Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Liberalism: Do You Agree We've Had Enough of It? -- Ignore unavailable to you. Want to Upgrade?

To: Geoff Altman who wrote (51220)	10/12/2008 9:16:38 AM
From: nigel bates	Read Replies (3) \| Respond to of 224703

That's the difference between us. I treat arguments on their merits, not on the basis of what I want the answer to be. Whoever Steven Warshawsy is, he knows little about polling. >>The first two issues would require a detailed analysis that is beyond the scope of this article<< In other words, he doesn't even make an effort to justify two thirds of his argument. >> First, there is the well-known problem of bias that results from how POLLS are worded.<< Strange, then that pretty well all the polls from every independent polling organisation, many of which are owned by Republican supporters, and each of which uses a different form of words, are coming up with the same answer. >>Second, the raw data for the POLLS almost always is "adjusted" by the pollsters to give more WEIGHT to the Democratic responses.<< Here Mr Warshawsky displays his utter ignorance of polling. The reason pollsters weight their polls is to make sure their sample matches the known percentages of the population (at local, state or national level, depending on the poll) who identify themselves as Democrats of Republicans. These are relatively stable figures. Pollsters started doing this not because they are biased (again, given the number of different polling organisations, that is an absurd idea), but because they have found that it gives them more accurate results (after the last few elections, they started to go back and see where they went wrong). If you interview 500 people at random, then you are likely to pick a sample which is skewed on way or the other. If you and other polling organisations have previously polled 50,000 people to get an idea of the relative numbers of Republicans and Democrats in the state (something they do every month), then you have a fairly accurate way of correcting that skew. Had you any interest whatsoever in polling, then you would read the 538 website. That it is run by democrats is completely irrelevant. They publish all the figures that they use, their sources, and their methodology. You don't have to agree with their conclusions, but you might learn something. >>But the third issue clearly proves my point that Obama is not "winning" the race.<< No it doesn't. It clearly proves that Mr Warshawsky doesn't understand either statistics, or probability. None of the pollsters are saying that Obama has won. They are saying that based on the available evidence of their polling, the probability of his winning has increased to somewhere between 75 and 90%. Sure, a McCain win is still possible. I just wouldn't bet on it. Anyhow, I've outstayed my welcome here, so I'll leave you folks for a while.