Florida

By DJ Drummond ~ October 24th, 2004 @ 10:48 pm

If John Kerry wins the election next week, he pretty much has to win Florida to do it. And if President George W. Bush is re-elected next week, it’s hard to imagine him doing it without taking Florida. The state polling has something for everybody. A couple polls have Kerry up by a point, others show the President ahead, and others simply call it even.

But I did find some interesting things, by reading some of the articles attached to the poll releases. Specifically, I looked at articles in the St. Petersburg Times, the South Florida Sun-Sentinel, the Quinnipiac poll, and the Mason-Dixon poll. They don’t sort out all the differences, but they do explain a few things.

The SP Times has Bush and Kerry even at 46%, with 7% undecided. The article concentrates on personal comments from individuals, which is interesting but not all that conclusive. It’s interesting though, that the SP Times hired two firms to do their polling; Schroth & Associates, who are partisan Democrats, and The Polling Co., who is identified as a “Republican firm”. If you use results from two admittedly partisan groups and weigh them evenly, why would anyone be surprised the result is a tie? With that in mind, especially the fact that the newspaper is mixing results from different sources, I will not waste time on further details from this poll, except to remind the readers to always check sources.

The Sun-Sentinel reported the results from the Research 2000 polling agency. Their description of the pool of respondents is interesting; “600 registered Florida voters who have a history of voting in statewide elections” (my emphasis). The Research 2000 poll also noted that the poll reflects “the state’s party affiliations and demographics by age, gender, race, and region“, which means they weighted the results to match the 2000 turnout, which I find rather suspect at the moment. The Research 2000 poll gave Kerry advantages with newly registered voters, but says Bush has strong leads “among middle-aged voters and Hispanics.” If that statement is true, then Bush has a lead the R 2000 crew is not showing, I think.

The Quinnipiac poll goes into demographic detail, and it matches the R 2000 statements. Kerry is close only if lots of new and young voters show up. Most notably, Q says Bush is pulling 12% support among Florida blacks, and if that’s true, Bush has already won Florida and Kerry just doesn’t know it yet.

The Mason-Dixon poll is much more the classic poll in its questions and demographics. Bush leads by 3 points in their summary, but they weigh Democrats more heavily, 45% to 41% of weighted respondents. Key in the results is that Bush still out-polls Kerry for favorable name recognition, 49% to 41%, indicating that Bush support is stronger than indicated by the polls, especialy if (as I suspect) the Republican/Democrat balance is more even than the weighting reports.

Just reading the news can brighten your day, if you pay attention.