Only 26% of unaffiliated voters also identify as independents. The remaining unaffiliated voters split evenly between Republicans (35%) and Democrats (34%). In other words, it’s probably okay to confuse registered Republicans (party registration) with self-identified Republicans (party identification), but unaffiliated voters are not always independents.
This analysis was performed by Zach Smith, a student research fellow at BYU’s Center for the Study of Elections and Democracy (like us on Facebook), in collaboration with CSED faculty. The writing is mostly his. Inquiries about this research should come to Quin Monson.
Even though they sound alike, party registration and party identification are conceptually different. In Utah and many other states, when you register to vote you can also register with a specific political party or as “unaffiliated.” This registration status can be changed online. Formal party registration is technically different from party identification—the “psychological attachment” or sense of identity that a voter has for a political party. Party identification is usually measured on a survey with a question something like, “Generally speaking, do you consider yourself a Republican, a Democrat, an Independent, or what?”
Registered Republicans/Democrats can also self-identify as Republicans/Democrats, but that is not always the case. Furthermore, “unaffiliated” registered voters can express an identity as a partisan when answering a survey question. This is important because political science has long shown that a voter’s party identification is a very strong predictor of voting behavior and political attitudes. If the overlap between the two is high, then party registration becomes a good substitute for party identification when campaigns wish to target specific voters.
We examine the overlap between registration and identification using Key Research data collected in cooperation with BYU’s Center for the Study of Elections and Democracy (CSED) for June 2012 and October 2012. The sample for both surveys was taken from the file of registered voters, which includes party registration. Both surveys also asked respondents to state their partisan identification. The combined samples give us 912 individuals for a comparison of party registration and party identification.
The figure below shows the percent registered and identifying with each party as well as the percent of unaffiliated (party registration) and independent (party identification)1 The proportion of voters registered as Republicans (60%) is comparable to the proportion who identify as Republicans (65%). The gap is wider for Democrats, but the proportions are similar. Here’s where it starts to get interesting: for both parties, the proportion of self-identifiers exceeds the proportion of registered partisans. This leads to a percentage registering as unaffiliated (30%) that is quite a bit higher than the percent self-identifying as independent (12%).
The figure below breaks out the percentage of registrants who identify with each party. Not surprisingly, registered partisans also overwhelmingly self-identify as partisans. For example, of registered Democrats, 87% self-identify as Democrats. Likewise, of registered Republicans, 89% self-identify as Republicans. Only 26% of unaffiliated voters also identify as independents. The remaining unaffiliated voters split evenly between Republicans (35%) and Democrats (34%). In other words, it’s probably okay to confuse registered Republicans (party registration) with self-identified Republicans (party identification), but unaffiliated voters are not always independents.
Utah Republicans have a closed primary that requires voters to register as Republicans to participate. Unaffiliated voters can change their voter registration at the polls on primary election day, while voters already registered with a party cannot. When they have a primary, Utah Democrats have open primaries that allow anyone to participate. You might expect that because of the closed rules for Republicans and the lack of regular primaries for Democrats that voters who self-identify as Democrats would register as Republicans to have influence in the Republican candidate selection process (or to make mischief). As the figure above shows, this isn’t happening much, if at all, in Utah. In fact, there is a slightly higher percentage of registered Democrats who self-identify as Republicans (5%) than registered Republicans who self-identify as Democrats (2%). Despite some incentives to register and participate in the other party’s primary, there are not many voters who are in a position to do so.
Party identification is also typically separated into seven categories so that partisan identifiers can express the strength of their identity. Independents are typically allowed to “lean” toward one party or the other (and these “leaners” are often thought of as “closet partisans” because they behave just like voters who openly identify with the parties).2
In the figure below, the unaffiliated voters are examined by all seven party id categories. The unaffiliated voters tend to clump in the middle as “pure” independents (26%) or independent “leaners” (36%). Altogether, 62% of unaffiliated voters are either “pure” independents or independent “leaners.” It’s also striking that such a significant proportion of unaffiliated voters strongly identify as Republicans (11%) or Democrats (15%). Despite their strong attachment psychologically to a party, these voters fail to register with their favored party on the voter rolls. They possibly represent the voters that are unhappy with their party but still feel closer to them than any other group (i.e. Tea Party voters). Or perhaps they are strong partisans who are reticent to reveal their affiliation on a public record.
From another angle, ideology, unaffiliated voters in Utah are somewhat more likely to identify as conservatives (41%) than liberals (33%) or as neither (31%).
So, what do we know about party registration and party identification patterns in Utah? First, while they aren’t perfectly correlated, partisans typically register with their chosen party. Second, the majority of “unaffiliated” voters are not all true independents, but instead are spread about evenly across Republicans, Democrats, and independents, especially when the “leaners” are correctly classified as partisans. Third, while most unaffiliated voters consider themselves “pure” independents or independent “leaners,” there is a sizeable group of strong partisans among them. So, party registration overlaps significantly with party identification among registered partisans, but when it comes to “unaffiliated” voters having their party identification is important before making any assumptions.