Not many polls in December for first states in Democratic primary

[Uncle Joe]

https://www.msnbc.com/hardball/watch/not-many-polls-in-december-for-first-states-in-democratic-primary-75772997897

Video is on the link

The moderator blames the lack of polls 14 in 2016 versus only 4 in 2019 on money, I believe that's bullshit, it shouldn't be that much more difficult to reach voters in 2019 than 2015.

[crazytown]

1. No idea.

What do you think?

[Uncle Joe]

2. They knew Bernie would be surging in December but didn't want to promote it. n/t

[beastie boy]

7. Not everything that happens is an anti-Bernie conspiracy, Uncle Joe. maybe they just got tired of

showing Biden consistently in the lead. Try to be more objective, like I am with my reply.

[Uncle Joe]

9. Biden is sliding in the polls and has been for months,

that's why Bloomberg is running.

[beastie boy]

10. He will need to do a hell of a lot more sliding before he can challenge anyone for the #2 position!

Except Iowa. But Biden is actually rising in Iowa. And he is pretty steady nationwide. Has been for months.

[Drunken Irishman]

16. This just isn't true.

For someone who seems to be pushing the idea for transparency and accuracy, it's remarkable to me that you have no problem attaching yourself to a lie.

Nationally, Biden is currently at 28.1% according to RCP - Bernie, who's second, is at 18.8.

On Dec. 1st, Biden was polling at 27.7.

On November 1st, Biden was polling at 27.6

On October 1st, Biden was polling at 26.2

Since October 1st, Biden has increased his lead by over a percentage point nationally.

In Iowa, Biden was polling third at 15.7 on Nov. 1st.

On Dec. 1st, Biden was polling at 4th at 16.3.

Today, Biden is third at 18.8% - which is essentially a statistical tie with Sanders for second (Sanders is polling at 20%). So, I struggle to see how he's been sliding in the polls the last few months.

Biden is better positioned today than he was on November 1st and December 1st - both nationally and in Iowa.

[DarthDem]

26. Thank You

I simply do not understand the need to push a false narrative. It's odd.

[sheshe2]

30. Please stop.

Uncle Joe (48,601 posts)
2. They knew Bernie would be surging in December but didn't want to promote it. n/t

They somehow knew Bernie would miraculously start surging in December (which he is not) so they decided to stop all polling just to make him look bad or perhaps have a media black out of BS and every other candidate in the race?! Dear Goddess that is just plain silly.

[Uncle Joe]

33. That's the difference between what one considers "miraculous" and

another highly logical.

[sheshe2]

34. Non answer. nt

[TexasTowelie]

35. .

[ritapria]

3. I was puzzled

when CBS News /YouGuv didn't publish Early State Poll results in December when they had done so for the previous 5 months ….I'm not buying the excuse that they ran out of funds to conduct them ….What happened CBS ?

[judeling]

4. Not money so much as fear of being wrong

I think most organizations are gun shy after 2016.
Not so much because they were very wrong, but because it was reported so breathlessly.
Also there hasn't been a real firming of support for any candidate as of yet. Bernie came in with that firm support. But there has been little consolidation.

[Uncle Joe]

6. I believe December polls would prove otherwise.

Whether they got it wrong in 2016 or not is irrelevant, that hasn't stopped them from issuing previous polls since this spring.

[judeling]

11. National Polls sure but early state level polls are greatly dimiished.

And it is those that are really important. In general most people are aware and interested in the nomination just not really engaged. That is normal as they still have months until they have to make a choice.

The lack of early state polls have really had an effect on the Race. Booker may have made the last stage if the pace of polling was anything like it was four years ago.

[Uncle Joe]

12. That changes in December; only six weeks away from the vote. In 2016 they had 14 polls

and now they're blaming it on money?

[Drunken Irishman]

15. See Reply 14...

Polling has decreased each of the last four election cycles.

[StevieM]

27. The polls weren't really wrong in 2016, the polls changed.

Comey turned that election upside down in the final weeks.

Then again, Comey turned that election upside down repeatedly. 2016 was his election, from beginning to end.

[msongs]

5. all part of the massive bernie blackout lol. nt

[Uncle Joe]

8. Yes. n/t

[justhanginon]

13. Oh my God, they are missing yet another Bernie surge!

Patience, i'm sure there will be another one along shortly.

[onetexan]

31. OMG i blinked & missed it

[Drunken Irishman]

14. It's not bullshit - & there's a clear trend. Polling has significantly decreased each election cycle

Most polls are conducted by, at least in conjunction with, a good amount of print media.

