Are the polls horribly wrong?

I keep seeing the polls on the Crap News Network(s), and I can't help but wonder if they are wrong.

First, the cross section of the population that they poll is incredibly small to be a good representation of the whole, and they're probably asking the same types of people they normally do.

By that, I mean people with homes and landline telephones.

We know that there has been unprecedented voter registration turnout, and (from some polls) we know that there is a relatively small portion of people that are still undecided on their vote.

Are the polls missing the young people that have turned out to vote this year? It seems to me that there might be an 'unseen and unheard' volume of people that could tip the scale toward Kerry.

I guess we'll see on November 3...

Polls are notoriously wrong, but it's not because of sample size, it's technique. No planning is done around polls except those contracted by the parties themselves, to make their messaging more popular. Take them with a grain of salt. In fact, since 2000, take the main poll (the election itself) with a grain of salt.

Those polls drive me crazy. They show Bush ahead one minute then Kerry, then they show them tying

I voted today. GO KERRY!

Also one must take into consideration the times that they do the polling, they mostly reach elderly folks who do not work. I know personally, no pollsters have called me, or just about anybody I know, not that we would have been home to answer the phone. The younger generation that leans toward Kerry is usually out working (or trying to find work thanks to Bush).

Originally posted by 27jd
Also one must take into consideration the times that they do the polling, they mostly reach elderly folks who do not work. I know personally, no pollsters have called me, or just about anybody I know, not that we would have been home to answer the phone. The younger generation that leans toward Kerry is usually out working (or trying to find work thanks to Bush).

You're one out of 300 million. Hardly concise proof of polling habits

Originally posted by Esoterica
You're one out of 300 million. Hardly concise proof of polling habits

The population of the US is less than 300 million.

The number of legal residents is somewhat less.

The number of legal residents of voting age is still less.

The number of legal residents of voting age who register to vote is still less again.

The number of legal residents of voting age who register to vote who vote at election is still less yet again.

The last election turned corruptly on less than 1000 votes in a farcical result.

Work the numbers.

Every vote counts (when it is counted).

No polls count.

I think I heard that 22million people actually voted in the last general election. Checking that...

Actually, you may want to polls not to show you in the lead. Voters would think that you'd win anyway, and not bother voting.

Originally posted by curme
Actually, you may want to polls not to show you in the lead. Voters would think that you'd win anyway, and not bother voting.

It works the other way too. If 1 person looks like a lock to win, voters for the other side may not come out.

Originally posted by MaskedAvatar

Originally posted by Esoterica
You're one out of 300 million. Hardly concise proof of polling habits

The population of the US is less than 300 million.

The number of legal residents is somewhat less.

The number of legal residents of voting age is still less.

The number of legal residents of voting age who register to vote is still less again.

The number of legal residents of voting age who register to vote who vote at election is still less yet again.

The last election turned corruptly on less than 1000 votes in a farcical result.

Work the numbers.

Every vote counts (when it is counted).

No polls count.

I was being facetious. He's still one out of many millions of legal voters.

And don't get in to the "Bush stole the election" again. I swear, the regulars around here are as bad as the one-off paranoiacs

These general election survey type polls don't even really show you whats going on anyway. We don't have a national election like this. We basically have 50 individual state elections. So what you'd really want to look at is a statewide poll of battleground states, and then compare those polls. That's just my $.02 anyway.

.
I think it would be an interesting study to try to form a psychological profile on what kind of people actually respond to polls. I would think that it would skew in some way towards people with time and patience. I know I have been phoned once this year and politely declined to participate. I�m sure a lot of people do the same.

On the other hand, the professionals who do polls, not affiliated with political parties, are supposed to have some idea what they are doing. No doubt some attempt is made to get a representative cross-section of respondents.
.

I worked for a political polling firm for a short time (about 5 months) right before the mid-term elections in '02, and I don't know about now, but at that time, all of the numbers on the list were landline telephones.

I think that if the lists are still comprised of landlines only, then the poll would be automatically eliminating a significant group within the whole, mainly the young.

Most people that I know that are my age or younger do not have a landline, and use a cellphone exclusively. Does anyone know if cellphone numbers are included in the call lists?

So if you have a media monopoly commissioning the polls, making them big chunks of their broadcast airtime ( revenue baby! ) and basing tedious 'analysis' ad infinitum for further broadcast filler, how can anyone look at them as legitimate?
They are quite literally creating their own News!!!

The only poll I care about is the big one that
will be taken on November 2nd - the one where
every registered voter can (if they wish) participate in.

These polls are inherently flawed.

For one they poll a small number of people and using attributable fractions, multiply this across the entire population. Flawed.

They only count likely voters. So newly registered voters aren't in there. Also, people who usually don't vote but will definitely vote in this election.

It also doesn't count people with a cellular telephone as their main form of being reached. These days that's pretty much the entire population under 21.

Polls are like vegas odds, you will not know the results until it's over. Oh yeah vote bush!!!!

Originally posted by coronamoz
It also doesn't count people with a cellular telephone as their main form of being reached. These days that's pretty much the entire population under 21.

This is exactly what I'm saying. I haven't seen the stats, but there has been a large voter registration turnout. I believe these new voters are the young. Very few of my friends have landline telephones, and I'm 30. In my experience, the people who are being polled are the people who have landline telephones, and are largely homeowners.

Because of a variety of practical limitations on polling procedures, current opinion polls really only sample one significant portion of the population: people who are too dumb to screen their phone calls.

They may just be emphasing that Bush is winning by a percent. This gives the impression to Americans to not bother voting for Bush as it seems guaranteed that he will win even if it is by a small margin. On the other hand you have Kerry supporters that don't usually vote but turn out after seeing that Kerry's behind.

The polls seem to be an addive incentive in Kerry's favor.