Why South
Carolina's Democratic presidential primary vote turned out the
way it did:
HOME GROWN: John Edwards appealed to voters who felt he was
like them. Six of 10 voters said they chose a candidate who
agrees with them on major issues. Edwards picked up nearly
half of those votes, twice what John Kerry won.
POCKETBOOK ISSUES: Nearly half the voters said they chose
based on jobs and economic concerns. Edwards picked up half
that vote - twice as many as Kerry. One in five said Medicare
and health care were their top concern. Edwards picked up four
of 10 of those votes, double Kerry's.
MODERATES: Four in 10 voters described themselves as
political moderates. Edwards picked up half that vote to three
of 10 for Kerry .
INDEPENDENTS: Almost one-fourth of the voters were
independents. Edwards won half that vote, more than twice as
much as Kerry.
VETERANS: Kerry, a Vietnam veteran, and retired Army Gen.
Wesley Clark were expected to capitalize on the military vote.
Both hit the airwaves with ads showing them in uniform and
with testimonials to their work. On Tuesday, Clark pulled
fewer veteran votes than Al Sharpton. Edwards, with no
military experience, pulled slightly more veterans' votes than
Kerry.
IRAQ: While almost three-fourths said they disapprove of
the war in Iraq, it was a deciding issue for only about one in
10 voters. Among the two of five voters who approved of the
war, Edwards picked up half the vote.
THE INTERNET: Howard Dean's campaign fanned out across the
Internet in the early months and his fund-raising knack
prompted some to say his efforts could mark a turning point in
strategy. Only a quarter of voters said they visited Web sites
at least occasionally or more frequently. Dean pulled less
than one in 10 of those voters, a third of Kerry's and a
quarter of Edwards'.
---
Exit poll conducted for The Associated Press and television
networks by Edison Media Research/Mitofsky International among
1,982 South Carolina Democratic primary voters as they left
precincts Tuesday. The poll has a margin of sampling error of
3 percentage points for the overall sample, larger for
subgroups.
