polls are stupid.

[351Cleveland]

Originally posted by: Zedtom
I just love the word "scientific" when polls are taken. The results are subject to manipulation and the objective goals can be rigged from the start. I have been polled before and have flip-flopped, (OOPS!), I mean- changed my mind.

Thus the reason they poll more than once...

 

[cquark]

Originally posted by: raildogg
Just because Bush is holding a big lead in many of the polls doesnt mean they are false. If all the polls give you the same result, it means they are fairly accurate

The issue is that the polls don't agree. There are two major groups of polls, one of which shows Bush way ahead, the other of which shows him barely ahead by less than the margin of error. There's likely some consistent difference in how they're sampling or reporting their data.

 

[Moonbeam]

Stools are pooped.

 

[jhu]

Originally posted by: 351Cleveland

Originally posted by: jhu
so there is are multiple polls out saying how gwb is in the lead. well, polls aren't always right. look at truman vs. what's-his-name. or even gore vs. bush back in the day.

Why does your post sound like my 3 week old when she is hungry?

perhaps because your three-week old is so advanced for his age that he can see through how mostly meaningless polls can be at this point.

 

[cquark]

Investor's Business Daily/Christian Science Monitor poll out today has exactly the same numbers as Zogby: Bush 46%, Kerry 43% among likely voters. Among registered voters Bush 44%, Kerry 43%.

 

[dullard]

Originally posted by: b0mbrman
There's a margin of error and things can always change at the last minute. However, all else equal, if you were to force me to bet my own money on a candidate, it would definitely be put on Bush.

Now, if you were to raise an argument such as "Pollsters don't call overseas servicemen who always send in absentee votes for XXX" or "Pollsters don't call cell phones and young people who vote for XXX use cell phones as their primary means of communication" or "Old people who vote for XXX are most likely to be at home when pollsters call," then this would be a worthwhile thread....

Otherwise, you're just arguing against the margin of error....and that's like arguing that the team that loses game 1 has a better shot at winning the best-of-7 series

I agree with you, good post.

Can I add more?
[*]The poorest of the poor still don't have phone lines, thus the polls exclude the poorest which is heavily democratic.
[*]The upper middle class is the most likely to have multiple phone lines (fax line, second line for teenager, second line for internet, second line for home business, etc). This group is much more likely to vote republican, so republicans are overemphasized.

 

[Perknose]

Originally posted by: raildogg
Just because Bush is holding a big lead in many of the polls doesnt mean they are false. If all the polls give you the same result, it means they are fairly accurate

No, it doesn't.

 

[Spencer278]

Originally posted by: dullard

Originally posted by: b0mbrman
There's a margin of error and things can always change at the last minute. However, all else equal, if you were to force me to bet my own money on a candidate, it would definitely be put on Bush.

Now, if you were to raise an argument such as "Pollsters don't call overseas servicemen who always send in absentee votes for XXX" or "Pollsters don't call cell phones and young people who vote for XXX use cell phones as their primary means of communication" or "Old people who vote for XXX are most likely to be at home when pollsters call," then this would be a worthwhile thread....

Otherwise, you're just arguing against the margin of error....and that's like arguing that the team that loses game 1 has a better shot at winning the best-of-7 series

I agree with you, good post.

Can I add more?
[*]The poorest of the poor still don't have phone lines, thus the polls exclude the poorest which is heavily democratic.
[*]The upper middle class is the most likely to have multiple phone lines (fax line, second line for teenager, second line for internet, second line for home business, etc). This group is much more likely to vote republican, so republicans are overemphasized.

Don't forget you can only vote once in a poll but you can vote early and often come election day. In all serinous the polls try to correct for the effects you listed. Oh and the margin of error that is cited in polls is based on the number of samples and total population. The margin of error does not account for any faults in the polling processes. For example if god decided to read 2000 peoples minds about who they voted for and he choose them at random then there is a 95% chance he would be correct +- 3%

 

[cquark]

Here's another set of poll results:
http://kirk.blogs.com/photos/uncategorized/polls.gif

 

[dullard]

Originally posted by: Spencer278
In all serinous the polls try to correct for the effects you listed.

My question to you is do they? Gallup says this:
"Today, approximately 95% of all households have a telephone and every survey reported in this book is based on interviews conducted by telephone...

In the case of Gallup polls which track the election and the major political, social and economic questions of the day, the target audience is generally referred to as "national adults." Strictly speaking the target audience is all adults, aged 18 and over, living in telephone households within the continental United States. In effect, it is the civilian, non-institutionalized population. College students living on campus, armed forces personnel living on military bases, prisoners, hospital patients and others living in group institutions are not represented in Gallup?s "sampling frame." Clearly these exclusions represent some diminishment in the coverage of the population, but because of the practical difficulties involved in attempting to reach the institutionalized population, it is a compromise Gallup usually needs to make...

In the case of the Gallup Poll, we start with a list of all household telephone numbers in the continental United States. This complicated process really starts with a computerized list of all telephone exchanges in America, along with estimates of the number of residential households those exchanges have attached to them. The computer, using a procedure called random digit dialing (RDD), actually creates phone numbers from those exchanges, then generates telephone samples from those. In essence, this procedure creates a list of all possible household phone numbers in America and then selects a subset of numbers from that list for Gallup to call."

Gallup clearly states that (a) they make no attempt to contact the 5% without phones, (b) they make no attempt to sample students, military personnel, etc, and (c) they don't claim to adjust the random phone numbers to make certain it is a representative sample. They do attempt to separate home from business numbers, but that is the extent that they claim they make adjustments.

So again, do they adjust for the 5% that don't have phones and do they adjust for the people with 2 home phone lines?