Free Republic
Browse · Search
Topics · Post Article

To: muawiyah
Some people are easily fooled by professional pollsters. They didn't accept the data they collected as is ~ they filtered it according to a number of standards to make sure they got the answer they wanted.

What is this? You're just babbling and hoping to bluff your way through this debate? Who's fooled? Conservatives that didn't believe the polls leading into November 2012 and had to admit they were wrong and the polls were right?

Man, you are not smarter than all these pollsters when it comes to methodology. You read a few articles and think you know what you're talking about, yet the poll results speak for themselves. These folks know their business pretty well - and are getting better and better at it.

You're STILL avoiding the fact that the 2012 election proves you wrong. The polls were very accurate - especially the state polls. And an aggregate of those polls, as evidenced by statisticians like Nate Silver, allowed him to predict all 50 states correctly. Almost all the polls showed Obama winning, and he won. The state polls painted an ever clearer picture in forecasting an Obama victory AND strong Democratic Senate victories. There were many conservatives, for example, that didn't believe the polls that showed Akin, Mourdoch, etc, losing. Yet they all lost just like the polls said they would.

The point being that the only way to beat the small minority effect is to increase your total sample size into at least 11 times as you've ever planned on doing for a given situation.

Look, you can drone on and on about your theory, but it's just wrong. The response rate is good enough for polling to be accurate. Obviously not every poll is trustworthy as some of them use loaded questions to achieve a result, but basic generic ballot questions, presidential approval, x versus y, etc, are extremely accurate (especially when you can generate an aggregate of polls). We know some polling, such as primary polling, is tough to do and the margin of error is often greater than normal. In the end though, the science and methodology behind polling has come a long way and the results are very accurate. As I keep saying, you only need go back to the 2012 election to see that.

48 posted on 03/24/2013 7:44:55 PM PDT by Longbow1969
[ Post Reply | Private Reply | To 45 | View Replies ]

To: Longbow1969
you aren't very well informed about the discussions on FR running all the way back to a year and a half ago regarding these polls.

I pegged Romney as a loser early on and demonstrated why he would lose ~ looking at a single voting event.

I began to take notice of the fact the progay polls weren't being supported by public referenda. This particular situation is hardly new ~ the polls have always said the progay stuff would win, then on the day the votes are counted they lose.

Maryland had the gaypolls saying gayness wins, and it won, but guess who counts the ballots.

Use your head ~ this is a Washington Post poll. It may well have originally said gays lose ~ but the Post has a reputation to keep up ~ so the poll will say what they say it means.

Get your argument into line with what PEW says and show something that tells me you know something about statistical sampling methods, even sigma 6 quality control sampling ~ something ~ you have no credentials that I know of. Thousands of Freepers know mine.

52 posted on 03/24/2013 8:00:15 PM PDT by muawiyah
[ Post Reply | Private Reply | To 48 | View Replies ]

Free Republic
Browse · Search
Topics · Post Article

FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794 is powered by software copyright 2000-2008 John Robinson