Why Virginia Loves George
Using exit poll data, Monika McDermott of the CBS News Election and Survey Unit analyzes George W. Bush's victory.
Texas Gov. George W. Bush swept to victory over Arizona Sen. John McCain in the Virginia GOP primary thanks to a solidly Republican and conservative electorate.
Virginia Is Bush Countrysize>color>
Virginia was friendly territory for Bush: 63 percent of primary voters were Republicans, and 55 percent called themselves conservative. In contrast, Michigan's electorate was only 48 percent Republican and 45 percent conservative.
Virginia's Republicans voted for Bush over McCain by 69 percent to 28 percent. Conservatives favored Bush by 42 points - 69 percent to 27 percent for McCain.
Voters who align themselves with the Christian right made up only 19 percent of the Virginia primary electorate, much less than the 34 percent in South Carolina earlier this month, and even less than Michigan's 27 percent. But these voters went overwhelmingly for Bush.
Bush's Strengthssize>color>
Since his image makeover in South Carolina, Bush has made headway as the "reformer with results." Fifty-one percent of Virginia's primary voters said that Bush was a real reformer, and 48 percent said McCain was. Bush has gained ground on this issue, even among his own supporters - in Michigan 61 percent of Bush voters said their candidate was a reformer, on Tuesday 74 percent of his supporters characterized him that way.
Bush also tied McCain for the straight-talk title. Fifty percent of voters said Bush was more likely to say what he believes rather than what people want to hear. Fifty-one percent said the same thing about McCain.
McCain's Weaknessessize>color>
McCain's image has worsened among Republicans in the past few weeks, making his task of attracting Republican voters even more difficult. In South Carolina ten days ago, 62 percent of Republicans viewed McCain favorably. But in Virginia, only 50 percent of Republicans did.
Republicans do not think McCain is conservative enough for them. Overall, 49 percent of voters thought McCain's ideology was about right. But among Republicans, only 41 percent thought McCain's views were about right, and 44 percent thought he was not conservative enough.
Another part of McCain's problem is the growing feeling that he has been attacking Bush unfairly. Forty-nine percent of Virginia primary voters said McCain had attacked Bush unfairly. In contrast, only 43 percent of South Carolina voters said that. Only one third of Virginia's primary voters thought that Bush had attacked McCain unfairly.
The McCain Majoritysize>color>
While some have argued that McCain's bipartisan appeal is transitory, and that Democratic, and perhapeven independent, voters would abandon him in the general election, the Virginia numbers say otherwise. Eighty-five percent of McCain voters said they would vote for McCain over Gore in the general election, given that option. Even 85 percent of independents and 53 percent of Democrats who voted for McCain said they would choose McCain over Gore in the general election.
Bush lacks this crossover appeal: only 48 percent of McCain primary voters would support Bush in November. But 70 percent of Bush's supporters would vote for McCain in a general election match-up with Gore.
Nonetheless, voters continue to believe that Bush is the stronger general election candidate - 55 percent said Bush would have a better chance of beating the Democratic nominee in November.
Voters and Issuessize>color>
Bush won among most demographic groups, performing particularly well among women voters. Among women, Bush beat McCain 57 percent to 41 percent. McCain carried Democrats and independents, as well as voters with the highest education levels and incomes.
Virginia's primary voters, like voters in other Republican primaries, chose moral values as their top issue. Thirty-three percent of voters chose moral values. Taxes came second with 15 percent. Voters concerned with both of these issues overwhelmingly supported Bush.
The CBS News exit poll was conducted among 1,832 primary voters as they left the polls February 29, 2000. The margin of sampling error could be plus or minus two percentage points for results based on the entire sample. The sampling error for subgroups is higher.