Sharing the Faith

By Kenneth R. Blake, Robert O. Wyatt, and Holly Warf

Since the 2004 election, the notion that evangelical Christians helped propel incumbent Republican George W. Bush to a win over Democratic challenger John Kerry in the presidential race has come to be regarded as gospel. The aid Bush received from evangelical "'values' voters" may be the latest example of the increasingly close ties between committed, white evangelical Protestants and the Republican Party. Still very much an open question, though, is what might be driving evangelicals' pro-Republican voting choices, and part of the problem in answering it is the methodological issue of how to measure evangelicalism in the context of a public opinion survey.

There are reasons to suspect that increasingly politically active evangelicals of the late twentieth and early twenty-first centuries may be developing into voters who possess the sort of ideologically constrained opinions--that is, opinions that relate both statistically and ideologically with one another--that Philip Converse in 1964 famously found so rare in the electorate. As Heather Hendershot observed in her recent book, Shaking the World for Jesus, evangelicals consume a wide array of evangelically produced media that parallel secular media and offer an evangelically flavored alternative on just about every conceivable topic, including politics. Popular evangelical leaders, including Moral Majority lobby-group founder Jerry Falwell, televangelist and one-time presidential candidate Pat Robertson, and James Dobson, psychologist and outspoken head of Focus on the Family, all comment routinely and publicly on political matters. And every Sunday, millions of evangelicals flock to churches where, during this past election season, political perspectives were shared and reinforced in conversations with fellow worshipers, printed voter guides, and more than a few sermons. Such activities have at least some potential to school evangelical congregants in an evangelically inspired political ideology.

And yet there is a chance that, despite such potential, evangelicals may end up just as reliant as everyone else on what Samuel Popkin termed "'gut' reasoning" when deciding which candidate to support in an election. Popkin characterized the task of gathering hard political information as both costly in terms of time and energy and unlikely to yield a tangible, short-term benefit. Voters, Popkin said, respond to the problem practically by relying on various shortcuts and proxies. They may rely on estimates of a given candidate's personal integrity or competence, for example.

Ultimately, Popkin asserted, voters end up making adequate political choices even though they sidestep a lot of the work. The many political messages evangelicals receive through their media, hear expounded from their pulpits, and discuss with friends while loitering in church hallways and foyers after Sunday services may, more often than not, resemble the sort of information on which Popkin-style shortcuts and proxies are based. Finally, it is also possible that evangelicals make voting decisions based mostly or exclusively on a handful of "hot button" issues they consider particularly important, like abortion or gay marriage.

Before investigating which of these views provides a better context for understanding voting behavior among evangelicals, however, one must first decide how to spot an evangelical. It is tempting to oversimplify. Researchers have taken essentially two approaches to the problem, one involving measures of religious beliefs and the other involving measures of religious behavior. Belief-oriented measures ask questions like, "Do you believe there are miracles?"; "Do you regard your problems as a consequence of personal sin?"; and "Do you believe there is life after death?" Behavior-oriented measures, meanwhile, cover topics such as witnessing the love of Christ, prayer and meditation, and church involvement, as well as association with other Christians.

One problem with these approaches is that they assume far more uniformity in religious belief and practice among evangelicals than actually exists. Given the diversity in belief and practice that evangelicals exhibit, it may be more economical simply to let people decide for themselves whether the term "evangelical," as they understand it, accurately describes their religious views, then try to discern the kinds of evangelicals who end up in such a self-selected group. Such was the approach taken in this study, in which we sought to answer two questions: Could voting preferences in the 2004 presidential election be significantly predicted by a single measure of self-identification as an evangelical or born-again Christian? And did voting preferences among such self-described evangelicals seem based on ideological constraint, "gut" rationality, and/or specific issues?