In U.S., Very Religious Americans Still Align More With GOP In U.S., Very Religious Americans Still Align More With GOP. Nonreligious Americans are much more likely to identify as Democrats by Frank NewportPRINCETON, NJ -- Very religious Americans remain significantly more likely than those less religious to identify as Republicans or lean Republican, and nonreligious Americans are more likely to identify as Democrats or lean Democratic. This strong relationship between religion and party identification has persisted over the past three years and four months, regardless of overall, broad partisan changes. The accompanying graph shows trends in the net Republican orientation of Americans in each of three religious groups, based on the percentage identifying as or leaning Republican minus the percentage identifying as Democrats or leaning Democratic. The three religious groups are defined by a combination of how important respondents say religion is to them and how often they say they attend religious services. The graph covers the period from February 2008 through May of this year, and is based on more than 1 million interviews conducted as part of Gallup Daily tracking. The overall percentage of the American population identifying as or leaning Republican dropped in late 2008 and early 2009, contemporaneous with the November 2008 presidential election of Barack Obama as president and his inauguration in January 2009. Republican affiliation began to pick up again in mid-2009 and has been higher since that point, although it showed a slight drop in May of this year. These broad shifts in party affiliation were evident across all three religious groups. All three became less Republican in late 2008 and all three have become more Republican since. But differences among the partisan orientations of the three groups remained virtually the same. Very religious Americans have remained the most Republican across the last three years and four months, moderately religious Americans less so, and nonreligious Americans remain the least Republican of the three groups. A multivariate analysis across this time period confirms that religiousness has remained significantly predictive of party affiliation in 2008, 2009, 2010, and in January through May of this year, while other variables related to partisanship such as age, gender, race, region, children, and marital status are statistically controlled for. In May, Very Religious Were 19 Percentage Points More Likely to Be Republican The detailed breakdown of approximately 30,000 interviews conducted in May 2011 shows that very religious Americans are 19 percentage points more likely to identify as Republicans than are the nonreligious, 48% vs. 29%. Very religious Americans are 10 points more likely to be Republican than Democratic -- 48% vs. 38%. Among moderately religious Americans, there is a Democratic advantage of 8 points. Among nonreligious Americans, the Democratic advantage increases to 25 points. In these May Gallup tracking data, 41% of Americans were classified as very religious, 29% as moderately religious, and 30% as nonreligious. Implications The influence of personal religiousness on party identification in the U.S. is not a new phenomenon, and has been observed in analyses of survey and voting data going back to the Reagan administration. The current analysis, based on more than a million Gallup interviews conducted between February 2008 and May of this year, provides more nuanced detail on the nature of this relationship in recent years. The analysis of month-by-month patterns reveals that all groups of Americans, regardless of religiousness, can be affected by broad tides of political change. This is seen in the shift to a more Democratic orientation in and around the election of Obama in 2008 and the shift back to a more Republican orientation in the months since. These changes are reflected in the party identification of very religious, moderately religious, and nonreligious Americans. Regardless, the partisan gaps distinguishing these three groups remained virtually constant across time. Very religious Americans remained consistently more Republican, and the nonreligious more Democratic. Statistical analysis confirms that the impact of religion remained constant and powerful across the last three years and four months. Obama managed to win the presidency in 2008 despite his relatively poor showing against John McCain among highly religious Americans. The current data suggest that he still enjoys relatively strong support among Americans who are nonreligious, but that in the coming election he will again face an uphill battle among those who are moderately and, in particular, very religious, given the latter's clear tendency to identify as Republicans. Survey Methods Results are based on telephone interviews conducted as part of Gallup Daily tracking January 2008 through May 2011, with a random sample of approximately 30,000 monthly interviews with adults, aged 18 and older, living in all 50 U.S. states and the District of Columbia. For results based on each month's total sample of national adults, one can say with 95% confidence that the maximum margin of sampling error is ±1 percentage point. Interviews are conducted with respondents on landline telephones and cellular phones, with interviews conducted in Spanish for respondents who are primarily Spanish-speaking. Each sample includes a minimum quota of 400 cell phone respondents and 600 landline respondents per 1,000 national adults, with additional minimum quotas among landline respondents for gender within region. Landline telephone numbers are chosen at random among listed telephone numbers. Cell phone numbers are selected using random-digit-dial methods. Landline respondents are chosen at random within each household on the basis of which member had the most recent birthday. Samples are weighted by gender, age, race, Hispanic ethnicity, education, region, adults in the household, and phone status (cell phone only/landline only/both, cell phone mostly, and having an unlisted landline number). Demographic weighting targets are based on the March 2010 Current Population Survey figures for the aged 18 and older non-institutionalized population living in U.S. telephone households. All reported margins of sampling error include the computed design effects for weighting and sample design. 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.