Kerried AwayThe myth and math of Kerry's electability.
Posted Wednesday, Feb. 11, 2004, at 12:41 AM ETBy media consensus, the race for the Democratic presidential nomination is over. Why? Because John Kerry has won 12 of the 14 primaries and caucuses held so far. And why has Kerry won these contests? Not because voters agree with him on the issues. The reason, according to exit polls, is that voters think he's the candidate most likely to beat President Bush. There's just one problem: The same polls suggest this may not be true.
PRINT
DISCUSS
E-MAIL
RSS
RECOMMEND...
Facebook
MySpace
Mixx
Digg
Reddit
del.icio.us
Furl
Ma.gnolia
Sphere
StumbleUponTwo weeks ago, Kerry beat Howard Dean by 12 percentage points in the New Hampshire primary, convincing Democrats around the country that Kerry was their most electable candidate. How did Kerry win? By racking up a 4-to-1 advantage over Dean among voters who chose their candidate because "he can defeat George W. Bush in November." Among voters who chose their candidate because "he agrees with you on the major issues," Dean and Kerry were tied.
Let me say that again: Among voters who picked the candidate they wanted based on the issues, not the candidate they thought somebody else wanted, Kerry did not win the New Hampshire primary.
OK, maybe Dean wasn't the most electable guy. But in the states that followed, voters applied the same theory to other candidates, padding Kerry's delegate count and aura of inevitability. They figured the guy who had won Iowa and New Hampshire was a winner. So, they voted for him, proving themselves right. The biggest delegate prize on Feb. 3 was Missouri, where Kerry beat John Edwards 2 to 1, filling the airwaves with talk of a juggernaut. How did Kerry thrash Edwards so badly? He won "agrees with you" voters by 10 points—a healthy but not awesome margin, largely attributable to the fact that Kerry was the candidate the media were talking about, since he had just won New Hampshire. No, the people who gave Kerry his enormous vote tally in Missouri—and nearly two-thirds of the state's delegates—were the "can defeat Bush" voters, who went for Kerry over Edwards by a ratio of more than 3 to 1.
Everywhere you look, Kerry collected big wins and delegates for this reason. In Arizona, he squeaked past Wes Clark by just two percentage points among "agrees with you" voters. But he crushed Clark among "can defeat Bush" voters, netting a 16-point victory. In Delaware, Kerry did twice as well among "can defeat Bush" voters as among "agrees with you" voters. In Oklahoma, both Clark and Edwards beat Kerry by 13 points among "agrees with you" voters, but Kerry got away with a competitive finish by thumping them among "can defeat Bush" voters. In South Carolina, Kerry lost "agrees with you" voters to Edwards by a 2-to-1 margin but escaped with a respectable second thanks to "can defeat Bush" voters.
Last weekend, the press wrote Dean out of the race after Kerry beat him 3 to 1 in the Michigan caucuses. A poll of Michigan absentee voters taken by the CBS News Elections and Survey Unit showed Kerry crushing Dean by 29 points among "can beat Bush" voters. But in the same survey, "agrees with you" voters chose Dean over Kerry by four points. To be fair, the poll showed Dean doing 19 points better, relative to Kerry, in the absentee sample than in the final returns. But the logical explanation for that gap is that many absentee ballots were cast before the race turned upside down. And the logical implication of that explanation is that while the poll understated Kerry's share of "can defeat Bush" voters, it was less likely to understate his share of "agrees with you" voters.
Tuesday, the electability factor wasn't just big; it was decisive. The networks anointed Kerry the nominee based on his sweep of Virginia and Tennessee. But Kerry wasn't the first choice of Tennesseans who selected their candidate based on the issues. Edwards was. The "can defeat Bush" voters were the ones who reversed the outcome and put Kerry on top.
All of which raises the $200 million question: Are these "can defeat Bush" voters correct? Is Kerry the most electable Democrat?
It's a hard question to answer, because most of the evidence is circular. If people support Kerry because they think he's electable, he goes up in the polls, which makes him look more electable. The best way to filter out this distortion is to focus on the voters least likely to make their decisions in November based on electability. These happen to be the same voters who hold the balance of power in most elections: independents, conservative Democrats, and moderate Republicans. They aren't principally trying to figure out which Democratic candidate can beat Bush, because they don't necessarily want the Democratic nominee to beat Bush. They're trying to decide which Democratic candidate, if any, would be a better president than Bush.
How well has Kerry done among these voters? In absolute terms, well enough. But in relative terms, the numbers show a disconcerting pattern. By and large, the closer you move to the center and center-right of the electorate, where the presidential race will probably be decided, the worse Kerry does. The opposite is true of Edwards.
