What the Exit Polls Said

The election is over, and the time has come to determine what the voters said by looking at the exit polls. Here’s what I see in the data served up with two caveats.

First, “the people” don’t say much except “No.” Of course, “no” to the ruling party also means “yes” to the alternative, in this case to the Democrats. But there is no collective mind that instructs the government in any detail beyond “no.”

Second, I’m going to compare 2006 exit poll data to a similar survey in 2004. But there are two differences. The electorate in the 2004 presidential contest was about 50 percent larger than the turnout last week. Many of the differences between 2006 and 2004 you are going to hear about in the next week or two will be relatively small. For example, self-described conservatives made up 34 percent of the voters in 2004 and 32 percent this year. That difference may just reflect the relative propensity of conservatives to vote in presidential and midterm elections (or it may just be random variation). (I can’t compare 2006 to the last midterm in 2002 because of problems with the latter poll). For this reason, I’m only going to pull out of the data largish differences between 2004 and 2006.

In general, the 2006 electorate was not all that different from the 2004 group. However, there are some differences:

Self-identified independents (26 percent of the vote) went from a 48-49 Republican-Democratic split to 39-57 favoring the Dems.

In 2004, 84 percent of self-identified conservatives voted for the GOP. This year 78 percent did.

Self-described moderates split 45-54 Republican-Democrat in 2004. This year the division was 38-60 against the GOP.

45 percent of the electorate said in 2006 that they attended church once a week or more. The GOP lost 6 percent of that group compared to 2004 while the Democrats were up 4 percent.

The Republicans also were down 5 percent among Protestants and 8 percent among Roman Catholics. Just over 80 percent of voters identify as Protestant or Catholic.

Asked about the importance of various topics, 67 percent said Iraq was either extremely or very important. 82 percent said the same about the economy; 74 percent said the same about corruption and ethics.

What does it all mean? The GOP lost significant support among independents, moderates, conservatives, and the pious, especially Roman Catholics. Each of these groups is a significant part of the electorate.

The question of the independent is especially interesting. 26 percent of the electorate told exit pollsters that they were independent of party. Earlier studies suggest only about 10 percent of the electorate are actually independent, and they are less likely to vote. Most “independents” do vote for one of the major parties whatever they say to exit surveys. My guess is that the GOP lost the votes this time of a lot of self-identified independents who normally vote Republican.

Exit polls do not ask people whether they are libertarians. My thought – yet another guess – would be that libertarians might identify themselves as independents in party and either as moderate or conservative in ideology. If so, the GOP lost their vote too.

Commentators often say the Republican party is a balancing act between economic libertarians and social conservatives. A GOP majority is always precarious: what pleases economic libertarians must alienate social conservatives and vice-versa. In 2006, the Republicans’ conduct of government along with the Iraq war alienated both aspects of its base along with many other voters who are not attached to one of the parties.

The picture is not all bleak for Republicans in 2008. In 2004, President Bush won two-thirds of the open and formerly Republican House districts that elected Democrats in 2006. Many of them could return to the GOP in 2008.

Democrats will have two years to dig in their incumbents in. If they can hold their majority in 2008, I would not bet against a Democratic House for a generation.

Finally, a strange result. When exit polls asked “Will the Democrats make American safe?,” fully 29 percent of Democrats answered “no.” The party apparently has yet to convince on national security.