Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

Serious Question: Are polls crap or not?

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU
 
renie408 Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:36 PM
Original message
Serious Question: Are polls crap or not?
I hear both sides of this all the time. I imagine that if you are a Hillary supporter, you are saying, "HELL NO polls aren't crap!!" and if you support someone else, you are saying, "HELL YES polls are crap!"


But here's the thing...how often do election track just like the polls? When do you trust polls and when don't you? And if the polls indicate one candidate leading...what does that really mean?
Printer Friendly | Permalink |  | Top
zonmoy Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:38 PM
Response to Original message
1. depends on if the polls are in any way tainted.
can the polls be possibly rigged should be the main question.
Printer Friendly | Permalink |  | Top
 
renie408 Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:39 PM
Response to Reply #1
3. Oh man...that is a whole 'nother layer I hadn't even thought about...n/t
Printer Friendly | Permalink |  | Top
 
Capn Sunshine Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 03:30 PM
Response to Reply #1
22. There's a certain degree of stacking that goes on
Edited on Fri Sep-14-07 03:31 PM by Capn Sunshine
Private pollsters for hire will start with a premise and build a poll that produces that result.
Just by example, if you want the insider candidate to poll well, ask insiders and exclude the rest from the sample or include set percentages of non insiders to give the appearance of fairness.

I see it all the time, particularly the polls quoted at sites like DU.
Printer Friendly | Permalink |  | Top
 
sailor65 Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:39 PM
Response to Original message
2. Whether they go our way or not,
I've bever believed in them. I don't think a sample of roughly 1000 people is sufficient to represent the voting public. And it also makes the poll easier to skew depending on your particular goal.
Printer Friendly | Permalink |  | Top
 
LeftCoast Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:39 PM
Response to Original message
4. Depends on the methodology and margin of error
Of course, even the best poll can be misinterpreted.
Printer Friendly | Permalink |  | Top
 
wyldwolf Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:40 PM
Response to Original message
5. only if your candidate is losing
Printer Friendly | Permalink |  | Top
 
Clintonista2 Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 03:09 PM
Response to Reply #5
17. I've noticed exactly the same thing
Edited on Fri Sep-14-07 03:10 PM by Lirwin2
I'm reminded of a certain poster blasting the accuracy of all polls showing Hillary in a favourable light, and then using a Fox News poll yesterday to prove how well Obama is doing.


http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=132&topic_id=3519082&mesg_id=3519082
Printer Friendly | Permalink |  | Top
 
napi21 Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:40 PM
Response to Original message
6. I don't think you can believe any single poll data, but if you look at
all of them collectively, they are a good indicator of the ups & downs of whoever or whatever they're tracking.
Printer Friendly | Permalink |  | Top
 
EVDebs Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:41 PM
Response to Original message
7. Not if done scientifically: GIGO
Garbage in, garbage out.
Printer Friendly | Permalink |  | Top
 
Debi Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:44 PM
Response to Original message
8. Depends on the size of the sample - the honesty of the person being polled
The wording of the questions - ...:crazy:

If my family were polled right now Joe Biden would be winning (my son has endorsed him while Mr. Debi and I are still undecided) - is that an accurate assessment of the Iowa Caucuses? No - are we a good cross-section of Iowa Democratic voters? Yes - One over 60 male - regular caucus attendee - college graduate/one under 40 female - regular caucus attendee - non-college graduate/one eighteen year old male - college student - first time caucus attendee. (one-third of those planning on attending the caucuses have never participated in past caucuses).


Printer Friendly | Permalink |  | Top
 
MonkeyFunk Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:44 PM
Response to Original message
9. almost all polls are right
within their margin of error, when done properly.

The problem is they measure things that can't otherwise be measured, so there's no way to check 'em. For instance, a presidential poll today doesn't purport to judge how people will vote 15 months from now. It only measures how people feel TODAY, and the only way to check that is with another poll.
Printer Friendly | Permalink |  | Top
 
evlbstrd Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:47 PM
Response to Original message
10. Shouldn't this be a poll?
Printer Friendly | Permalink |  | Top
 
