Flaws in Polling Data Exposed as U.S. Campaign Season Heats Up

[Jesus Malverde]

Polls for the 2016 U.S. presidential race have been defying all expectations: Donald Trump as the persistent Republican frontrunner even as he insults large swaths of the country and brushes off policy questions; Hillary Clinton haunted by an email controversy Democrats shrug off while a Vermont socialist keeps gaining on her.

Are the polls correct? While that is hardly a new question, doubts are intensifying after a series of high-profile misfires around the world in the past year, notably in Greece, Israel and the UK. As politics and business lean increasingly on surveys and data, technological and social shifts are combining to challenge polls’ reliability in an entirely new way. Polling professionals have no solution; investors are wary.

"There isn’t a pollster out there who thinks about this seriously who isn’t a little bit uneasy," said Kirby Goidel, editor of the book "Political Polling in the Digital Age." Interviews with more than a dozen pollsters in the U.S. and around the world revealed similar anxiety.

Brad Schruder, a director of foreign exchange at Bank of Montreal, said what many in the investment world have been thinking: "It makes you wonder, how much weight should we attach to these polls?"

Telephone Avoidance

Global Failures

A single voter

Robocalls

Rising Anger

http://www.bloomberg.com/politics/articles/2015-09-29/flaws-in-polling-data-exposed-as-u-s-campaign-season-heats-up

[still_one]

1. This is actually very interesting. The shift from land lines to mobiles definitely changes the

demographics, but I also think caller ID has an impact to, especially because people are tired of robocalls, so if they see a caller ID they do not recognize, they won't answer it.

Something to think about for sure.

Thanks

[Scuba]

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

[Jesus Malverde]

3. Exactly..

People follow a crowd when it comes to down to it. If someone tells you the crowd thinks this, most people want to be part of it.

Sanders momentum is proof of that, seeing his massive crowds is contagious. In spite of the media push polls promoting others. Push polls are only effective with the boob tube Generation. Most people get their news elsewhere these days.

[Scuba]

4. That's part of it. The clever wording of questions is another, as is the selection of participants.

[Jesus Malverde]

5. Who these days answers their home phone

Old people who skew conservative. Haven't answered my home phone in probably five years.

Friends know my mobile, even then won't answer if I don't know the caller ID. The days of answering the phone are last century

[postulater]

6. We are considering dumping our landline.

Just curious, why do you keep yours?

I can't think of a good reason why we keep ours.

[Jesus Malverde]

7. dsl

For Internet access in the states.

[newfie11]

8. This old person hasn't had a landline for over 10 years

And will never go back.

[LiberalElite]

10. Old habits are hard to break - one of my friends is 80 y.o.

and uses a walker. It's not easy for her to get around. Nevertheless she insisted on answering the phone despite having an answering machine. She doesn't have caller I.D. When she recently complained about answering the phone only to find it was a telemarketer or a fraudster from Microsoft, sometimes several times a day, I had to tell her to stop answering the phone. It is not necessary anymore. It was a hard sell but she finally agreed.
P.S. she's a Democrat.

I also had a friend in her '70s who had a Tracfone and a landline and intended to keep the landline in case of emergency. She also was a registered Democrat.

[Gormy Cuss]

11. Independent polls using accepted, scientific practices are not designed to influence opinions.

In fact, just the opposite is true. Admittedly there are lots of junk polls out there (DU polls, for one) and most people don't understand the difference.

It's also true that legitimate, rigorous polls can miss the mark. That's why AAPOR* members spend so much time testing methodologies and trying to understand the influence of technology on the representativeness of the responses.




*http://www.aapor.org/AAPORKentico/default.aspx

[Scuba]

12. Most polls seem to start with "If you knew so-and-so ate puppies for breakfast ... "

[Adrahil]

13. No, they don't.

There are most certainly push polls out there, but it is not "most." Check out a site like 538, which actually assesses poll methodology and poll performance.

[Scuba]

14. You should remember that most polls have nothing to do with politics.

[Gormy Cuss]

16. That's the perception here and other places where internet "polls " are done.

but again, scientific polls have questions that are carefully crafted to mitigate bias and have sampling methodology designed to poll a broad spectrum of demographics, again to mitigate bias.

[Scuba]

17. Most polls are not scientific nor carefully crafted to mitigate bias.

Some polls, sure. Most, no.

[Gormy Cuss]

19. Sure,because "most poll" include internet polls.

But if we're talking about national political polls, it's not a matter of my opinion. There are those who abide by the AAPOR standards and those that don't. One big tell is whether the pollster's methodology is open and reported so that other researchers can critique it and try to replicate the results, just as is standard in hard science data collection.

I realize that to those outside of any profession it's hard to ferret out the good from the bad and that unfortunately the internet has made us feel like armchair experts because there's so much data readily available. The problem is that GIGO is more true now than ever.

[canoeist52]

9. The most accurate poll to date - the 25,000+ people who showed up yesterday in Mass. fro Bernie!

And in a cold wet and windy weather. We vote - we win!

[Sisterwife15]

15. We don't like the results so

They must be wrong!!

Seems to be how it goes around here.

This site ranks them. Some polls are better than others I suppose.

http://fivethirtyeight.com/interactives/pollster-ratings/

[Ace Rothstein]

18. It kind of reminds me of unskewedpoll.com from 2012.

[Sisterwife15]

20. That's a blast from the past!

I tried going to that site and it's now German as-seen-on-TV products