Poll: Electability Boosts Kerry

Feb. 4, 2004 -- Electability boosted John Kerry in three of five states where exit polls were conducted Tuesday, but he had less strength among issue voters — and he didn't get much in the way of Southern comfort.

The attribute that did the trick for Kerry in Iowa and New Hampshire — a perceived ability to beat President Bush — scored again in Missouri, Delaware and Arizona, giving him far-flung bragging rights.

But electability worked less well for Kerry in South Carolina and Oklahoma, opening the door for others — and raising questions about his prospects in two more Southern states, Virginia and Tennessee, that vote next week.

Here's a summary of state-by-state exit poll results:

ARIZONA

Electability was the critical difference for Kerry in Arizona. Among the 28 percent of voters looking mainly for a candidate who can beat Bush — the top attribute — Kerry won a crushing two-thirds of the vote. He also won among those looking mainly for a candidate with the "right experience," winning 51 percent of them.

In another electability measure, 58 percent of Arizona voters said they cared more about a candidate they agree with on the issues than one who can beat Bush (34 percent). But while Kerry ran about evenly with Clark among the issue voters (32 to 30 percent), he dominated among the "beat Bush" voters, with 59 percent, beating Clark in this group by nearly a 3-1 margin. Clark did better than Kerry (36 to 30 percent) with voters looking for a candidate who "cares about people like me." And Dean, Kerry and Clark split the vote among those looking for a candidate who "stands up for what he believes."

As elsewhere, Arizona voters expressed broad discontent with the economy (79 percent rated it negatively), with Bush's administration (46 percent even "angry" about it) and with the war in Iraq (70 percent disapprove). The economy topped their issues list (31 percent), followed by health care (23 percent) and the war (19 percent). Kerry did particularly well among those citing health care as their main issue (with 50 percent of their votes) and the economy (44 percent). Clark snatched some votes from Kerry among those citing the war as their top issue — Kerry won 38 percent of them, Clark 34 percent.

Strong across the board, Kerry did better in Arizona with seniors (48 percent), white Catholics (49 percent), and with voters who decided last week (56 percent) — just after Kerry's victory in New Hampshire. He did less well with conservatives (34 percent) than with liberals (45 percent) or moderates (40 percent), but still won them, by 10 points over Clark.

DELAWARE

Strong essentially everywhere in Delaware, Kerry did especially well among senior citizens — nearly one in four voters, he won 63 percent of them; the least educated (54 percent); union members (54 percent); Catholics (59 percent); voters who made up their minds last week or last month (winning about six in 10 in each group); voters who cared more about a candidate's ability to beat Bush than his position on issues (71 percent); health care voters (54 percent); Iraq voters (55 percent); and those who picked beating Bush or having the "right experience" as the top candidate attribute (71 and 64 percent, respectively).

Dean ran closer to Kerry among voters who wanted a candidate who stands up for what he believes (24 percent for Dean, 32 percent for Kerry). But Kerry's support was much more broadly based; he also won four other attribute groups — those looking for someone who cares about them, has the right experience, can beat Bush and has a positive message. Edwards did well with this last group, too, but still trailed Kerry among those voters by 12 points.

Kerry also won more than four in 10 blacks (who accounted for about one in six voters in Delaware). Sharpton was second in this group, winning three in 10 black voters.

MISSOURI Kerry dominated in Missouri, with his support peaking among those who placed top importance on beating Bush (78 percent) and having a candidate with the right experience (67 percent). He won roughly equal numbers of voters who picked the economy and health care as their top issue (56 and 57 percent respectively) — the top two issues. He also won 52 percent of voters from union households and 54 percent of military veterans. Kerry did less well with conservatives (41 percent) than with liberals (51 percent) and moderates (53 percent). He did better with high school graduates (58 percent) than with college graduates (48 percent), and particularly well among seniors (59 percent); they accounted for a quarter of Missouri voters.

Edwards won voters seeking a candidate with a positive message (50 percent), and ran evenly with Kerry among those looking for a candidate who "cares about people like me." Edwards won 31 percent of voters who called education their top issue, but Kerry won 39 percent of this comparatively small group.

OKLAHOMAThe strengths of individual candidates played a much larger role in Oklahoma than did anger with the Bush administration and electability — two hot-button issues in other states. Edwards won 42 percent of voters who picked "cares about people like me" as the top candidate attribute, while Clark won 34 percent of those who picked "stands up for what he believes." (These two tied as top candidate attributes.)

The next two attributes — "can defeat George W. Bush" and has a "positive message" — went to Kerry and Edwards, respectively. Kerry won more than half of those who picked beating Bush; Edwards won 45 percent of those who picked "positive message," a quality that served him well in South Carolina.

Fewer than one in 10 voters chose "right experience" as the top attribute; while strong for Kerry in other states, Clark handily won "experience" voters here, with almost half their votes.

On issues, nearly four in 10 voters who cited the economy as their top issue chose Edwards. Clark had a narrow advantage on the second-tier issue of health care, a bigger edge on education, and ran about evenly with Kerry among voters most concerned about the war. While most voters disapproved of the war in Iraq (by 55-39 percent), that was milder disapproval than in other states.

