Guest Pollster | May 12, 2010
Topics: Indiana , Interpreting polls , midterm , Tea Party movement
Michael Wolf is an Associate Professor of Political Science at Indiana University. He can be reached at email@example.com.
Andrew Downs is Director of the Mike Downs Center for Indiana Politics and is an Assistant Professor of Political Science at Indiana University. He can be reached at firstname.lastname@example.org.
Craig Ortsey is a Continuing Lecturer of Political Science at Indiana University. he can be reached at email@example.com.
The authors would like to thank Brian Schaffner for his suggestions on an earlier draft of this piece.
Tea Party observers have floated two explanations for the group's emergence since their unexpectedly intense protests last year. The first explanation - embraced by conservative commentators and the movement itself - is that the Tea Party is comprised of grassroots citizens upset at the direction of the country and the deficit. Democrats champion a second explanation, that the Tea Party is composed of Republicans upset that President Obama and the Democrats control Washington. If the Tea Party is a movement against Washington politicians no matter their political stripes, then establishment Republicans must be wary of disaffected voters picking off their incumbents in primaries and President Obama faces a genuine rejection among voters he attracted in 2008. If it is simply Republicans upset at losing the presidency, 2010 looks more like a normal midterm election than an anti-incumbent revolt.
To get a better feel for the political dynamics behind the Tea Party, The Mike Downs Center for Indiana Politics asked registered Indiana voters whether they identified with the Tea Party, their vote intention for the Republican primary, and a series of election-related questions. Our first noteworthy finding is that 36% of registered likely Hoosier voters identified themselves with the Tea Party, while 61% of Republicans did.
Contrary to the "throw the bums out" rhetoric surrounding the movement, however, a plurality of Tea Partiers intended to vote for Dan Coats to be the Republican nominee for US Senate. Coats, a former senator and lobbyist who had homes in North Carolina and Washington, D.C. (but not Indiana) prior to jumping into the race, was recruited by the National Republican Senatorial Committee and was clearly the Washington establishment candidate. The candidates who reached out most aggressively to the Tea Partiers, Bates, Behney, and Stutzman, did relatively better with Tea Partiers than with non-Tea Party identifiers, but they still lagged behind Coats. Between this poll and the election, Stutzman's support surged, but that movement was more likely due to Senator Jim DeMint's Senate Conservatives Fund's late but strong support of his candidacy than due to a grassroots shift of non-Republican Tea Partiers looking his way.
So where were the pitchforks and torches against establishment Washington? Our findings demonstrate that Tea Partiers are overwhelmingly Republican. The blue bars in Figure 1 show the percentage of Indiana Tea Partiers in each partisan category. Four in ten Hoosier Tea Partiers are strong Republicans, and when weak Republicans and independents who lean Republican are added to the strong Republicans, nearly 80 percent of Tea Party identifiers are Republican Party adherents. Less than 10 percent of Tea Partiers are Democrats or independents who lean Democratic. True independents make up less than 13 percent of Tea Partiers.
The second piece of evidence that supports the position that the Tea Party is a Republican phenomenon comes from the red bars of Figure 1. Here the percentage of Indiana Tea Partiers who voted for Obama in 2008 is presented across each category of party identification. Less than seven percent of all Tea Party adherents voted for Obama, and they are largely comprised of a handful of disappointed Democrats. The differences between the red and blue bars represent McCain supporters, implying that the great majority of Tea Party independents were McCain voters and even half of the Tea Party Democrats were McCain voters. The genesis of Tea Party identification does not result from a rejection of Obama by his own supporters; rather, it arises more from upset McCain supporters - hardly a broad-based grassroots movement. What explains this pattern of Tea Party identification that looks as if it may have begun on November 5, 2008 rather than after the stimulus bills or auto bailouts? If the Tea Party were a response to the conditions of the country or frustration with spending, then a negative view of the direction of the country or a concern over the deficit should lead to an even distribution of Tea Party identification across party identification, or perhaps a bell-curve distribution concentrated among those independents who identify with the Tea Party. To test for these possibilities, we ran a logit model that yielded four significant explanatory variables. Two of these variables are issues associated with the Tea Party: believing that the US is on the "wrong track," and holding that the deficit is the most important issue facing the US. Another two significant variables are longer-term determinants: party identification and voting for John McCain in 2008. A factor analysis shows that party identification, view of national direction, and 2008 presidential vote all hang together as a single factor (the results of the logit model and factor analysis are available upon request). These outcomes imply that it is unlikely that the distribution of those viewing the national direction poorly is separate from Republican identifiers who voted for McCain. However, the salience of the deficit issue may still lead non-Republicans to be more apt to identify with the Tea Party and that distribution may be concentrated outside of Republicans.
