Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

kentuck

(111,052 posts)
Thu Sep 18, 2014, 03:40 PM Sep 2014

The inconsistency of polls.

Last week, I saw a poll in Colorado where Udall was comfortably ahead of Cory Gardner. Today, a poll shows Gardner ahead. The same with the Governor's race. Hickenlooper was ahead about a week ago and yesterday, he was 10 points down??? Which are we supposed to believe?

Obviously, both cannot be right. Last week, I saw a poll that showed Alison Grimes ahead of Mitch McConnell by one point. This week, I have seen polls with McConnell ahead by as much as 8 points?
I'm sure this is going on across the country?

What could be the purpose of such polls? We are told they are just a snapshot. Are the American people that fickle? That they can totally change from one week to another?

Are they really any indication of how people will vote on election day? I tend to think not.

8 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies
The inconsistency of polls. (Original Post) kentuck Sep 2014 OP
What could be the purpose of such polls? Glassunion Sep 2014 #1
First we need a poll... Whiskeytide Sep 2014 #4
At best, they H2O Man Sep 2014 #2
I tend to agree with the latter point. kentuck Sep 2014 #3
Important things to look at: Spider Jerusalem Sep 2014 #5
Very important! H2O Man Sep 2014 #6
I've discussed this point before and if you're not bored here it is rock Sep 2014 #7
The only poll that matters is on election day. bigwillq Sep 2014 #8

H2O Man

(73,506 posts)
2. At best, they
Thu Sep 18, 2014, 03:46 PM
Sep 2014

are a glimpse of opinions at a specific time; I've heard polls described as "snapshots."

Polls can also be used to manipulate public opinion.

 

Spider Jerusalem

(21,786 posts)
5. Important things to look at:
Thu Sep 18, 2014, 03:51 PM
Sep 2014

Methodology, and sample size. Polling is tricky anyway (online polls are very much self-selecting and tend to be wildly inaccurate; telephone polls skew older because they phone landlines, and a lot of people with caller ID just won't answer calls from "unknown number", or whatever) .

rock

(13,218 posts)
7. I've discussed this point before and if you're not bored here it is
Thu Sep 18, 2014, 04:57 PM
Sep 2014

The mathematics and methods are there for highly reliable polls to be developed. But politics invariably enters the picture.

A public poll is a poll whose results will be made public. These are typically politically motivated and requested and paid for by someone who desires a specific outcome (which the pollster is aware of). "Push" polls naturally are of this type.

A private poll is a poll whose results will remain private. These too of course may be politically motivated. But these are requested and paid for by someone who desires to know the truth, at least as far statistics can provide it. Notice that "push" polls are never of this type.

And now a personal rant: it is possible that the polling firm has such integrity that they eschew money for providing "doctored" results but this is highly unlikely because every large organization is overrun with MBA's.

Latest Discussions»General Discussion»The inconsistency of poll...