by Jonathan Singer, Fri Nov 10, 2006 at 10:02:35 PM EST
Despite the fact that the data do not support the conclusion that a more mobilized religious base was the key to George W. Bush's win in 2004*, it still remains a commonly held assumption by the political media. So it should not come as a surprise then that both The New York Times and The Washington Post run stories today trumpeting the supposed importance of Democratic gains among religious and Evangelical voters in the party's win in Tuesday's midterm elections.
The Post's Alan Cooperman, writing on the front page of Saturday's paper, is a particularly egregious offender, citing numbers that don't really prove his thesis that "faith" voters were key to the Democrats' win. Cooperman writes that Democrats "sliced the GOP's advantage among weekly churchgoers to 12 percentage points, down from 18 points in 2004 congressional races" and that while "in 2004, 74 percent of white evangelicals voted for Republicans and 25 percent for Democrats, a 49-point spread ... This year, Republicans received 70 percent of the white evangelical vote and Democrats got 28 percent, a 42-point spread."
Both of these sets of figures sound really impressive -- until you look deeper at their actual meaning. When comparing results between elections, it is not only important to look at absolute numbers but also relative numbers: How a specific subset of the electorate moves from one election to another relative to the change in the electorate as a whole. Although vote totals in districts have not yet been finalized, making it difficult to tabulate the exact national popular vote for the House, it appears that the nation voted a net 10 points more Democratic in 2006 House elections than it did in 2004 (a move from a Republican advantage of about 49 - 46 to a Democratic advantage in the ballpark of 53 - 46). So for a group to have disproportionately helped the Democrats take the House this year, it would have to have increased its net support for the Democrats by more than 10 points -- which none of the groups cited by Cooperman actually did.
So which groups did noticeably boost their support for Democrats over the past two years? As I noted on Wednesday, Latino voters backed Democrats by a net 28 more points in 2006 than they did in 2004, as did those without a high school diploma. Jewish voters upped their net support for Democrats by 21 points. Those earning $200,000 or more per year voted a net 19 points more Democratic in 2006 than they did in 2004.
Looking at vote broken down by church attendence, Democrats gained a net 13 points from those who never attend church, 10 points from those who attend church a few times a year, 16 points from those who attend monthly, 8 points from those who attend weekly and just 2 points from those who attend more than weekly. Even discounting for the fact that voters who say they attend church at least one a week increased as a share of the overall electorate by about 3 points over the past two years, it's overwhelmingly clear that the Democrats gained many more votes from less churchgoing voters than from more churchgoing voters.
Now I wish that a closer look at this data would cause political reporters and Beltway pundits to second-guess their beliefs that religious voters were key to Democrats' wins in 2006, but I'm not holding my breath for that to happen.
Yet more importantly, Democratic strategists need to come to the realization that religious voters, specifically White Evangelicals and/or those who attend church weekly or more often, are not a part of a Democratic majority coaltion and aren't going to be any time soon. In fact, secular voters are significantly more important to the Democratic base than more observant voters. This isn't to say that Democrats shoudn't talk about their values or even try to reach out to the "faith voters" or "values voters." But if the Democratic Party wants to maximize the returns from its efforts in bringing out its strengths among the broader electorate, it would be much better served by targeting its own base, including those who rarely if ever attend church, rather than trying to win over a rather small small set of votes from the Republican base, most notably the "churched" and Evangelical Christian communities.