BY AL CROSS
The Courier-Journal
Democrat Scotty Baesler has largely erased Republican Jim Bunning's advantages in Kentucky's tight race for the U.S. Senate, but still faces a challenge in getting his supporters to the polls Tuesday.
Those are the major findings of the Courier-Journal's latest Bluegrass State Poll, taken Oct. 20-25.
The survey showed Mr. Baesler leads his fellow congressman by 1 percentage point, a statistically insignificant edge because it's within the poll's error margin of 4 percentage points.
The results are based on interviews with 589 registered voters who said they were likely to vote, but history shows many of them will not cast ballots.
The poll showed at a more likely level of voter turnout, Mr. Bunning had the lead, but still not beyond the margin of error.
Thus, as in many elections in the middle of a presidential term, turnout could decide the winner.
"It's absolutely critical," said J.B. Poersch, Mr. Baesler's campaign manager.
And turnout could hinge on a host of factors, from voters' feelings about President Clinton to the weather on Election Day.
"Hope it rains," said Mr. Bunning's campaign coordinator, Kyle Simmons.
"Everyone's known from day one this was going to be a race that goes down to the wire, and that's exactly what's reflected in the Courier's poll."
The survey showed Mr. Baesler is attracting more support from traditional Democratic voters than he did in the last Bluegrass State Poll, taken Sept. 25-30, but such voters have been less likely to turn out in recent mid-term elections.
Not only has Mr. Baesler firmed up his base, he has largely eliminated the biggest advantage Mr. Bunning enjoyed in last month's poll -- a great disparity in the voters' personal impressions of the candidates. Last month, more than half of the likely voters said they viewed Mr. Bunning favorably.
Now, after suffering a month of attacks in TV commercials from Mr. Baesler and the Democratic Party, he is viewed favorably by only 44 percent and unfavorably by 32 percent.
Republicans' TV attacks on Mr. Baesler appeared to cause him serious problems in the last poll, but little damage since then. Impressions of Mr. Baesler were about the same as a month ago: 38 percent favorable and 31 percent unfavorable. He did show a noticeable gain in favorability among voters 50 and older, who are more likely to vote than younger age groups.
Mr. Baesler's gain in support appears to have come partly from voters who were undecided last month, because his share of the vote went up by the same amount the undecided share declined, 7 percentage points.
Mr. Bunning's support rose by 2 points in the past month.
While Mr. Baesler may appear to have some momentum, it has come almost entirely among voters who tend to lean Democratic such as women, people over 50, middle-income voters, high-school dropouts and union households.
Mr. Baesler's newfound firmness in the Democratic base also is reflected geographically. The former mayor of Lexington, he gained in his home Bluegrass region and the two areas that have provided key margins for recent statewide Democratic victories eastern Kentucky and Jefferson County.
In Jefferson County, Mr. Baesler led Mr. Bunning 48 percent to 43 percent, but that is also within the margin of error, which is 10 percentage points for that smaller sample.
The other big battleground of the election is western Kentucky. There, Mr. Baesler leads by 45 percent to 40 percent, also with a 10-point margin of error.
Mr. Bunning also appeared to solidify his base, picking up support among Republicans who only weakly identify with their party and among residents in his home region of Northern Kentucky, where he has been running TV commercials for weeks and Mr. Baesler has run no ads.
Mr. Baesler appeared to gain among voters who say President Clinton should remain in office. Mr. Baesler voted Oct. 8 for a limited impeachment investigation in the House, while Mr. Bunning supported a wide-open investigation, but Mr. Baesler has made only glancing references to the issue.
The issue cuts strongly both ways, according to voters who were polled and agreed to follow-up interviews.
Tim Leake, a Jeffersontown electrical contractor, is a registered Republican who says he usually votes Republican, but will vote straight Democratic for the first time ever to protest GOP treatment of Mr. Clinton.
"I'm sick of the way our party is doing," said Mr. Leake, 35. On the other hand, registered Democrat Mike Deeley of Louisville said he has largely abandoned his party because of Mr. Clinton and Gov. Paul Patton. Mr. Deeley, 67, a part-time maintenance worker, said he will vote for Mr. Bunning.
Judith Hendrix of Louisville, an independent-minded Democrat who is undecided in the Senate race, said she is less likely to vote Republican because "I think they've gone too far" against Mr. Clinton.
Political observers in both parties think the Clinton controversy is likely to boost turnout among strong partisans.
Any boost for Democratic turnout in a midterm election is welcome in light of the depressed Democratic turnout that led to the Republican landslide in the 1994 midterm election, said Mr. Baesler's pollster, Mark Mellman of Washington, D.C.
But Mr. Mellman said he expects more voters to be spurred to the polls by local races, which are on the same Kentucky ballot as federal races for the first time in memory.
Mr. Mellman said he expects turnout to be above 50 percent a level at which the poll indicated that Mr. Baesler has about an even chance of winning. Below 50 percent, Mr. Bunning has the advantage.
Bunning pollster Jan von Lohuizen, who puts less stock in the influence of local elections, said he expects turnout to be about 45 percent. But he and Mr. Mellman agreed on one thing about turnout, in the same words: "It's hard to predict."
The Bluegrass State Poll is based on telephone interviews conducted Oct. 20 to 25 with 901 Kentucky adults. Of these, 589 said they were registered and definitely or probably would vote in the Tuesday election.
Trained interviewers called households whose telephone numbers were randomly selected by a computer. Then, one adult in each household was selected randomly to be interviewed for the poll.
The results have been weighted to properly balance the sample by age, sex and regions of the state.
The poll's margin of error for the likely voters is 4 percentage points.
This means, in theory, in 19 out of 20 cases the poll results will differ by no more than 4 points above or below the results that would have been obtained by questioning all likely voters in Kentucky who have telephones. Percentages based on subsamples are subject to a higher potential margin of error.
The Bluegrass State Poll conforms to the standards of the National Council on Public Polls.
