March 16, 2004Most national polls, as well as surveys in the 16 crucial battleground states, show the race between President Bush and his Democratic challenger to be so close that the election seems mere days away instead of almost eight months. Already, factors that might seem almost trivial in a less evenly divided nation have begun to loom large -- whether Ralph Nader and other minor candidates will play spoiler roles, for example, and whether geography will defeat the Democrats.
The Democrats are more likely than the Republicans to "waste" a large share of their votes by racking up wide margins in several states. The problem for the Democrats is that, in terms of Electoral College votes, a California landslide is worth no more than a one-vote squeaker. Being more evenly distributed, Republicans tend to "waste" large numbers only in Texas. So, once again, this year's popular vote might well be more Democratic than the electoral vote.
With Sen. John Kerry of Massachusetts the de facto Democratic nominee, we are now awash in polls. Some just pit Bush against Kerry; others include Nader. Some survey registered voters; others home in on "likely" voters. Some push the undecideds to say which way they "lean"; others don't. So, naturally, the polls don't produce identical results.
The latest Gallup poll conducted for CNN and USA Today shows Kerry 5 points ahead in a two-way race among registered voters, 50 percent to 45 percent, and 8 points ahead among likely voters, 52 percent to 44 percent.
Likely voters once tended to be more Republican than registered voters. But, in politics, anger is a stronger emotion than love, and dissatisfaction with Bush is so great among older Americans, a very high turnout group, that Democrats now have an edge among likely voters.
When Nader is thrown into the mix, Kerry drops 3 points to 47 percent among registered voters, Bush stays at 45 percent, and Nader pulls 5 percent. Among likely voters, Kerry drops just 2 points, to 50 percent, Bush remains at 44 percent, and Nader draws 2 percent. Clearly, Nader's support comes at Kerry's expense.
In the nation's "reddest" states -- ones that Bush carried by at least 5 points -- Bush leads in a two-way trial heat among likely voters by 3 points, 50 percent to 47 percent. Kerry leads by 13 points in the "bluest" states, the ones where Al Gore won by 5 points or more. In the "purple" states -- where the 2000 contest was closest -- Kerry is ahead by 16 points, 55 percent to 39 percent. For Kerry, the problem is that the Electoral College magnifies the clout of states with small populations. And except for Delaware and Hawaii, they tend to vote Republican.
Other polls vary somewhat. The Fox/Opinion Dynamics poll presents only a two-way trial heat, which has the race tied at 44 percent among registered voters. ABC News and The Washington Post have Kerry ahead among registered voters by 9 points in a two-way contest and 7 points when Nader is added. A joint poll for National Public Radio by the Republican firm of Public Opinion Strategies and the Democratic firm of Greenberg Quinlan Rosner Research puts Bush ahead by 2 points in a two-man contest among registered voters.
The Associated Press/Ipsos poll found Bush ahead by 1 point overall in a three-way race, ahead by 8 points among men, and behind by 7 among women. According to that poll, the race is tied at 46 percent among voters with college degrees, but Kerry is ahead by 5 among those with a high school diploma or less, and Bush has a 7-point lead among those with some college. The poll also shows Kerry leading by 10 in the Northeast and 5 in the Midwest, while Bush leads by 5 in the South and 10 in the West. Perhaps the most telling statistic is that 78 percent of Republicans say they will "definitely" vote for Bush and 75 percent of Democrats say they "definitely" won't. Among independents, 38 percent definitely won't support Bush, 28 percent definitely will, and 31 percent would consider voting for someone else. With Nader in the race, the three-way polls are the ones to watch. But never give too much credence to any one poll. And avoid the temptation to embrace whatever poll produces the result you'd most like to see. The "Olympic scoring" method works best: Throw out the highest and lowest poll findings and average the rest. The result is likely to be a fairly good snapshot of this exciting and unpredictable race.
March 16, 2004