Debi Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:54 PM
Response to Reply #10
13. Oh, you are an evlbstrd!!
:rofl:
Printer Friendly | Permalink |  | Top
 
redqueen Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:52 PM
Response to Original message
11. So many factors...
sample size... wording... margin of error... methodology (phone polls? how many people answer their phone these days if they don't already know the caller?)
Printer Friendly | Permalink |  | Top
 
theoldman Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 01:53 PM
Response to Original message
12. Polls may not be very accurate but they tend to show a
direction on peoples thinking. As an example if HRC is 2% ahead in the polls she may lose because the poll is not that accurate. If the poll shows that she is ahead by 20% she may win by 30% or only 20%.
Printer Friendly | Permalink |  | Top
 
Dem_Wit Donating Member (20 posts) Send PM | Profile | Ignore Fri Sep-14-07 02:52 PM
Response to Original message
14. Polls reflect the opinion of the pollster
If you've ever been to Journalism 101, you will quickly learn that a poll can be swayed any way the pollster wishes to sway it.

Example:

Go to a poor section of town. Go door to door. Whoever answers the door, ask what they think of our current President, favorable or unfavorable. Those are the only two choices. You will get almost a total agreement of unfavorable.

Now then, go to a rich section of town. Ask whoever answers the door the same question using the same wording. You will now get almost a total agreement of favorable.

Which poll is accurate?

Another way to get opposing results is to run a poll on CNN and Fox News similtaniously. Same wording, different audience. Which poll is accurate?

Air America and Rush Limbaugh. Same question, different audience. Which poll is accurate?

According to NORML, 70% of Americans want marijuana legalized. Wow, 70%! That was a poll conducted in about 1978 that said "Do you think marijuana should be legalized for medicinal purposes?" 70% said "yes." Did NORLM lie? No, they didn't tell the whole story. They took words out of context. Shame on them. Sometime later, someone else took another poll. "Do you think marijuana should be legalized and made available for any person of legal age to buy as easily as tobacco and alcohol?" Guess what? The result plummeted to about 37%.

There must be 100 ways to sway a poll. Polls are to be taken with a grain of salt.

Printer Friendly | Permalink |  | Top
 
DemocratSinceBirth Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 03:20 PM
Response to Reply #14
19. Those Are Biases In The Poll Or By The Pollster
If you have controls on your poll and you know what you're doing you can measure what you want to measure...
Printer Friendly | Permalink |  | Top
 
depakid Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 02:58 PM
Response to Original message
15. It depends on the poll!
Edited on Fri Sep-14-07 02:59 PM by depakid
Who's commissioned it and what the methodology is. What are they trying to "demonstrate?"

Most cheap media polls aren't worth the time to look at. At best, they may show trends over times- provided that they're reliable (they ask the exact same question to the exact same "type of group). Even then, they're usually not generalizable to the population at large- or even among the actual subgroup that they supposedly represent.

Bottom line is that cheap media polls are meant to influence public opinion (like a herd effect)- they're NOT meant to accurately reflect what might actually be "out there."

And of course- it works. Just watch how many DU'ers jump to the tune every time some ridiculous result comes out.



Printer Friendly | Permalink |  | Top
 
Golden Raisin Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 03:09 PM
Response to Original message
16. Frankly, what difference do polls
make when elections are stolen/manipulated by Repubs and the Supreme Court annoints presidents?
Printer Friendly | Permalink |  | Top
 
Hersheygirl Donating Member (353 posts) Send PM | Profile | Ignore Fri Sep-14-07 03:10 PM
Response to Original message
18. Right now, yes.
Printer Friendly | Permalink |  | Top
 
DemocratSinceBirth Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 03:24 PM
Response to Original message
20. Here's A Link To Dozens Of Final Polls From The 04 Presidential Election
http://www.ncpp.org/files/2004%20Election%20Analysis.pdf


If the pollsters were just making shit up they did a great job...
Printer Friendly | Permalink |  | Top
 
Adelante Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 03:28 PM
Response to Original message
21. Tracking the horse race, that's it
It's a slice of time and a picture of where the public is at that time only. When voters start voting is when we know whether predictions prove out. I don't take them entirely seriously, but I do have fun watching them.
Printer Friendly | Permalink |  | Top
 
PATRICK Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 03:59 PM
Response to Original message
23. Most polls
tell you what you can largely deduce without them if you have some contact with people AND the real world and some little wisdom based on experience. When you give your own bias free rein out of hope or determination you have the opportunity to seeing it at work and whether to measure it as denial or creeping pessimistic despair. In most cases Thomas the Doubter would be right.