Figure 2 indicates that this hypothesis is not correct. It presents the predicted probability of identifying with the Tea Party when one views the deficit as the most important issue (blue bars), the probability of Tea Party identification when the respondent believes that the deficit is the most important issue and views the US as being on the wrong track (maroon bars), and these two factors combined with voting for John McCain in 2008 (yellow bars) across each category of party identification. The overall message of this figure is that party identification conditions all of the factors that increase the probability of Tea Party identification. The distribution of deficit hawks' likelihood of identifying with the Tea Party is not a bell-shaped curve centered around independents, and in fact follows the strength of party identification in a nearly perfect progressive step-by-step pattern. When combining this factor with the view that the country is on the wrong track and (in a second step) with having voted for McCain, it is clear that the robust explanation for Tea Party identification is related to Republican Party identification rather than a populist reaction to national direction and deficits. What makes this result even more dramatic is when Figure 2 is juxtaposed against Figure 1. The predicted probability of Democrats identifying with the Tea Party given these attitudes looks impressive in Figure 2 (roughly 0.5 to 0.6 probability of Democrats identifying with the Tea Party when they hold these attitudes and voted for McCain). However, there are almost no Democrats who hold these attitudes and who voted for McCain. Only nine of the 343 Tea Party identifiers are strong Democrats, weak Democrats, or Democratic-leaning independents who hold these attitudes and voted for McCain. In other words, the maximum 0.6 probability of Tea Party identification by Democrats of any stripe given these conditions is deceptively strong. On the other hand, it is very telling for Republicans. Indeed, the variable with the largest marginal influence on Tea Party identification is voting for McCain in 2008, but for Hoosiers this act is intertwined tightly with Republican Party identification and viewing the country on the wrong track. Of the two explanations for the Tea Party's rise (a grassroots non-partisan movement upset at Washington policy versus Republican frustration with losing the 2008 election and the Obama administration's policies), our evidence from Indiana supports the latter. The Tea Party in Indiana is a Republican phenomenon whose effects will most likely be on voter mobilization rather than voter choice in next November's elections. While these results are only from one state, there is reason to think that a state with a culture of Midwestern agricultural individualism would be more likely than most states to have a Tea Party movement independent of partisan politics. The fact that it is not bodes ill for the grassroots explanation being correct in other states. The Tea Party is popular because it has provided aggrieved Republicans with a "reset" button unconnected to the past. Rather than voicing their frustrations by placing "Don't Blame Me! I Voted for McCain!" bumper stickers on their cars (which we do not expect to see soon in Indiana or elsewhere), the development of the Tea Party has operated as a convenient vehicle for Republican grievances that is unconnected to the unpopular end of the Bush era.
Note: Statement on Methodology: This SurveyUSA poll was conducted by telephone using the voice of a professional announcer. Respondent households were selected at random, using a registration based sample (RBS) provided by Aristotle of Washington DC. All respondents heard the questions asked identically. The calls were conducted from April 22-26, 2010. The number of respondents who answered each question and the margin of sampling error for each question are provided. Where necessary, responses were weighted according to the voter registration database. In theory, with the stated sample size, one can say with 95% certainty that the results would not vary by more than the stated margin of sampling error in one direction or the other had the entire universe of respondents been interviewed with complete accuracy. There are other possible sources of error in all surveys that may be more serious than theoretical calculations of sampling error. These include refusals to be interviewed, question wording and question order, weighting by demographic control data, and the manner in which respondents are filtered (such as determining who is a likely voter). It is difficult to quantify the errors that may result from these factors. Fieldwork for this survey was done by SurveyUSA of Clifton, NJ.