In Missouri, Kerry's vote share was 19 points lower among independents than among Democrats, and another seven points lower among Republicans than among independents. Edwards' trend moved in the other direction: He scored five points higher among independents than among Democrats, and another nine points higher among Republicans than among independents. Kerry performed about as well among moderates as he did among liberals, evidently because Dean took a solid chunk of the liberal vote. But Kerry's share of the conservative vote was 10 points lower than his share of liberals or moderates. Edwards, meanwhile, came in four points higher among moderates than among liberals, and another two points higher among conservatives than among moderates.
In Oklahoma, Kerry's vote share was 11 points lower among independents than among Democrats, and another 11 points lower among Republicans than among independents. (Republicans were self-identified, not registered.) Clark followed the same pattern, scoring five points lower among independents than among Democrats, and another 14 points lower among Republicans than among independents. Edwards, on the other hand, scored six points higher among independents, and two points higher among Republicans, than among Democrats. Kerry came in seven points lower among moderates than among liberals, and another eight points lower among conservatives than among moderates. Clark's trend was similar: His vote share was one point lower among moderates than among liberals, and another eight points lower among conservatives than among moderates. But Edwards' trend went the other way: He scored seven points higher among moderates than among liberals, and another three points higher among conservatives than among moderates.
In South Carolina, Kerry's vote share was 11 points lower among independents than among Democrats, and another six points lower among conservatives than among moderates. Again, Edwards reversed the curve: He scored five points higher among independents than among Democrats, and another six points higher among Republicans than among independents. Kerry came in two points lower among moderates and conservatives than among liberals, while Edwards scored seven points higher among moderates, and four points higher among conservatives, than he did among liberals.
In Arizona, Delaware, and New Hampshire, the pattern was a bit different. Joe Lieberman campaigned hard in these states, taking a significant number of the moderate and conservative voters who, in other states uncontested by Lieberman, went to Edwards or Clark. Moreover, in New Hampshire and Delaware, Dean took a sizeable chunk of the liberal vote from Kerry. Still, Kerry performed slightly worse among conservatives than among moderates and liberals. In New Hampshire, he came in four points lower among independents than among Democrats, and another eight points lower among Republicans than among independents. In Arizona, he came in 11 points lower among independents than among Democrats. He did five points worse among moderates than among liberals, and six points worse among conservatives than among moderates. In Delaware, he came in 10 points lower among independents than among Democrats. Clark, Edwards, and Lieberman went the other way, scoring higher among independents than among Democrats. (Self-identified Republicans were too few to cross-tabulate in Arizona and Delaware, because both states closed their primaries to registered Republicans.)
The Michigan exit poll was somewhat unique, since the event was a caucus and the sample was confined to absentee voters. Nevertheless, Kerry's numbers ran in the same direction, putting him 11 points lower among independents than among Democrats. There were too few Republicans to cross-tabulate. Kerry performed somewhat better among moderates than among liberals, as did Clark and Edwards, evidently because Dean took much of the liberal vote.
Tuesday, the pattern was particularly stark. In Tennessee, Kerry's vote share fell from 48 percent of liberals to 39 percent of moderates to 32 percent of conservatives. Edwards went the other way, attracting 26 percent of liberals, 32 percent of moderates, and 35 percent of conservatives. In Virginia, Kerry's trend was less clear—he did slightly better among moderates than among liberals before plummeting among conservatives—but Edwards' trend was the same, ascending two points from liberals to moderates and another 11 points from moderates to conservatives. While Kerry fell from 59 percent of Democrats to 41 percent of independents to 13 percent of Republicans, Edwards rose from 21 percent of Democrats to 31 percent of independents to 45 percent of Republicans.
If I were a Kerry believer, I'd make three arguments against this analysis. The first is that Kerry's higher score among liberals shows strength on the left rather than weakness in the center. But unless you think liberals wouldn't vote for Edwards against Bush, it's logical to assume that Edwards, as the nominee, would end up matching Kerry's strength on the left. Building support in the center is a lot harder.
The second objection is that in addition to the issues-versus-electability question, the exit polls asked voters a question that included other factors, such as "has the right experience" and "cares about people like me." On "has the right experience," Kerry routinely whipped the field, and deservedly so, given his military service and his expertise in national security and foreign policy. But on "cares about people like me," Edwards did, on average, slightly better than Kerry.
