Non-believers, Seculars, the Un-churched and the Unaffiliated: Who are Non-religious Americans and How do we Measure them in Survey Research

This paper was presented at the annual conference of the American Association of Public Opinion Research in May 2009.

The bulk of research conducted by public opinion scholars on the subject of religion and politics has consistently focused on the largest religious groups: Catholics, white evangelical Protestants, black Protestants and white mainline Protestants. These groups are significant to American politics and American culture not only because of their size but also because of their organizational capacities and theological distinctiveness. However, over the last 20 years another important group has emerged to play a more prominent role in American politics: Americans with no formal religious affiliation.

Americans with no formal religious affiliation now represent 16% of all American adults. They are increasing in size and as a share of the electorate. Over the past 4 presidential elections this group has nearly doubled in size as a proportion of all voters. Furthermore, this group has steadily grown more liberal and Democratic in their politics. In 2008, Americans with no religion supported Barack Obama over John McCain by a stunning 52-point margin (75% to 23%).

Yet despite the increasing size and electoral clout of the unaffiliated, less attention has been paid to investigating who they are and what they care about. In fact, there is significant uncertainty about just how to define this group. For instance, it is not uncommon for people to use the terms like non-religious, atheist, agnostic and secular interchangeably, despite the fact that there are significant differences between and among these groups.

This paper analyzes the ways in which public opinion research has attempted to operationalize the unaffiliated and compare the effects of different definitions on both estimates of group size and composition. Using the landmark 2007 Religious Landscape Survey that has a sample size of over 35,000 respondents, we examine the religious and political diversity among the unaffiliated. Then, using the General Social Surveys we analyze how this group has grown over the last 20 years and examine how the political and demographic composition of the group has shifted as it has grown.