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 : Foreign Affairs Discussion Group -- Ignore unavailable to you. Want to Upgrade?

To: Jim Willie CB who wrote (150294)	11/1/2004 4:38:23 PM
From: Neocon	Read Replies (1) \| Respond to of 281500

I thought that was more or less what I said, but no matter. I never studied statistics. I just picked up a few concepts along the way, so I am not too worried about it. Getting a better handle on it is always worthwhile, of course.

To: Jim Willie CB who wrote (150294)	11/3/2004 10:23:13 PM
From: Bilow	Read Replies (4) \| Respond to of 281500

Hi Jim Willie CB; Re: "... 3% margin of error individually means 4.24% margin of error for the difference ..." This simply isn't true. You're failing to take into account the obvious fact that among any individual who happens to be in your poll, there is a nearly perfect *inverse correlation* between votes for Kerry and votes for Bush. Consequently, you can't add percentages with the square root scaling law that applies to differences of *independent* random variables. In other words, the sample sets for the two measurements, the Bush vote and the difference vote, are not independent measurements. They are almost perfectly correlated. Neocon, miraculously, is correct despite his not having studied statistics. If there is a 3% margin of error on the vote for Bush, and the vote is approximately 50% each, then there will be a 6% margin of error in the vote difference between Bush and Kerry. Hey, you could use the 4.24% figure if you were using different sample sets for the two measurements, but you're not. -- Carl P.S. I did study statistics, among other things.