Leaving the community

Peter Duniho wrote:

"Matt Whiting" wrote in message
...

Polls are facts about statistics.

I'd say just the opposite. Polls are statistics about facts.


Statistics are always about facts. Polls are the facts about the
statistics.

No, a fact is an invariant. If you take a poll and then take another
poll, you'll get a different result. That isn't factual, sorry.


Yes, most of these polls have significant biases.


Such as?

Such as who they talk to, where they conduct the poll, what time they
poll (as mentioned earlier, the working Republicans may not vote until
after the welfare liberals are done), and many other factors.


Matt

No, a fact is an invariant. If you take a poll and then take another
poll, you'll get a different result. That isn't factual, sorry.

Actually a poll is a statement of fact: the people polled did in fact say
what the poll says. The problem arises with how you extend the poll to
represent the unpolled.

On Fri, 5 Nov 2004 18:15:05 -0600, "Greg Butler"
wrote:


No, a fact is an invariant. If you take a poll and then take another
poll, you'll get a different result. That isn't factual, sorry.

Actually a poll is a statement of fact: the people polled did in fact say
what the poll says. The problem arises with how you extend the poll to
represent the unpolled.

And whether those being polled told the truth. I wouldn't and many
told them their vote was private. I think it's becoming trendy to lie
to the pollsters.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com

Greg Butler wrote:

No, a fact is an invariant. If you take a poll and then take another
poll, you'll get a different result. That isn't factual, sorry.


Actually a poll is a statement of fact: the people polled did in fact say
what the poll says. The problem arises with how you extend the poll to
represent the unpolled.



Which is the entire purpose of a poll, so I think that is a given. And,
yes, I agree that herein lies the problem.


Matt

"Matt Whiting" wrote in message
...
No, a fact is an invariant.

Really?

So, a statement regarding the position of the sun during the day isn't a
fact? After all, it varies continuously throughout the day.

You have an odd definition of what's a "fact".

Yes, most of these polls have significant biases.


Such as?

Such as who they talk to, where they conduct the poll, what time they poll
(as mentioned earlier, the working Republicans may not vote until after
the welfare liberals are done), and many other factors.

Only an ignorant asshole would seriously claim that all Republicans work
while all Democrats are on welfare.

Oh, I'm starting to see what your problem is...

In any case, if you have a legitimate beef with the polls in question, state
them. So far, you've made no suggestions about why those polls are
significantly wrong, and as I've already pointed out, the chances of those
polls being correct are MUCH greater than the chances of them being
drastically incorrect.

Pete

Peter Duniho wrote:

"Matt Whiting" wrote in message
...

No, a fact is an invariant.


Really?

So, a statement regarding the position of the sun during the day isn't a
fact? After all, it varies continuously throughout the day.

Yes, that is a fact because it includes the element of time. If you
take two polls at the same time in the same place you will get two
different answers.


You have an odd definition of what's a "fact".

Much better than yours though.


Matt

"Matt Whiting" wrote in message
...
Yes, that is a fact because it includes the element of time. If you take
two polls at the same time in the same place you will get two different
answers.

Not if you poll the same people. If you're going to argue variant versus
invariant, you need to hold ALL relevant factors invariant when you're
talking about invariant.

You have an odd definition of what's a "fact".

Much better than yours though.

Huh? That's the best you could come up with?

Peter Duniho wrote:

"Matt Whiting" wrote in message
...

Yes, that is a fact because it includes the element of time. If you take
two polls at the same time in the same place you will get two different
answers.


Not if you poll the same people. If you're going to argue variant versus
invariant, you need to hold ALL relevant factors invariant when you're
talking about invariant.

Maybe, maybe not. People are notoriously fickle. And even if you get
the same responses twice, it only tells you one thing: what those people
said at that instant. It doesn't tell you that much about the broader
population because of the biases we discussed earlier. And thus exit
polls are a complete waste of time because nobody cares what some subset
of the voters think.


You have an odd definition of what's a "fact".

Much better than yours though.


Huh? That's the best you could come up with?



Beats "huh?"

Matt

In article ,
"Peter Duniho" wrote:

Yes, that is a fact because it includes the element of time. If you take
two polls at the same time in the same place you will get two different
answers.

Not if you poll the same people.

assuming that they will answer truthfully each time....

--
Bob Noel

Peter Duniho wrote:
So far, you've made no suggestions about why those polls are
significantly wrong, and as I've already pointed out, the chances of those
polls being correct are MUCH greater than the chances of them being
drastically incorrect.

So which is it? Is the poll a fact or nearly a fact?