Regarding the question of whether voting preferences among self-described evangelicals seemed based on ideological constraint, "gut" rationality, and/or specific issues, little evidence was found to exempt today's evangelicals from Converse's critique of American voters as generally lacking a constrained political belief system. If anything, evangelicals exhibited significantly less ideological constraint and issue knowledge compared to nonevangelicals, although neither group's average scores turned up all that far from the zero mark.

And while evangelicals did, indeed, differ from nonevangelicals on the issues of restricting abortion and allowing gay marriages or gay civil unions, indicators of the sort of "gut" rationality described by Popkin dominated as predictors of actual voting choice among white evangelicals. The strongest predictor in Figure 2 is perceived candidate intelligence, and the most popular predictor is perceived candidate morality. Only among white nonevangelicals did ideological constraint play something of a supporting role in voting preference. Score 0 for Converse with evangelicals when issues with less prominence than abortion or gay marriage are involved--issues about which the evangelicals don't know the "right" answers.

But interesting differences can be noted between the evaluative approach employed by evangelicals and nonevangelicals. Specifically, while perceived candidate intelligence was an important consideration for many white evangelicals, most considered the two candidates equally intelligent and relied more on perceived candidate morality as a guide to voting choices. Relatively speaking, intelligence was more important to nonevangelicals, and morality simply not a distinguishing factor.

But how do evangelicals judge a candidate's morality, and why is it so important to them? The data at hand provide no ready answers, but the follow-up study suggests an intriguing possibility. Evangelicals' top three expectations of the Bush administration--securing God's blessing on the nation, increasing belief in God, and improving the country's moral values--have little in common with political policies the Bush administration officially pursues, or even could pursue, but they have a great deal in common with evangelical yearnings to promote evangelical-flavored religion in the wider culture. According to John MacArthur, many evangelicals interpret the Bible as asserting that God will bless a nation with peace and prosperity only if the nation's people believe in God and strive for moral purity. Thus, evangelical support for Bush may be one more example of what Hendershot describes as the evangelical propensity for adapting all sorts of secular cultural products--movies, rock music, jewelry, clothing, and even teen magazines--to the purpose of spreading the evangelical version of the Gospel.

If so, it may be helpful to try understanding evangelicals' political behavior from the perspective of W. Phillips Davison's "third-person effect" hypothesis. In brief, Davison found evidence that people tend to perceive media messages as more influential on others than on themselves, and to behave accordingly. The hypothesis has been used to explain several kinds of behavior, including voting behavior. As David Domke observes, evangelicals may vote for Bush partly or even mostly because they perceive his habit of referring publicly to his personal evangelical religious beliefs as highly effective at persuading nonevangelicals--the "others" in the minds of evangelicals--to take evangelical religious faith more seriously. Score 1 for Davison.

Kenneth R. Blake is associate professor of journalism at Middle Tennessee State University; Robert O. Wyatt is on leave as professor of journalism at Middle Tennessee and serves as a priest at the Episcopal Church of the Transfiguration, Palos Park, Illinois; Holly Warf is an undergraduate student at Middle Tennessee.

Caveat

Additional reading