Americans with no religious attachment are much less likely than religious Americans to identify themselves as "pro-life" on abortion, according to a new Gallup poll.
Only 19 percent of those with no religious attachment identified themselves as "pro-life" in the May 3-6 poll of 1,024 adults. Sixty-eight percent called themselves "pro-choice." This represents the largest propensity toward the pro-choice position on abortion of any major demographic group.
The 49 percentage point margin between pro-life and pro-choice positions among the nonreligious is much larger the 16 percentage point margin among Catholics (54 percent pro-life, 38 percent pro-choice) and the 24 percentage point margin among Protestants and other Christians (57 percent pro-life, 33 percent pro-choice).
The nonreligous category included those who identified themselves as atheist or agnostic and those who said they have no religious preference.
Besides self-proclaimed Democrats and the nonreligious, the only other major subgroup to show a majority identifying as pro-choice are those with a postgraduate education. Fifty-eight percent of those with a postgraduate education identify as pro-choice while 37 percent call themselves pro-life.
Among college graduates and high school graduates, 53 percent say they are pro-life. Those who have had some college but no degree are about evenly split. Forty-nine percent are pro-life and 41 percent are pro-choice.
Gallup also found that men are more pro-life than women, nonwhites are more pro-life than whites, those 55 and older are more pro-life than 18- to 54-year-olds, Southerners are more pro-life than non-Southerners, and those who make less than $75,000 per year are more pro-life than upper-income Americans.
Gallup observed that there has been a shift from 2001 to 2008. Groups that used to be mostly pro-choice -- women, young adults, middle-aged adults and college graduates -- are now evenly divided on the issue. And some who used to be evenly divided -- men and those 55 and older -- now clearly tilt toward the pro-life position.
The results are based upon the same survey that recently reported the number of pro-choice identifiers are at a record low.
The margin of error for the full sample is plus or minus four percentage points.