Thinking Arizona
Facts to Inform the Public Discourse
Splitting The Divide On Candidates, Policy
August 18, 2011
Democrats and Republicans may be unalike in most regards but they share this: They’re completely, totally, absolutely predictable.
Nine times out of 10, Republicans will vote for the Republican candidate for president. Or at least they did in 2004 and 2008. And nine times out of 10, Democrats will vote for the Democratic candidate for president.
Their ballots might be cast with an open mind or mindlessly. They might vote blindly or eyes wide open. Their choices might be out of loyalty, or conviction, or habit. But they’re completely, totally, absolutely consistent. They are partisans.
A Less Predictable Mix
Unaffiliated voters aren’t similarly afflicted. The only predictable thing about them, at least in aggregate, is that they will come down somewhere between the Democrats and Republicans on every candidate and ballot issue.
Call them conflicted, call them ‘tweeners, call them centrists, but the explanation for their middleness is probably not that they’re a bunch of hand-wringing, fence-straddling, can’t make up their mind milquetoasts. The more likely reason is that they collectively cover such a wide part of the political spectrum, they even each other out. For every pro among them, there is a con.
When Barack Obama and John McCain ran for president in 2008, 90 out of 100 Democrats voted for Obama and 90 out of 100 Republicans voted for McCain. But of the unaffiliateds, 55 of 100 voted for Obama, 40 voted for McCain, and the remainder penned in someone else. In 2004, it was 55 for John Kerry and 42 for George Bush. The unaffiliateds land in the middle, particularly when one discounts for the skew of the exit polls.
When John Kyl and Jim Pederson ran for the U.S. Senate in 2006, 90 out of 100 Republicans voted for Kyl, 90 of 100 Democrats voted for Pederson, and – at least according to the exit polls – the unaffiliateds split 56-35. The remaining nine voted for someone else.
When Arizonans were asked to decide a constitutional amendment in 2004 requiring proof of citizenship to vote, 65 of 100 Republicans said yes, 58 of 100 Democrats said no, and the unaffiliateds divided 47 to 49 (with the remainder abstaining).
The middling pattern is repeated over and over. Unaffiliateds, for example, are less white than the Republicans, more white than the Democrats. They are more Hispanic than the Republicans, less Hispanic than the Democrats. They were less likely than Republicans but more likely than Democrats to strongly approve of the war in Iraq. They were more likely than Republicans but less likely than Democrats to favor granting legal status to illegal immigrants. And so it went. Their percentages of yes’s and no’s landed somewhere between the Republicans and Democrats on 181 of the 274 possible answers to the political and demographic questions contained in the 2004, 2006, and 2008 exit polls.
In a few cases, the “somewhere between” was exactly midway. For instance, just over 90 of 100 Republicans voted to re-elect John McCain to the U.S. Senate in 2004. Here was a case where a majority of Democrats agreed with their rivals. Just over 50 of 100 voted for McCain. The unaffiliateds, with just over 70 of 100 voting for him, were exactly halfway between the voting pattern of the two parties.
But in most cases, “landed somewhere between” doesn’t mean exactly midway. Looking at which way they shaded on any particular question provides some insight as to their preferences.
Take, for example, the war in Iraq. In 2008, nearly 90 of 100 Democrats expressed some level of disapproval. By comparison, 25 of 100 Republicans were to some degree critical. The midpoint of the two would be a disapproval rating of 57 or 58 of 100. Anything lower than that could be construed as the unaffiliateds nudging closer to the Republican view. A higher number would put them closer to the Democratic view. The latter was the case as 67 of 100 unaffiliateds registered some level of discontent with the war.
Some Answers Tilt Democratic
Other questions where the unaffiliateds leaned more to the Democratic viewpoint were:
- Opposition to the 2006 proposition to ban same sex marriage. (Democrats: 63 of 100. Republicans: 21 of 100. Midpoint: 42. Unaffiliateds: 57.) The measure failed in 2006. It then passed in 2008 due to less opposition from Democrats and unaffiliateds (Democrats: 59 of 100. Republicans: 20 of 100. Midpoint: 40. Unaffiliateds: 50.) Of the remainder of the unaffiliateds in 2008, 46 favored the amendment and four said they did not vote on it.
- Agreement as to whether the country was more or less safe from terrorism in 2004 than four years earlier. Democrats answered less safe. (Democrats: 70 of 100. Republicans: 15 of 100. Midpoint: 43. Unaffiliateds: 49.) Republicans answered more safe. (Democrats: 26 of 100. Republicans: 82 of 100. Midpoint: 54. Unaffiliateds: 44.) The remainder saw no difference.
- Agreement in 2008 that illegal immigrants now in the country should be granted legal status, as opposed to being deported. (Democrats: 69 of 100. Republicans: 46 of 100. Midpoint: 58. Unaffiliateds: 62.) The same question in 2006 brought similar responses, although unaffiliateds were right on the midpoint that year with 56 of 100 in favor.
On a few ballot questions, the unaffiliateds were split down the middle. For example in 2004, 47 of 100 favored while 49 of 100 opposed the proposition that proof of citizenship be required to vote.
On a few others, everyone took the same side. No group favored the 2008 proposition that many saw as an attempt to dilute the accountability of employers to ascertain the legal status of new employees. (Democrats: 40 of 100 in favor. Republicans: 35 of 100. Unaffiliateds: 38.)
Republican Tilt Lacking
But in no cases did the exit polls show the unaffiliateds tilting more heavily toward the Republican position on any of the ballot propositions that were polled.
Conversely, it stands to reason that the pendulum would have to swing toward Republicans and away from Democrats on one or more other types of issues. Otherwise the unaffiliateds would be shadow Democrats. For instance, one possible difference is over fiscal policy. However, as noted elsewhere in this report, the three exit polls studied here didn’t ask participants about fiscal policy or other possible sticking points.
On those untested issues, one guesses that the overall pattern would continue. Even if the unaffiliateds were to tilt more toward the Republican perspective, it wouldn’t be by all that much. They still would land as they usually do . . . somewhere between the extremes.
– Richard Gilman