NBC does theirs with the Wall Street Journal.
ABC does theirs with the Washington Post.
CBS does theirs with the New York Times.

FOX and CNN do theirs without a print partner.

Print journalism is dying.

Moreover, polling IS becoming increasingly difficult to accurately, and properly, pull off. The lack of landlines, the overall inability of people to actually answer their cell phones, has made it a much bigger challenge for organizations to successfully poll, and when you couple that with cash-strapped mediums, you get fewer polls.

As I mentioned in the headline of my post, this isn't a new thing. It's been the trend now the last few election cycles.

Just looking at Iowa, the first state to cast their vote, and you'll see the trend over the last few election cycles.

In November and December of this year, there have only been a total of six caucus polls. Over the same stretch throughout the past few election cycles, it's been:

2016 (Dem): 12
2012 (GOP): 28
2008 (Dem): 43

That is a clear, dramatic shift in polling and it's not just from 2016 to 2020. In November & December, 2007, 43 polls were conducted in Iowa. Four years later, that was down fifteen and four years after that, it was down sixteen and now, four years later, it's down half that.

And it isn't just Iowa, either. 2016 is actually a good example of how the lack of polling missed Trump's late surge in Wisconsin, Michigan and Pennsylvania.

Let's take Pennsylvania and compare the amount of polls the final week of the campaign compared to 2012, 2008 and 2004:

2016: 6
2012: 8
2008: 14
2004: 12

Polling has dropped significantly across the board and will only continue to do so as it becomes increasingly difficult to weigh sample size to accurate results. The lack of final week polls in 2016, and even 2012, missed the shift in support both the winning candidate saw in the final days of the campaign (despite the narrative - in 2016, it was that Hillary would win and in 2012, it was that things were just too close to call, with maybe a slight lean toward Obama). What we know is that Trump won and Obama won by a far more comfortable margin than many anticipated. Had there been as many polls run in 2016 in Pennsylvania as there were in 2008, we probably would have seen Trump's rise before it was too late.

Alas...

With that said, campaign polling was correct. Hillary's campaign picked up on Trump's surge in these states and it's why they made a last-minute push in each one of 'em - including their massive rally in Philadelphia with the Obamas. A lot of people questioned why she was putting time in Pennsylvania despite leading in the polls, as she held a huge rally with Katy Perry the Sunday before the election, and then again, as I mentioned above, with the Obamas the day before the election, well that's why.

Polling is going to become increasingly rare due to the costs/ineffectiveness of it.

[Uncle Joe]

17. I guess that's one advantage individual donors and donations have over polls,

they're self-funding.

14 polls four years ago in the one month of December to 4 today when when we have a 1600+% increase in candidates is just too drastic of a drop to be blamed solely on money in my book.

[Drunken Irishman]

18. Except the decrease aligns perfectly with the trend.

Not everything is a conspiracy. Leave that to the Qers.

Polling across the board has decreased each election cycle. It's not a new phenomenon. We saw it four years ago - and four years prior to that.

[Uncle Joe]

19. When are polls most valuable or accurate

when compared to the vote?

[Drunken Irishman]

20. If it's a conspiracy against your candidate, why have they decreased each cycle since 2008?

Maybe you should look into it more instead of throwing around conspiracy theories. The reality is that polling is far harder today than it was in 2008. It doesn't take someone with an advanced degree in statistics to understand this.

In 2008, 80% of U.S. households had a landline, while only 20% used cellphones only.

Today? Only 41.7% of U.S. households have a landline - 54.9% of households are 'cellphone only':



That is a dramatic, massive decrease from 2008 - an election that was only a little over a decade ago.

Fact: polling cellphones is much harder today than it is to poll landlines.

Fact: Americans willing to participate in polls is at an all-time low.

Pew’s data suggest that decline is underway again, with response rates slipping to 7 percent in 2017 and 6 percent in 2018. What’s to blame for the recent slippage? Kennedy says it’s harder to get people to complete polls over cell phones because they are getting more calls they don’t want, which makes them less likely to talk to pollsters.

“It’s our sense that that exponential increase in robocalls, spoofing of incoming calls, pretending they’re a local number, has really changed the environment in using a cell phone,” said Kennedy.

What does this change mean? It means pollsters have to weigh their sample differently to be as accurate as possible, which also means more effort is placed into mining the results - and yes, more money:

“(W)ith the rise of cellphones and the increasing prevalence of people who decline to answer or complete surveys, how do you ensure that those you are interviewing are in fact representative of the underlying population?” Cook said. “This is not an intractable problem, but it places a premium on rigorous methodology, which can be complicated and costly.”

