beam me up scottie]

6. If I had your knowledge and experience I'd probably be angry all the time.

[Godhumor]

11. Heh, it was the first debate wrap up that caused me to take a GDP break last time

Not a good start post second debate either

[tex-wyo-dem]

53. Thanks, Godhumor!

Kinda calls into question all PPP polling if they are going to be using questionable/bullshit polling here.

[azurnoir]

78. Thank you

[edgineered]

8. a new funny for you bmus - correct the turntable! nt

[beam me up scottie]

12. If I admit I know what that is I'll be dating myself.

But it's too funny to pretend I don't!

[edgineered]

15. the record is fine - this turntable though,

only spins at 45, I've heard better acoustics on a weather vane!

[TDale313]

5. Thank you. n/t

[jeff47]

7. Not a statistical analyst, but this is what I spotted so far.

1) Didn't publish the screening questions.
2) Didn't rotate the candidate's names.
3) Answer for Q1 is very odd...100% of their sample watched the debate. Nobody had something come up??
4) ...I don't see any random selection of their sample in their methodology.

[hopemountain]

50. & they were screened prior to the debate. nt

[jeff47]

68. Screening would be fine if they told us exactly how they were screened

and how they created a random sample of people to screen. Since they didn't talk about how they got their random sample. So that heavily implies it wasn't a random sample.

The 100% watched in Q1 clearly indicates the sample was not random, and was set up to give the results Correct the Record wanted.

[BainsBane]

9. I thought PPP was supposed to be a top polling outfit

and highly accurate?

[Godhumor]

13. They are. But they also do sponsored polling at customer behest, including directions on conducting

In any other time this would be a "ignore all sponsored polls" post. But the PPP one came out really fast and needed to be addressed specifically.

Unfortunately, some posters are already using this to claim PPP had a pro-Hillary bias, which is an incorrect but logical outcome.

[BainsBane]

14. Then the sponsored polls aren't accurate?

They use a different methodology?

[Godhumor]

16. Customers can specify how a poll is conducted, if they wish to do so

It is one of the best kept and least savory secrets of the industry.

And this poll was designed to reach one outcome.

[BainsBane]

18. I see

I would think that would do damage to PPP's reputation.

[Godhumor]

20. Actually, it might. This is a job they should have turned down

Even Nate Silver is calling them on it:

People should know that this poll was paid for by Correct The Record, a Hillary Clinton Super PAC. https://t.co/DkOTatB8ln

https://twitter.com/NateSilver538/status/665755635288571904


Not a good day for PPP.

[Weidman]

33. So from now on, all PPP polls will be dealt with a heavy dose of skepticism.

Since Clinton's spending a lot of her money on polling while Bernie hired a pollster to do some focus testing.

I wonder what the real numbers look like. The real numbers with the 18-34 demographics, especially. The one that supposedly have 75 million Americans in that age group that may or may not vote in for Bernie.

[Godhumor]

39. And you just illustrated perfectly why PPP screwed up taking the job

It will make it harder to argue the real polls that come out are right. Just stupid on their part.

[reformist2]

21. It's kind of hard to say now whether PPP is unbiased or not. They have more $$$ to make from HRC!

[Godhumor]

22. They take money to do polling for any Democratic leaning group or organization

But they should have known to turn this one down.

And it will make people question there results going forward. I will not deny they stepped in it, big time.

[tammywammy]

10. I agree that No one should put stock into either the PPP poll or silly internet polls. n/t

[Juicy_Bellows]

17. K&R nt.

[bravenak]

19. I won't even bother to look at it in that case.

[Godhumor]

23. Don't. It is not a good night for PPP

Really disappointed in them, to be honest.

[bravenak]

25. I hate when Folks jump the shark like that.

I always wait a few days until people who understand statistics and polling let me know what's up. I defer to them. I always find the first numbers that come out right after the event to be silly.

[restorefreedom]

24. i generally look at polls for fun

the only one that matters is election day

but thanks for the honest assessment!

[joshcryer]

26. What's wrong with their methodology?

http://www.publicpolicypolling.com/pdf/2015/DemocraticPostDebatePoll111415.pdf

One of the things I didn't like about the post debates polls before is that they polled everyone after the media had their input, which would bias the people who watched (and in some cases they didn't even poll people who watched the whole thing).

[tammywammy]

27. What were the screening questions? n/t

[joshcryer]

28. "Are you a democrat planning to watch the debate and answer a poll afterward?"

What else could the question be?

[tammywammy]

30. They could have screened for people who already have a preferred candidate. n/t

[joshcryer]

35. If they did that then that's trash.

And the poll is trash. And PPP should be ashamed for not making that clear in their poll. I didn't see that when I read the poll though, that's why I asked about the methodology.

[tammywammy]

38. It's best to just ignore this poll. n/t

[Godhumor]

34. And that is the problem with sponsored polls right there

Your question is the sensible one.

