The KGB Oracle Forums General Public Discussion (SPECIAL ACCESS) KGB Political Debate Central Opposition to polling data on Header

Pfft, I knew you were just another RINO.

You have confused me for some one else, I have never said I was a republican.

Hey how the hell are the polls doing nowdays ? Bahahahhaha !

Interesting! When they go your way they are accurate. When they don't it is a conspiracy. Create your own reality if you don't like the truth, is what to many repubs do. All I can do is hope that people expose Romney for all the lies he is telling.

LMAO You sure you want to go there !

You must be taking lessons from Mitt. You completely dodged the comment.



No comment I take it from the conservative folks who doubted the polls a week or so ago?

Theyre all crap, for various reasons. I could actually wax poetic for hours about the crap the pollsters do, and I have even better authority than most on the topic because I used to work for one and saw more than enough lollerstats to fill 100 million spreadsheets.

Never trust any poll at all that doesnt have fully published methodology, weighting, and questionnaire. And only half trust those that do. Not that they are intentionally trying to jilt things (the ones that publish their methods anyway, they are typically relatively clean), but polling is an extremely imperfect science that has gotten even more difficult over the years as people switch their usage patterns from landlines to cellphones. The spread that day can often be determined by the pollsters intuition, going by how they think the data should be weighted to be most reflective. Obviously, humans misjudge these types of things quite often.

Also, much like video game reviews... if other pollsters come out with a view, it *can* cause a certain pressure to somewhat conform...unless you are *really* confident that you have some sort of advantage in your evaluation.

Or in other words, the Romney dip - and conversely the Romney rise post-debate could easily be as caused by the *expectations of the pollsters and how they think they should be interpreting the data, as actual changes in the electorate.

On that note, I am pretty nuetral on the whole recent polling crap... havent had any time to spend digging. Just saying theyre all fucked up.

But yes, polls do intentionally skew things quite often. Some more than others. They also get things wrong. Though it happens to both parties equally, over any reasonable timescale the amount and degree of skewing is actually about 1:1 in most circumstances. The people who get hosed the most are 3rd parties, who often get "weighted" out of the running altogether to create the image of no possible support.

And if you think *that* is a conspiracy theory, even though 90% of the major polls are directly funded by the parties, campaigns, or partisan interests.... then I also have some prime swampland in Florida to sell you.

edit: edit

Nope, we learned that from you Jet. :)

That is why I like real clear. It is an average of all.

Third parties still get the shaft. And it still reflects the same groupthink.

Best polls generally for the two-party comparison is Rasmussen. One well-run poll by a talented pollster is worth any amount of averages of incompetents.