Edwards had an advantage among conservatives; he did less well among liberals. Blacks (a small group here, eight percent of voters) divided about evenly between Clark and Kerry, with Edwards a weak third.

Clark, Edwards and Kerry were about even among military veterans (nearly three in 10 voters) and voters from union households (one in five).

SOUTH CAROLINAThe economy topped the issues list in South Carolina (cited by 47 percent), and Edwards did especially well among those voters, beating Kerry by 2-1. It was a closer contest with Kerry among health care voters, and Kerry did better with voters who cared most about the war in Iraq. (Education, a much lower-tier issue, also worked well for Edwards.)

South Carolina voters by almost a 2-1 margin said they cared more about a candidate who agrees with them on the issues than one who can beat Bush. Among issue voters (59 percent), Edwards won by more than 2-1; among "beat-Bush" voters, though, it was a dead heat between Edwards and Kerry.

Edwards did particularly well on his trademark "nice guy" image, and benefited from his native-son advantage. He won by huge margins among people looking for a candidate who "cares about people like me" (with 57 percent) or who "has a positive message" (58 percent) and especially among the one in 10 looking for a candidate who "understands South Carolina" (not an especially large group, but Edwards won 78 percent of them). Kerry, by contrast, did best on the qualities that were his best in general — electability ("he can defeat George W. Bush" — 58 percent for Kerry) and experience (63 percent).

Agreeing on Issues or Beating Bush

In South Carolina, native son John Edwards beat Kerry by 26 points among voters who cared more about a candidate who agrees with them on the issues than one who can beat Bush.

And in Oklahoma, Edwards and Wesley Clark tied among issue voters, with 32 percent apiece, while Kerry trailed with 19 percent.

In all five states, given these two choices — agreeing with a candidate on the issues, or beating Bush — majorities said agreeing on the issues was more important to them.

Even among those who cared more about beating Bush, Kerry was softer in the South. He tied Edwards among "beat Bush" voters in South Carolina; in Oklahoma he won them, but by far less of a margin than in Arizona, Delaware and Missouri.

Electability, while important, wasn't the only game in town. In another gauge, when voters were asked what single candidate quality was most important to them, the ability to beat Bush was first (by just five points) in Arizona and Delaware, tied for first (with "stands up for what he believes") in Missouri, but tied for second in South Carolina and ended in a three-way tie in Oklahoma, where most everything was in a three-way tie in a very close Clark-Edwards-Kerry contest.

Kerry nonetheless had a strong night, with some good results even in states he lost. In South Carolina, for example, he ran roughly even with Edwards among blacks, a core Democratic group that essentially hadn't been heard from before these contests. Blacks accounted for nearly half the voters in South Carolina; Edwards won 37 percent of them, while Kerry was right there with 34 percent (Al Sharpton had 17 percent). Among South Carolina whites, by contrast, it was Edwards by 2-1 over Kerry.

Kerry Scores With Core Democratic Voters

Indeed Edwards showed some vulnerability among minorities in three other states: He did less well with blacks than with whites in Missouri, Delaware and Oklahoma alike. (He also did perhaps slightly less well with Hispanics than with whites in Arizona.)

Kerry did his best with core Democratic voters as opposed to independents.

In South Carolina he lost Democrats by 10 points but independents by 26; in Oklahoma he did 11 points better with Democrats than with independents. It was similar in states where he cruised: Kerry did 19 points better with Democrats than with independents in Missouri, 10 in Delaware, 11 in Arizona.

Kerry also did less well with conservatives than with liberals in some states — Arizona, Missouri and particularly in Oklahoma, where he won a third of liberals, a quarter of moderates, but just 18 percent of conservatives. Conservatives accounted for 21 percent of Oklahoma voters, more than in any other state Tuesday night. And there may be a good number of conservatives out and about next week: In the 2000 Democratic primary in Tennessee, they accounted for 24 percent of voters.

Edwards Plays the Nice Card

Kerry did better in general with disgruntled voters — those angry at the Bush administration, who strongly disapprove of the war in Iraq, who're especially unhappy with the economy — while he tended to do considerably less well among gruntled ones.

For example in Oklahoma, Kerry's weakest state, with a third-place finish, he won 35 percent of angry voters, but less than half as many, 15 percent, of those satisfied with the Bush administration or even enthusiastic about it — more than a fifth of voters there.

Edwards continued to play the nice card, doing well even outside the South (in Missouri) among voters looking mainly for a candidate who has a "positive message" and who "cares about people like me." As noted, he did well with issue voters (as opposed to "beat Bush" voters) in South Carolina and Oklahoma.

Clark, too, made inroads among issue voters, running nearly even with Kerry in this group in Arizona (where he finished second), as well as even with Edwards among issue voters in Oklahoma.

There was little for Howard Dean in any of these states, and less for Joe Lieberman, who read the writing on the wall and headed home.

Methodology

This analysis is based on exit polls of 2,093 Democratic primary voters in Arizona, 1,197 in Delaware, 1,400 in Missouri, 1,501 in Oklahoma and 2,033 in South Carolina. The results have a margin of sampling error of plus or minus four percentage points in each state. Field work by Edison/Mitofsky for the National Election Pool.

Previous ABCNEWS polls can be found in our Poll Vault.