The real world of the pollster starts with the phone caller and the pressure to get answers. Then the science of statistics based again on experience tempered with critical wisdom, sometimes daring to call itself objective blankets over all those thousands of mushy encounters. Then the goal gets moved along with that cover of plus or minus 4 per cent. Do they ever admit to screwing up the goal post badly? then they hide in AVERAGING all the different polls.

But in particular what should be disturbing considering some of the dubious characters(WSJ) jumping on the band wagon, is their particular interest in what is usually an internal polling goal by candidates. They are trying to measure and interpret Hillary's "likeability" and for once, after seeing many many frontrunners go up in a puff of ballots, try to bolster the momentum of a perceived poll leader by helping it erase the doubts. That, without any other factor being obvious why they should be changing other than "success" of the moment.

The numbers themselves might be significant but they are too small for me to notice anywhere in the real world. People are more depressed than ever that the candidacy is slumping toward inevitability. these are democrats. Maybe I am missing the wildly enthusiastic women, but there may be a rallying factor based on that huge segment of the population tired of seeing one of there own bullied and then successful in spite of it. That is not much of a change in seeing any depth in the voter or the candidate or of polling at this stage. The numbers are not comforting for whatever reason, all the more because such movement can vanish in a twinkling, under real pressure, but the interpretations by the pollsters seem eager to please.

I would love to see the leading candidate, the eventual winner perhaps, take the public opinion by the short hairs and deal actively with things set in stone. This is more like surf boarding on mush. Where are any people being actively won over except by the aura of coronation and the need for hope in any form allowed? There are passive strategies that are workable, but always they are dangerous for the choice of leadership. When Clinton, in allergy season, THIRD in polling and definitely THIRD in the media circus, took time out to nurse voice and finances and get some sax playing in. He could chill because the other two were destroying each other. Kerry's respite continued under the barrage of MSM supported swiftboating, now a household term, whereas our justly elected president just became another voice in the Senate, one not granted full leadership even by the centrists who helped ruin another successful candidacy. What worries me, after all our debates and controversies of the past is how set in stone the likely scenarios are going to be, and the biggest millstone that gives a stable base to all theories is the Clinton campaign steaming along on entitled cruise control.

Kerry convinced with his personal campaigning on the stump after being written out of the circus. He convinced with organizational and winning acumen that spoke of the hero of Vietnam and the BCCI takedown. Now it's coasting with Clinton with a red carpet poll being laid at her feet like the submission to George Bush after 9/11, a popularity and strength that was not rational and could not last by the predictability of those in power.

If Hillary smashes through the primaries, people must try to strengthen the public perception. If she gets a bruising, doubts will linger and likely again be coasted over. For everyone's sake she must risk victory the hard way. If the real leadership presentation campaign, dealing with fixed attitudes(not all loony), does not start until she takes the podium at the Convention, then she has lost an opportunity that adversity might have forced her to. Whatever the polls suggest to the hopeful campaign
staff itself the real trouble is still out there. Doubling the misgivings is how this reflects on her very likely presidential performance when she gets elected and the unnecessary costly sacrifices caution usually exacts.
Printer Friendly | Permalink |  | Top
 
Forkboy Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 04:50 PM
Response to Original message
24. The real problem is how they are used by various groups.
Misleading, manipulative and a subtle form of peer pressure...that works.
Printer Friendly | Permalink |  | Top
 
LWolf Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-14-07 07:17 PM
Response to Original message
25. Yes. Polls are crap.
Maybe not the kind of crap you mean, though.

Polls are crap because they are used to manipulate public opinion. How many people will stick to the top "3" only because of polls, rather than issues or record? That's corruption of the democratic process, imo.

Poll questions can be, and often are, slanted to get responses that often don't really reflect a person's true response.

Name recognition affects polls. People are much more likely to narrow their choices to names they hear about, inflating the poll numbers for those getting the most air time with the msm.

Of course, campaign funds are a factor in getting the word out, and building name recognition. Those with the most money get the most time, and therefor e the highest polling numbers, which leads to more air time, in a vicious closed circle.

The msm sorts candidates and awards more air time and more positive press to some long before all have had an opportunity to build that name recognition, basically slamming the door on their exposure.

The whole process is corrupt, and the polls are part of that corruption.
Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Fri Apr 26th 2024, 04:46 AM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC