Need comments on LTTE on recent Gallup poll

[T Roosevelt]

Limit is 250 words, so I have some room to play...


Understanding Gallup
News organizations love to point to pre-election polls as evidence of trends, winners, and losers. Unfortunately, too few of them bother to dig into the techniques used and assumptions made by the polling organization when presenting the results. The current Gallup poll provides a perfect example:

Gallup Poll of September 13-15, which shows Bush winning by 55%-42%, assumes that this November 40% of those turning out to vote will be Republicans, and only 33% will be Democrat (and 28% Independent).

According to John Zogby (well known pollster): "If we look at the three last Presidential elections, the spread was 34% Democrats, 34% Republicans and 33% Independents (in 1992 with Ross Perot in the race); 39% Democrats, 34% Republicans, and 27% Independents in 1996; and 39% Democrats, 35% Republicans and 26% Independents in 2000."

So despite the fact that more Democrats than Republicans turned out in each of the last three Presidental elections, Gallup has made the decision to bias this poll using a faulty sample, spreading a false impression of this race. Statisticians will recognize the invalidity of this, and so should the public in general. Interestingly enough, Gallup's CEO, James Clifton, is a GOP donor. This is a serious conflict of interest, and brings into question the Poll's legitimacy.

[kstewart33]

1. That's a great letter

You make your points without sarcasm or attitude. I wouldn't change anything. What to add? If you have the space, you might point how how cell phones and caller ID are making polls less accurate, which further challenges the reliability of all these polls.

Good job!

[glitch]

2. very well done - where are you sending it? nt

[T Roosevelt]

3. Orlando Slantinel

I've had a couple letters posted...the last one was over a month ago.

[glitch]

4. Be sure and let us know if they print it! nt