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 : Sharks in the Septic Tank -- Ignore unavailable to you. Want to Upgrade?

To: Lane3 who wrote (12545)	4/28/2001 11:34:48 AM
From: epicure	Respond to of 82486

I am not sure his purpose was served- unless you mean by baiting, simply answering. I think he has demonstrated to all of us here how he thinks and what he thinks is a logical argument. And I have set out what I think is a logical argument. I think those two arguments can stand together for anyone to see. His intent may have been to bait, I don't know, I certainly didn't feel that. I felt he was just desperate to try to prove his own point. Such desperation often leads to faulty reasoning. jmo

To: Lane3 who wrote (12545)	4/28/2001 2:12:06 PM
From: The Philosopher	Respond to of 82486

All it goes to show is the danger of the misuse of statistics. In a previous stage of my life I was engaged in the business of creating and analyzing polls. It was totally fascinating. After grad work in statistics I spent a year digging in to the psychology of polling. Wonderful and terrifying stuff. We all know about the self-serving politician "surveys." "What do you think should be done to save Social Security? ___ I support Senator Dingabat's proposal to save Social Security at no cost to the taxpayers. __ I think Congress should let Social Security die and throw all the folks who were relying on it onto the welfare rolls." Gee -- Senator Dingbat finds that 78% of his constituents support his plan to save SS. Whee!!! But things are lot more complex than that. For example, do you know that given a range of numeric options most people will exclude the bottom and top options (unless it's obvious that those are what's wanted, and even sometimes then)? So let's say the number of gun deaths in a city in a given year was 30. Your question is: "Which of the following numbers is closest to the actual number of deaths in Metroville in 19__?" If your client is a gun control agency and their bias is to show that the public underestimates the number of gun deaths, you give as the options 20, 24, 27, 30. If your client is a pro-gun lobby and you want to show that the public overestimates the number of gun deaths yuou give as the options 30, 34, 37, 40. Both questions have the correct answer in them. But neither poll is likely to get the correct answer as its result. Instead, each one is likely to give the result the client wants. That's intentional distortion. Far more dangerous, IMO, is unintentional distortion, bias introduced inadvertently by the way questions are asked, the order, etc. From having been in the field, I can tell you that I believe virtually no polling results. I also know that I can prove virtually any proposition with statistics. Don't care what it is. I can find statistics somewhere to support it. I think no American should be allowed to read a single poll until they have read Darrell Huff's simple, short, and totally superb book "How to Lie with Statistics." I can tell you, it's an eye-opening read. If you have never read it, do so!