Share “Who is voting and why from NC exit polls”

Who is voting and why from NC exit polls

Associated Press Modified: November 7, 2012 at 12:00 am •  Published: November 6, 2012


About 1 in 6 Obama voters split their ticket and cast ballots for McCrory.


Just one-third of voters said they supported same-sex marriage, similar to support for an amendment voted on last May that defined marriage as only between a woman and a man. Obama has thrown his support behind gay marriage since the 2008 election.


About 3 of 5 voters want some or all of Obama's health care law repealed. Seven of 10 voters who want to see some of the law repealed voted for Romney. Two in 10 voters said health care was their most important issue, and three-quarters of them voted for Obama.


Slightly more than half of all voters said that government is doing too many things better left to individuals and businesses. Romney was picked by about three-quarters of those voters.


About 3 in 4 voters said they decided on a presidential candidate before September, when the Democratic National Convention was held to Charlotte. About 3 in 10 voters said they had been contacted on behalf of Obama during the campaign, while slightly fewer than that share had been contacted by Romney's campaign.


A slim majority of voters were born in North Carolina.


About 1 in 5 voters said they go to church more than once a week. About 3 of 5 of them voted for Romney. About 1 in 8 voters said they never attended church and Obama picked up two-thirds of their votes.


McCrory got his biggest boost from the 6 in 10 people who said the economy was their biggest issue. He won 3 out of 5 of their votes. Most of the 1 in 5 people who said health care was the most important issue voted for Democrat Walter Dalton.


The survey of 4,341 North Carolina voters was conducted for AP and the television networks by Edison Research. This includes results from interviews with voters as they left a random sample of 50 precincts statewide Tuesday, as well as 926 who voted early or absentee and were interviewed by landline or cellular telephone from Oct. 26 - Nov. 3. Results for the full sample were subject to sampling error of plus or minus 3 percentage points; it is higher for subgroups.