Religious Whites Still Tilt Toward GOP, Bush

PRINCETON, NJ -- An extensive Gallup analysis of more than 90,000 interviews conducted since 2004 shows that highly religious Americans are not drifting away from the Republican Party any more than are Americans generally, or nonreligious Americans specifically.

All Americans have become less likely over the last three years to identify with the Republican Party, and less likely to approve of the job President Bush is doing. Highly religious Americans have followed this same pattern, but no more -- and no less -- than anyone else. This is particularly true among highly religious white Americans, who have constituted a core base for the Republican Party. Thus, the gap between religious white Americans and whites who are not religious in terms of GOP identification and Bush approval is just as large today as it was in 2004 and 2005. Highly religious white Americans remain one of the strongest pockets of support for the Republican Party in the United States.

Background

A recent article in The New York Times Magazine by David Kirkpatrick (entitled “The Evangelical Crackup”) advanced the hypothesis that “the extraordinary evangelical love affair with Bush has ended, for many.” Kirkpatrick built most of his article around interviews with conservative Christian pastors and religious leaders, but at one point he bolsters his argument using poll data:

Today the president’s support among evangelicals, still among his most loyal constituents, has crumbled. Once close to 90 percent, the president’s approval rating among white evangelicals has fallen to a recent low below 45 percent, according to polls by the Pew Research Center.… And the dissatisfaction extends beyond Bush. For the first time in many years, white evangelical identification with the Republican Party has dipped below 50 percent.

But to what degree has the love affair between highly religious white Americans and the Republican Party really ended? Is the president’s support crumbling among evangelicals? And does it appear that a degree or type of disaffection is occurring among religious Americans that is distinct from what is occurring among the broader, national population?

To help answer that question, Gallup has assembled a dataset of more than 90,000 interviews from approximately 90 national surveys conducted between January 2004 and October 2007. Each survey included questions asking respondents about their church attendance, their political identification, and their approval of President Bush. (The analysis that follows focuses exclusively on white respondents, given the near-monolithic identification with the Democratic Party and disapproval of President Bush among blacks.)

The analysis underscores first and foremost the degree to which both the Republican Party and President Bush have slipped significantly in their standing among all white Americans between 2004 and 2007.

The percentage of white Americans identifying themselves as Republicans has fallen by 7 points from 2004 to 2007, from 39% to 32%.

The percentage of white Americans approving of the job George W. Bush is doing as president of the United States has fallen by 19 points, from 56% in 2004 to 37% for the first 10 months of 2007.

These conclusions are based on a comparison of more than 25,000 interviews with white Americans that Gallup conducted in 2004, and more than 18,000 interviews with white Americans conducted between January and October 2007.

The fact that the type of poll data cited in the New York Times Magazine article may show Republican identification and Bush approval dropping among evangelicals is not in and of itself surprising -- because Republican identification and Bush approval have been falling in general. The more important issue is the degree of this change relative to other groups within the U.S. population. The question becomes: Has the Republican Party suffered disproportionate losses among religious whites, or do the changes among this group appear to reflect broad, general trends?

Gallup ’s 2004 to 2007 Data on the Relationship of Religiosity to GOP and Bush Support

For the purposes of this analysis, church attendance is being used as the measure of religiosity.

(Church attendance is highly correlated with other measures of religion, including the self-reported importance of religion in one’s daily life.) Based on the 2007 Gallup Poll data, 29% of white Americans go to church weekly. Twenty-three percent of whites say they attend church almost every week or monthly and 47% say they attend seldom or never.

The tables presented at the end of this article show on a year-by-year basis the relationship between religiosity and party identification, and religiosity and Bush job approval, among white Americans.

The key relationships are summarized in the accompanying graphs.

The data show that there has been no diminution in the relative advantage that Bush and the GOP enjoy among highly religious whites over less religious whites in 2007, compared to what it was in 2004.

The overall “tide” has lowered over these three years, so that now each group defined by church attendance has lower GOP identification and lower Bush approval than it did in 2004. But the relative gap between religious groups has remained constant.

The accompanying table summarizes the relative change in identification with the Republican Party and Bush job approval between 2004 and 2007, by religiosity.

It appears likely that highly religious white Americans have been affected by the same issues, events, and trends that have weakened the position of the GOP and President Bush across all of American society. Identification with the Republican Party and support for Bush have fallen among religious whites, but on a proportionate basis, and Republicans continue to enjoy a significant advantage among religious whites compared with less religious whites -- as they have in the past.

The gap in Republican identification between highly religious whites and nonreligious whites was 22 points in 2004, and is exactly the same -- 22 points -- in 2007.

The gap in Bush job approval between highly religious whites and nonreligious whites was 22 points in 2004 and is 23 points in 2007.

It is true that the Republican leanings of highly religious whites did drop. They were not uniquely immune to the pressures that have negatively affected the Republican Party since 2004. But religious whites did not succumb at a disproportionately greater rate than other white Americans. Thus, to this day they continue to be one of the subgroups in the American population most likely to identify with the Republican Party and most likely to approve of the job Bush is doing as president.

It appears less that there has been a unique “crackup” within the highly religious segment of the U.S. population, and more that this group has drifted away from the GOP and Bush to the same degree as everyone else.

Survey Methods

The results in this analysis are based on an aggregated sample of 109,859 telephone interviews conducted across Gallup surveys with randomly selected national samples of adults 18 and over between Jan. 1, 2004, and Oct. 14, 2007.

For results based on 25,342 white Americans conducted in 2004, the maximum margin of sampling error is ±1 percentage point.

For results based on 28,399 white Americans conducted in 2005, the maximum margin of sampling error is ±1 percentage point.

For results based on 22,827 white Americans conducted in 2006, the maximum margin of sampling error is ±1 percentage point.

For results based on 18,294 white Americans conducted in 2007, the maximum margin of sampling error is ±1 percentage point.

For results for smaller subgroups based on church attendance, the maximum margin of error is between ±1 and ±2 percentage points, depending on exact sample size.

In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.