This article was from January, 2018 - long before Bernie announced he was running for president. And they came to the same consensus as the article you posted: it takes more money to poll today than it did a decade-plus ago because of the changing demographics. That is why there's fewer polls today than four years ago and eight years ago twelve years ago.

[Uncle Joe]

21. You never answered my question.

"When are polls most valuable or accurate when compared to the vote?"

[Drunken Irishman]

22. You never answered mine.

Asking a question to deflect from the point at hand is a losing tactic. The issue is you claimed they're lying about the fewer polls by citing cost. Your initial point in your original post said:

The moderator blames the lack of polls 14 in 2016 versus only 4 in 2019 on money, I believe that's bullshit, it shouldn't be that much more difficult to reach voters in 2019 than 2015.

I just explained to you why it was harder to reach voters in 2019 than it was in 2015 - just as it was harder to reach voters in 2015 compared to 2012.

I assume you'll continue to gloss over all the evidence I've presented because you have no answer for it and it doesn't fit your narrative that the polls just decided, the last few weeks, to purposely hold back in reporting results, despite this being a trend now for each of the last four election cycles, because they want to undercut the supposed Bernie surge. That's a crackpot theory and I showed you why.

[Uncle Joe]

23. You never asked a question but I did and you still haven't answered it. n/t

[Drunken Irishman]

24. I did.

If it's a conspiracy against your candidate, why have they decreased each cycle since 2008?

You've failed at backing up your initial assertion that it shouldn't be harder to reach voters in 2019 than it was in 2016 and I outlined, pretty reasonably, I might add, why it is, in fact, more difficult to reach voters today than four years ago. You had no answer for my evidence, so, you do what many who can't concede defeat do and you deflected by asking an avoiding question. It's cute but you're the one who posted the initial claim and I refuted it. You started the dialogue and I responded. It's my fault that I expected actually having an intelligent conversation based around facts with someone on here - oops. Instead, you took the Donald Trump approach and made an easily disprovable claim and then, when confronted with evidence about why that claim doesn't hold up, you pivoted to something different altogether.

Whatever. Your initial post was wrong. I showed you why you were wrong. I feel I did my due diligence in proving how wrong you were. If you refuse to answer the information I provided, that's on you. It just indicates to me to never enter into another conversation again with you because you're clearly not here for actual debate and rather just lobbing bullshit accusations with the occasional gas-lighting. So, good luck.

But to answer your ridiculously vague question: polls are most valuable and accurate when the resources, and costs, are invested into 'em, which means, maybe fewer, but more accurate polls, are the wave of the future despite your whining about there being less polls conducted than four years ago. Since more resources go into mining polling data today than a decade ago, the cost + effort likely means fewer overall polls so not to undermine the data.

It's a big reason I am skittish about the accuracy of the weekly polls from the Economist/YouGov and Morning Consult, even if they show Biden doing well.

So there you go. Peace.

[Uncle Joe]

25. The point I was getting at and to which you couldn't/wouldn't acknowledge

is that the closer to actual vote day is when polls are most accurate and valuable, not three, four, five, six months or longer.

I believe the first four states Iowa, New Hampshire, Nevada and South Carolina being first to vote should have a preponderance of polls six weeks before they vote and just four polls in 2019 dropping from 14 polls in only four years just doesn't get it.

[StevieM]

29. I guess it is possible that a slew of new polls will come out on Monday.

Either way, I don't think it is a giant conspiracy.

[Uncle Joe]

32. Three days left in the month,

they have a lot of ground to make up.

[TexasTowelie]

36. Perhaps you would like to kick in some money to finance those polls?

Personally, I believe that polls conducted over the holiday season have a higher margin of error because the participant is more likely to be intoxicated.

[Uncle Joe]

37. or perhaps feeling the spirit. n/t

[judeling]

28. Thank You it is a huge problem.

Even in 2016 the late surge in swing states of Trump was masked and since the narrative of Hillary was going to win was set, it may have cost her the election.

Dewey defeats Truman is the most famous example of the lack of polling, but there are many and subtle.
Cory Booker may have very well made the last debate stage if the polling in early states was near the same. MOE should have given him at least a couple of qualifying polls.

The entire Rational of early states is that they are small enough that candidates without initial Name Id and or Money to gain one have a chance. Without polling they lack the media narrative to prove that. Without that proof it is essentially over.

Pete Buttigieg is only considered a contender right now because of Iowa polling.
Iowa is important not only because it is first but because as one of the regionals it gets polled the most because of the Register.
If Nevada or South Carolina were even polled half as much as Iowa the Narrative and the candidates would be different. Castro and Beto may have been able to join the conversation in the fall.