On Reconsidering The "Myth of the Middle"

Steve at The Carpetbagger Report brings up an excellent discussion today:

It’s one of those political truths that “everybody knows” — the party that wins the elusive middle wins the election. It’s all about the “center,” where most Americans are and where campaigns are decided. This seemed particularly true in 2006, when, the conventional wisdom tells us, the middle expressed its disgust with the status quo and backed a divided government so that both sides would govern from the center.

But is any of this true? Political scientist Alan Abramowitz and journalist Bill Bishop suggested this week that we may want to reconsider the “myth of the middle.”

The Cooperative Congressional Election Study (CCES) surveyed more than 24,000 Americans who voted in 2006. The Internet-based survey compiled by researchers at 30 universities produced a sample that almost perfectly matched the national House election results: 54 percent of the respondents reported voting for a Democrat, while 46 percent said they voted for a Republican. The demographic characteristics of the voters surveyed also closely matched those in the 2006 national exit poll. If anything, the CCES respondents claimed they were more “independent” than those in the exit poll.

The CCES survey asked about 14 national issues: the war in Iraq (the invasion and the troops), abortion (and partial birth abortion), stem cell research, global warming, health insurance, immigration, the minimum wage, liberalism and conservatism, same-sex marriage, privatizing Social Security, affirmative action, and capital gains taxes. Not surprisingly, some of the largest differences between Democrats and Republicans were over the Iraq war. Fully 85 percent of those who voted for Democratic House candidates felt that it had been a mistake to invade Iraq, compared with only 18 percent of voters who cast ballots for Republicans.

But the divisions between the parties weren’t limited to Iraq. They extended to every issue in the survey. For example, 69 percent of Democratic voters chose the most strongly pro-choice position on the issue of abortion, compared with 20 percent of Republican voters; only 16 percent of Democratic voters supported a constitutional amendment to ban same-sex marriage, while 80 percent of Republican voters did; and 91 percent of Democratic voters favored governmental action to reduce global warming, compared with 27 percent of Republican voters.

When we combined voters’ answers to the 14 issue questions to form a liberal-conservative scale (answers were divided into five equivalent categories based on overall liberalism vs. conservatism), 86 percent of Democratic voters were on the liberal side of the scale while 80 percent of Republican voters were on the conservative side. Only 10 percent of all voters were in the center. The visual representation of the nation’s voters isn’t a nicely shaped bell, with most voters in the moderate middle. It’s a sharp V.

OK, if this is true, and Abramowitz and Bishop certainly make a compelling case, what does this tell us about how the political process should work?

Your thoughts?

I'm not sure I "buy" that there is so little "middle". But I also wonder if those most willing to participate in surveys are those with strong ideological viewpoints compared with others who may not be willing to discuss national issues.