The third objection is that the simplest way to measure electability is a national head-to-head poll, and by this standard, Kerry does a bit better than Edwards. The problem with this method is that most voters haven't seen the candidates in their states and haven't been obliged to focus on the election. Only those in early primary states have. So, while Kerry, having received favorable nationwide press coverage for his primary victories, scores well on the "if the election were held today" question, the underlying data are often less auspicious. A week ago, for example, a CNN/USA Today/Gallup Poll found that Kerry would beat Bush by seven points, while Edwards would beat Bush by just one. (A follow-up poll this week shows Bush beating Kerry by one point and beating Edwards by four.) But among Republicans and Republican leaners, Kerry's image was on balance unfavorable, while Edwards' image was on balance favorable.
Could I be wrong about all this? Sure. We pundits have been wrong before. Punditry is a dangerous game. But according to the exit polls, that's exactly the game Democratic voters have played in nominating Kerry. And if they're as shaky at it as we are, the price isn't just embarrassment. It's defeat.
After Fort Hood, There's No Excuse for the Ban on Women in Combat
What Does "Stable Condition" Mean? Absolutely Nothing.
Jim Carrey's Admirably Restrained Scrooge
The Great New Single That's a Little Bit Whitney Houston and a Little Bit Rusted Root
Joe Biden Explains His Gaffe-Evasion Strategy
The Box: A Creepy, Confusing Thriller From the Guy Who Brought You Donnie Darko
Remarks from the Fray:
Just because 'electability' was a top-rated reason for voting for Kerry -- and in fact precisely because of what underlies his very electability -- you leap to a premise-basing conclusion before you even get started that therefore voters don't "agree with him on the issues."
Let me say that again: Among voters who picked the candidate they wanted based on the issues, not the candidate they thought somebody else wanted, Kerry did not win the New Hampshire primary.
…The only way to interpret the poll in question that is remotely a) common sense, b) not full of sensationalizing projections, and c) sticks to the facts, is to note that among voters polled who said the issues was the MOST important thing to them (not the ONLY thing, which what you have managed to twist the poll into meaning, to suit your own substandard-journalism purposes), that means they were saying merely that the best thing their candidate had going for them was issue positions they liked. For voters who chose electability as their MOST important reason for voting, it merely means that the thing they liked best of all is that Kerry could beat Bush. It says NOTHING -- and in fact implies the opposite -- of what you choose to daftly intuit.
What kind of idiot do you assume cares ONLY about electability REGARDLESS of positions on the issues? Of COURSE, these voters ALSO like and support Kerry's positions on the issues, but they chose to say that what was decisive for them in choosing Kerry (among a field of candidates among which they probably agreed with almost all on the issues more or less) was that he could win. It was a given for them that Kerry's positions on the issues also appealed to them.
--zinya
(To reply, click here)
The fatal flaw of Saletan's analysis is his use of descriptive polls for predictive purposes. Exit polls capturing the reason an activity was just undertaken have very little bearing on what a person plans to do in the future (and, as some else noted - the question included the relative language 'more' which more than somewhat negates the differentiation value Saletan relies upon so heavily).
The one predictive poll update that shows Bush over Kerry and Edwards by 1 and 4 points respectively is too far out. The polls in the top half of the article are about history. Between the 2, we're left knowing a little about the past and less about the future.
--HillDweller-2
(To reply, click here)
So the most important factor that has been shown in the exit polls has been the perceived ability to beat Bush in November. So John Kerry ranks higher on that score, but less so for those most concerned about specific issues.
So what?
First of all, there is very little difference between the remaining major candidates on any major issue. The candidates all oppose Bush's radical foreign policy that led to the Unnecessary War. They all support expanding health care coverage to as many as possible or all citizens. They all are progressive on the environment. To the extent they disagree, it is only a matter of degree and style. And the campaign process itself -- by the introduction and initial success of the positively agressive Howard Dean -- has strengthened Kerry's style quite a bit.
Secondly, all the defeated candidates -- including and most importantly Dean -- will fall in behind Kerry in absolute unity. Kerry will promise them input into his fall campaign message and will bring some of them into his administration (most obviously: Edwards as VP, Clark at State or Defense). The party will not only be united as to its goal of defeating the radical Bush, but also will be united on policy and message.
Those like Saletan who insist on looking for or creating trouble within the party will find it -- after all, we are Democrats. Al Sharpton, for one, even with little or no constituency, will try to cause all kind of trouble -- that's what his campaign is really all about. But the party is as united in purpose as it's ever been.
Bush will hold us together.
--PBoySr
(To reply, click here)
(2/11)