Posted on 09/26/2012 12:18:38 PM PDT by smoothsailing
We are speaking of the difference in vote totals of each party ~ not the total percent of the total vote.
Thank you for your insight.
No, my math is not wrong. I simply reported the comparable statistics from the last Presidential election.
Simply dividing the difference in the two vote totals by the total number of votes don't tell you anything about the difference!
Yes, it does. The difference is simply expressed in a proportion of the entire sample.
Look, I see your point. You can compare the two subsets to each other, instead the entire group, and come up with a different percentages. But, that's not how this difference was reported. It is the percentage of the entire sample, not a percentage of any subset.
You can't calculate it your way, then say that it means that a percentage calculated the other way (and the way everyone else is doing it) is comparable. If you know as much about statistics as you claim, you should know better.
Personally, I don't think that you can just adjust a poll according to self-reported political affiliations. However, I think it's both intellectually and ethically dishonest to report a poll as "correct" or "accurate" when there is such a clear imbalance in the sample.
Pollsters are just not willing to admit their sampling methodologies aren't really random. If they did, they would be out of business. These days, telephone poll respondents are effectively self-selecting. Between caller ID and call screening, most poll responders know exactly who they will be talking to before they even pick up the phone.
When you add in the different behaviors of the major voting groups (especially on the weekends), the possibility of intimidation (do I really know who is calling me for my opinion?), and telemarketers using the poll/survey exclusion to get their foot in the door, there is simply no way to get a genuinely random sample.
I'll also note that the party identification that I cited above is also suspect. It was derived from exit polls, and those are also self-selecting.
Pardon me but I got a couple of thoughts on this matter.
Your reputation precedes you as far as I’m concerned. You are always involved in some dispute with another Freeper or some such.
HOWEVER....this time I agree with you. I think to properly reflect the populace that the polls should reflect , oh I dunno, maybe the % Dem/pub turnout from the prior election.
HOWEVER again, jeez louise, I once worked for five years on a job requiring specific knowledge of samples and sample sizes....quality control to be exact.
I learned then and still now that sample sizes are fairly accurate. Gotta find the coveted “mean” and all that.
But come on, today all these polls come out showing Obamer running away with this thing in swing states. That’s just not believable I say based on my experience with statistics. It’s not likely at all that polls would suddenly change so dramatically overnight.
Still and so, hey, if the American people re-elect Obama I have decided to accept their choice and not grouse about it.
Do I see it as the end of America as I know it? Yes.
But we have a majority rule and so the majority shall rule.
I believe with all my heart America will make the right decision as regards this election.
But if not I will stoicly accept the rule of the majority.
Unhappily, to be sure. But what shall be, shall be.
“Personally, I don’t think that you can just adjust a poll according to self-reported political affiliations.”
__________________________________
That might be singularly the stupidest thing I’ve ever read on FR regarding polls. Voter turnout is NOT random!! One more time invade you we’re unable to let it sink in the first time. Voter turnout is NOT random.
Usually, I ignore personal attacks like this. You should know better. And if you don't dial it back a notch, this will be my last response to you.
No, voter turnout is not random. But, you can't expect a telephone poll to predict it reliably, especially when the poll results are well within the margin of error.
Instead of taking my comment out of context (like the Obama Media Group does for Romney), maybe you should read the rest of my response.
Poll responders are not randomly selected, either. The "opportunity" may be random, but the choice to respond is not random. There's too many factors that essentially make a poll respondent self-selecting. So, you can't adjust the results of one self-selecting poll with the results of another self-selecting poll and say that it is "better".
However, as I noted in the part that you deleted, I think that it is ethically and intellectually dishonest to claim that a poll is "accurate" when it is clear that your samples have a significant number of outliers.
Either:
1) There’s a vast leftwing polling conspiracy
or
2) We are part of the slant.
Maybe the left has a weak turnout for BO, but also, Romney has a weaker base demonstrated daily on FR by people who’d rather teach the GOPe a lesson than send Obama packing.
#2 seems more and more likely.
We assume the left won’t turn out as well as 2008. But I think a lot of spite is going to materialize in the Laz-e-boys and popcorn sales when the puritanicals sit home and passively protest against republicans.
....”a guide to where a politician might want to spend scarce campaign resources.”
Please explain to me why Obamy is campaigning today in Ohio, while enjoying a 53% to 43% lead over Romney?
I have my own theory about what the polls are telling us and why even the pollsters don't know what they mean.
It's not all bad news, just confusing news.
As you know all political parties attempt to stir their followers to go vote for their candidates. Billions have been spent based on the idea that this stuff works ~ that you can campaign and raise your candidate's popularity, and more people will come out to vote for him. BILLIONS!
I have no doubt that's the way it is almost all the time. But usually the parties and their campaigns start with candidates of more than uncommon popularity. For example, Eisenhower was very popular. He had a large edge with the populace and could draw on Democrats for votes as well. Truman had even tried to get him to run as a Democrat. He'd won either way.
The others we know about were not in the Ike category of popularity, but they weren't like Charlie Manson either. Sure he had his following, but you get tired of the glow in the dark eyeball trick in a really short time.
Obama was highly popular before folks got to know him. Now he's down in the dumper.
What about the other guy? Romney ~ is he popular, or is he just the guy who managed to snag the nomination? What if BOTH major party candidates are perceived by most voters as not being all that popular, or even someone to respect?
Then, as the campaign progresses and we get to know both of these guys better their existing popularity could drop leaving more and more voters confused, upset and disturbed that they don't have a candidate in this race.
The pollsters should be detecting this in the number of non-respondents they encounter ~ those are the folks who hang up on the pollster right off, or refuse to answer questions, or who hector the poll caller, or even preach to them before hanging up.
None of these guys will tell you, but if it's going on, and there are fewer and fewer actual probable voters in the pool within which the sample is being taken, the results should, at some point, come to a resting point where any number of additional samples (calls to people) will not change the results beyond random chance ~
I think we have begun hearing that with ever increasing incidences of pollsters reporting ties, and once one guy does it the others will bravely come forward and they'll tell us the same thing.
This before the debates and without any real campaigning having taken place.
Does not look good
It is what it is. Compare A to B, not A to A+B
I’ll take #1 sprinkled with just a tiny pinch of #2.
Good point. We need to keep our principles, but take the long view. All this hand wringing isn’t going to produce a conservative majority.
Several reasons ~ it's on the way to Chicago. Michelle has friends in Cleveland. ........ just hundreds of things like that.
Then I suggest we name the gulags after you.
“It’s in a symposium i reffed the other day ~ the lead lecture was by Zogby ~ and he went right to comparing his approach to Rassmussen’s approach.”
Good grief are you freaking kidding me? Is this the same John ‘the Retard’ Zogby that just this week made the claim that Obama (a black, Marxist, pro gay marriage, failed president) has a significant lead among NASCAR fans??? WOW I say!! If you are stupid enough to follow this dipshit then you really have ZERO credibility.
Can't argue with that. They wanted to find Obama in the lead and did..
Everybody has a bad poll or two lying about ~ but say, tell us how you’d go about sampling specifically for NASCAR fans. Bet that problem is a bit more difficult to crack than you imagine.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.