But this poll produced a margin of error, which means population parameters were used which means the sample was weighted. Based on the results, I am pretty certain it was weighted by perceived current level of support for each candidate.

So more than likely the opt in questions were:

1) Are you a Democrat
2) Are you watching the debate
3) which, if any, candidate do you currently support

[joshcryer]

36. Ah-hah. Now you've convinced me.

Thanks. Makes sense.

[Godhumor]

43. No problem. Easiest response is to shrug and call it a poorly designed poll

But I think PPP took a reputation hit here on DU, at least, tonight.

[joshcryer]

47. I always have to understand the thought process.

Especially because those other polls were after the media put its mouth in to it, I like the idea of pre-screening for debate watchers only and getting their opinion right after the debate as to not be biased by the media. But if PPP did as you suggest it is pathetic.

[Godhumor]

48. I mentioned in another post, even Nate Silver called PPP out

https://twitter.com/NateSilver538/status/665755635288571904

People should know that this poll was paid for by Correct The Record, a Hillary Clinton Super PAC. https://t.co/DkOTatB8ln

-----------

I may be off on the specific particulars, but PPP definitely provided for the customer in this case.

[Godhumor]

29. A lot, to be honest, but start with the participants weren't undecided voters

Polls discussing debates only almost always have an initial control of undecided voters to prevent candidate bias. Based on the results given I am comfortable saying they purposely picked a sample that were primarily Clinton backers. Considering the sample was pre-screened, another huge no-no to preserve randomness, I put the likelihood they loaded the sample at basically 100%.

There are other issues with the questions and responses themselves, but the big one is how the sample was put together.

[joshcryer]

32. Then we need the pre-screen questions.

I wouldn't understand if there was a problem for simply screening for "people planning to watch the debate." But if they screened for supporters beforehand the whole thing is really fucked up and I thought PPP was above that, even with paid polls.

(They did start out as a Democratic pollster if you recall.)

[JonLeibowitz]

37. They also didn't rotate candidate names

That means a candidate, say, with their name starting with the letter 'C' has an inherent advantage over anyone starting with 'O' or 'S'.

[riversedge]

49. question number

one said that they all watched the debate tonight. Are you talking about this poll?

[MrMickeysMom]

31. Well, it' served one purpose being posted here...

We can see who the best cheerleading squad is... Rah! Rah! Rah!

[Godhumor]

40. The poll was designed to trick people

Look, I do this kind of stuff for a living. I'm devastated from a professional point of view that this was put out. And I feel for any supporter who tries to use it to make their point. It sucks.

[MrMickeysMom]

51. Good for you...

Mighty powerful people working for her. I share your feeling of icky-ness.

[Bluenorthwest]

66. I don't do this for a living but I do at times survive professionally based on stats and gathered

data. So from my point of view a poll designed to 'trick people' is morally lacking, it is in fact fraud, an attempt to deceive.
The persons who carried the polling out are unethical, but those who paid for it and used it even more so.
In business, I'd never, ever even take a call from anyone associated with that sort of deceit.

Explain to me why I should not blame the candidate who benefits from this at the expense of others.

[Godhumor]

69. In all honesty, I can't

I could talk about the firewall between candidates and PACs, etc, but there is no wisdom a Hillary ally bought the poll. It doesn't feel right to me, either.

[BlueStateLib]

41. PPP interviewed 510 Democratic primary voters nationally

PPP interviewed 510 Democratic primary voters nationally by telephone after the debate who had been pre-screened on Thursday and Friday as planning to watch the debate and willing to give their opinions about it afterward. The survey’s margin of error is +/-4.3%. This research was conducted on behalf of Correct The Record.
http://www.publicpolicypolling.com/pdf/2015/DemocraticPostDebatePoll111415.pdf

[Godhumor]

42. Read through the thread for my other responses. It was a cooked poll

The exchange with joshcryer in particular.

[hay rick]

44. One source of bias: "Democratic primary voters."

Limiting the poll to "Democratic primary voters" excludes the youngest voters and all independent voters- two sources of Sanders support. In my part of Florida, Democrats who voted in the 2010 primary (the last non-presidential year) represented 27.7% of Democratic voters in the 2012 general election and just 8.9% of all voters in the 2012 general election. We didn't have a presidential primary in 2012 so that comparison isn't available to me. If PPP used a more stringent super voter criterion than just voting in the previous primary, the sample becomes even more unrepresentative.

[SleeplessinSoCal]

45. trying to be neutral here. Bernie has yet to make a positive pitch

It shouldn't be so hard. Move to the next sentence after how much we're getting screwed to how much the economy will benefit from a more equal playing field. From the system is rigged to how to essential it is to a stable future in a changing world.

Problem is, it all winds up being expensive. And he made a dreadful mistake by admitting changes won't happen soon. That's where we are now. Moving slowly to fairness IF we can Congress to act.

[Godhumor]

46. ? Was this in right thread?

