Friday, October 20, 2006

Why Do We Even Pay Attention To Polls?

Take a look at these Zogby polls taken two days before the 2002 election, contrasted with the actual results, at Riehl World View (linked above).

Combine that info with what Jim Geraghty dug up from 2004; again, what was being reported by pollsters and the media didn't correlate with the actual results.

Betsy's Page also notes a study which shows Clinton polled more strongly in 1996 than correlated with the actual votes, and suggests it wasn't just random chance.

I think more often than not, in this day and age polls are largely used not to report possible news, but to attempt to impact the news.

2 Comments:

Blogger Jim said...

I agree with you.

Many polls are biased, and/or misinterpretted by opportunists on the basis of their results.

I have one friend who suggests we purposefully mislead the pollsters, while others suggest that the gap is evidence of election tampering.

What should we do about them though? I really don't know.

12:18 PM  
Blogger Laura said...

"I have one friend who suggests we purposefully mislead the pollsters, while others suggest that the gap is evidence of election tampering."

I think both things happen...the former probably happens more often than it used to (as those suspicious of polls deliberately contribute to their unreliability), but we've seen a lot of examples of polls used to "push" the electorate in the last decade, too. Here in California we've seen that in gubernatorial races in years past, as well as in the national elections.

The question you raise is one I sure don't have an answer for, either! I think it would be quite interesting if some political season everyone -- media, politicians, consultants, etc. -- voluntarily agreed to forego polls. But that would never happen. (Grin)

Best wishes, Laura

12:46 PM  

Post a Comment

<< Home

Newer›  ‹Older