Bush with lead in St. Pete Poll
Posted by
According to this poll, Bush has an 8 point lead and has crested the 50% mark. The margin of error is 3.5%, so this is a statistically significant difference.
This poll, however, shows the biggest gap between the two candidates in a long time. That could mean that:
1)The poll is just veering to the extremes of the margin of error
2)Mcbride did bad enough in the last debate to swing a significant number of voters to Bush
3)It is under counting Dem voters, over counting Repub votes, or not counting likely voters appropriately.
That last is interesting to me, because from everything I know, likely voters are defined as voters who voted in two of the last three elections. If they used that standard, then I think there is a pretty good chance that they are undercounting Dem voters. That standard will not take into account the fact that the Dems appear to be very motivated this year, and that the bitterness from 2000 is still an issue. If it is, then the Dems will turn out in better numbers than the standard formulas would indicate.
There will be more polls in the next few days, and, taken together with this poll, should give us a better feel for how the race’s last week will play out.
UPDATE: I missed this from the original story:
“”The statewide telephone survey of 800 registered voters who described themselves as likely voters in the Nov. 5 general election…”
So, the question is partly answered. I still could not find any reference to the partisan breakdown of the respondents, so they could still be modeling the dem turn out incorrectly.
keep in mind that the term “likely voters” most likely excludes people who weren’t _allowed_ to vote.
Comment 10/26/2002
Very similar results to Survey USA poll out on
Friday which showed a bounce from the final Jeb-McBride debate:
Jeb 52% Bill 45% Undecided 3%
http://www.surveyusa.com/home.asp
Comment 10/27/2002