Research Design

Note: For ARIS Series technical and sampling details including response rates please visit this link.


ARIS 2008 is the third in a landmark time series of large, nationally representative surveys that track changes in the religious loyalties of the U.S. adult population within the 48 contiguous states from 1990 to 2008. The 2001 and 2008 surveys are replicas of the 1990 survey, and are led by the same academic research team using an identical methodology of random-digit-dialed telephone interviews (RDD) and the same unprompted, open-ended key question "What is your religion, if any?" Interviewers did not prompt or offer a suggested list of potential answers. Moreover, the self-description of respondents was not based on whether established religious bodies or institutions considered them to be members. To the contrary, the surveys sought to determine whether the respondents regarded themselves as adherents of a religious community. The surveys tap subjective rather than objective standards of religious identification. The value of this unique series of national surveys, which allows scientific monitoring of change over time, has been recognized by the U.S. Bureau of the Census. The Bureau itself is constitutionally precluded from such an inquiry into religion, and so has incorporated NSRI/ARIS findings into its official publication the Statistical Abstract of the United States since 2003.

The key religion question is part of an inquiry that also probes a range of socio-demographic, political, social, and life-cycle issues as well as attitudes that add richness to the main findings. These responses reveal the nation's pattern of religious beliefs, behaviors and belonging. The ARIS 2008 survey was carried out from February through November 2008 and collected answers from 54,461 respondents who were questioned in English or Spanish. In order to fill the information gap on the growing number of people who do not have a landline but use cellular telephones mainly or exclusively, we supplemented the traditional RDD sample with a separate national cell phone survey. Results for the ARIS key open-ended question on religious self-identification indicate no statistically significant differences between the RDD sample and the cell phone sample.1 ARIS 2001 interviewed 50,281 respondents and the 1990 NSRI interviewed 113,713 respondents. The huge number of cases in these surveys provides unparalleled, in-depth profiles of the social make-up of religious groups and detailed geographical coverage with a high degree of statistical precision and a standard error of under 0.5 percent for the full sample in 2008.

As one might expect with over 220,000 interviews recorded over three surveys, the ARIS respondents offered a vast number of theological, religious and denominational responses to our key question. These open-ended answers have to be aggregated down to a manageable number of categories for analytical purposes. This requires using a simplified aggregation that helps highlight the major trends in religious sentiments across five major theological blocs as utilized in Tables 1, 2 and 12. The category Catholic is comprised of (1) Roman Catholics, (2) Eastern Rites Catholics, and (3) all others who used the term "Catholic" in their response. The "Other Christians" bloc is composed of all non-Catholic respondents who self-identified with a religious group which claims to be Christian as well as any theological term that related to Christianity. The "Other Religions" bloc comprises all the other faiths, world religions and religious groups that are not Christian. The "Nones" are an amalgamation of all the respondents who provided answers to our key question which identified them as having no religious identity or connection. The most common response was "None" or "No Religion." This bloc can be described as the non-religious, irreligious and anti-religious bloc. It includes anti-clerical theists, but the majority are non-theists. For reasons of scientific integrity we have also included data on the "Unknown" category, composed of those who said they did not know the answer to our key religion question and those who refused to reply to our key question. We have no religious identification data on this population but we do have demographic and attitude data.

A further re-classification of the responses that offers a finer-grained taxonomy identifying 12 religious traditions and some of the larger religious groups is provided in Table 3 and other subsequent tables. However, this summary is just the tip of the iceberg of statistical data on a much larger number of religious groups than can be handled here and many more social variables than are highlighted here. The 1990 and 2001 studies were fully analyzed and reported in One Nation under God: Religion in Contemporary American Society (1993) and Religion in a Free Market: Religious and Non-Religious Americans (2006).2


1 Detailed analysis of the religious and socio-demographic profiles of the cell phone users will be provided in a later report.

2 Barry A. Kosmin and Seymour P. Lachman, One Nation under God: Religion in Contemporary American Society, New York, Harmony Press, 1993; Barry A. Kosmin and Ariela Keysar, Religion in a Free Market: Religious and Non-Religious Americans, Ithaca, N.Y., Paramount Market Publishing, 2006.

ARIS 2008