In not debating who won the debate. I'm just trying to say that you should disregard the after debate poll.

[SleeplessinSoCal]

57. having been blocked by both Clinton and Sanders groups

I felt an urgency to express why I think the poll #'s were closer to being accurate than not.

[stillwaiting]

60. Sanders is being completely honest with you (and all of us).

Electing him President will NOT change things overnight. He can't "wave his magic wand" as some of his detractors like to say and change things with how stacked the Congress is with corporate sell-outs.

The BIG progressive changes that need to happen can't happen simply with his election.

BUT, having him in the White House is a HUGE first step forward. And, we really do need to take a first step one of these days.

Bernie will use the bully pulpit and the power of the Executive to TRY to get his agenda moving forward. He can do this by his cabinet appointments (which will be better than HRC's) and by putting someone in at the DNC that actually works to develop a PROGRESSIVE Democratic Party. His election could be the start to changing the Democratic Party back to the party of FDR and away from the "New Democratic" Party that we are currently saddled with.

Please note that I said "could" start to change the Democratic Party. It absolutely will not change if we, the people, do not stay engaged and continue with a political revolution that demands elected Democrats represent OUR interests over Wall Street's, the multinational corporations, and the financial elites interests. If we can't do that then the Democratic Party will continue to exist as the lesser evil to the Republican Party, voter turnout will continue to be abysmal as so many Americans believe that no political Party will fight for their economic well-being, and Republicans will therefore have the edge NO MATTER HOW CRAZY THEY ARE since they have a base that has been radicalized to hate many of us and they will turnout to vote against us. That is a scenario that we are currently experiencing, and it is a scenario that will continue as long as the Democratic Party continues to be co-opted by Wall Street interests.

Edited to add: AND, you best believe that Bernie will be a strong DEFENSE to any legislation that inflicts further hurt or pain on the poor, working class, or middle class. That is another huge benefit to having him in the White House over Clinton. And, it's one worth fighting for.

[KoKo]

73. .....!

[SleeplessinSoCal]

74. Clinton referenced the revolution that didn't come.

I hope to live long enough to be a part of it.

[Bread and Circus]

52. well I will give credit where credit is due and you deseve credit for this op

[840high]

56. Agree.

[JackInGreen]

54. Thanks Gh

Everyone jumping the gun so willing to call the set

[billhicks76]

55. Clinton Has To Purchase Twitter Followers

Keep defending her with all that myopic loyalty.

[bigdarryl]

61. What utter B/S you might not believe this but

Hillary has a lot of support you wouldn't know that reading DU though

[reformist2]

62. The only times I hear about her support is when they release a poll...

... otherwise I see no evidence of support. I've seen several Bernie yard signs, bumper stickers, and t-shirts already. Not one for Hillary.

[sufrommich]

65. If we went by yard signs and bumper stickers,we

would be coming off Ron Paul's second presidential term. Most people don't wear their political choices on their sleeve.

[billhicks76]

63. Agreed

I know this. But she still purchases Twitter followers. She got caught.

[DemocratSinceBirth]

58. If the findings are specious we will find out soon enough

The scientific method is a way to ask and answer scientific questions by making observations and doing experiments.
The steps of the scientific method are to:

Ask a Question
Do Background Research
Construct a Hypothesis
Test Your Hypothesis by Doing an Experiment
Analyze Your Data and Draw a Conclusion
Communicate Your Results

There will be more scientific polls released in the coming days. They will either verify or contradict PPP's findings. I would literally wager on the latter.

[Scuba]

59. Many (most?) polls are designed to influence opinion, not measure it.

[sufrommich]

64. Thanks for pointing this out,Godhumor.

I'd hate see see HIllary supporters promoting unscientific polls.

[ismnotwasm]

67. Exactly

That would be bad.

I will wait for a decent analysis. I have my personal opinion, but that's the same as a unscientific poll..

[riversedge]

70. no, not in light of this poll....

both trending same direction--UP for HIllary!


http://www.democraticunderground.com/1251812152#post24

[Godhumor]

71. A second similar result doesn't excuse the sins of the first

And this one is very problematic.

[DemocratSinceBirth]

76. Flash polls are inherently problematic

But the CBS poll validates the scientific method... Two scientists working independently of one another using the same methodology should get similar results.

[Hepburn]

72. Thank you for an honest professional opinion.

Much appreciated.

[DemocratSinceBirth]

75. Yet its results are confirmed by a CBS poll thus validating the scientific method

http://www.democraticunderground.com/1251812304

[Godhumor]

77. Similar results does not mean same rigor

And the PPP poll had very serious issues.

I haven't looked into the CBS poll, and doubt I will, but, even if every single poll from here on out show Hillary winning, this poll should be ignored due to issues highlighted elsewhere in the thread.

[beam me up scottie]

79. Kick!

For those who haven't seen your thread.

[UglyGreed]

81. I tip my hat

to you