In 1994, a Mason-Dixon poll showed Nashville Mayor Phil Bredesen, then the Democratic gubernatorial nominee, within the margin of error—and therefore in a statistical dead heat—with Don Sundquist, then a Republican congressman from West Tennessee. But when Nov. 8, 1994, rolled around, Sundquist carried Tennessee by a full 9 percentage points.
Incumbent Democratic U.S. Sen. Jim Sasser consistently led his Republican opponent, Bill Frist, in Mason-Dixon polling that year, with the poll rounds closest to the election showing a dead heat. In the end, though, Frist wound up carrying the state by an impressive 14 percentage points.
That’s not to say Mason-Dixon’s polling is chronically unreliable, but Nashville’s mayoral campaigns are using these figures to express their doubts about the mayoral election poll recently commissioned by, and published in, The Tennessean. While each candidate said at the time they were pleased with the poll results, the three leading campaigns nevertheless now find some fault, minor or major, with either its methodology or its sampling—something the research company is used to.
Perhaps predictably, former Mayor Dick Fulton’s general campaign consultant, Bill Fletcher, is most critical of the poll, even though it shows his candidate ahead of the others. What he may not be so happy about is that it showed Fulton with a staggering 96-percent name recognition but a much less healthy 26-percent support among voters. He calls Mason-Dixon polling consistently “inexpensive and unreliable” and says, “If you know what you’re doing, you can actually take a Mason-Dixon poll and figure out what’s going on, but the newspaper has yet to do it.”
While the poll showed Fulton leading with 26 percent, it showed Vice Mayor Jay West at 17 percent and former state House Majority Leader Bill Purcell at 15 percent. With 626 respondents and a margin of error of plus or minus 4 percentage points, it also showed a huge undecided block of 36 percent.
Fletcher’s primary beef is with Mason-Dixon’s methodology. The company, he says, doesn’t make enough of an effort to “screen” respondents—in other words, to identify them as likely voters. What’s more, he says, it doesn’t make an effort to nudge the “leaners”—those who are leaning toward a particular candidate—away from the undecided column. In reality, Fletcher claims, Fulton’s support is closer to the high 30-percent or low 40-percent range. (The Scene’s mailbox or fax machine, however, has yet to receive evidence of such assertions.)
What’s more, Fletcher concludes, the poll was not controlled enough and was too heavily weighted toward certain parts of the county. As evidence, he points to the Mason-Dixon poll results showing at-large Metro Council member Ronnie Steine—who hails from Green Hills—with a healthy 6-percent lead in his race for vice mayor over popular state legislator and district Council member Tim Garrett. That, Fletcher says, shows response from West Nashville was too heavy.
Mason-Dixon says such criticism is par for the course. “The people we interviewed indicated they were registered voters and were likely to vote,” says Brad Coker, managing director of Mason-Dixon Polling & Research Inc., who handled the project. “Did they lie to us? Maybe a few did. We didn’t show up at their door and make them show us their voter registration cards.”
Coker says the grumbling is typical “insider stuff,” and that the Fulton camp is upset at the results because “the very large name-recognition lead and very large money advantage [aren’t] translating into the kind of lead they’d like to see.” Nashville, Coker says, “is a growing city, and maybe people are saying this is the new Nashville, [and Fulton is] the old Nashville.”
Kendell Poole, campaign chairman for West, agrees with some of Fletcher’s claims. “My perspective, particularly recently, of Mason-Dixon polls, is not very good,” he says. “Because the sampling size is so low, it’s not weighted for different areas of the county like we would weight it.”
Poole questions, for example, the efficacy of Fulton’s 96-percent name recognition. “There were a couple of results like that we knew were way off,” he says. Still, “there’s some credibility in the poll, and I would say the indications from it bear some similarity” to what West’s campaign polling shows.
Meanwhile, the Purcell campaign, which has been struggling with the perception of being in a solid third place—as opposed to being, as Mason-Dixon shows, in a statistical tie for second—acknowledges that the sampling poses some difficulty for getting a handle on demographic trends. “A 600 sample is OK,” says Purcell political director Patrick Willard. “They could have gone with 800, which would have reduced their margin of error, but I think it accurately shows a deadlock for second place.”
To reach Liz, call her at 244-7989, ext. 406, or e-mail her at liz@nashvillescene.com
