Previous studies have found that after an election, supporters of the losing candidate tend to be more likely to believe that votes were counted improperly. So how did Donald Trump’s claims that the 2016 election was ‘rigged’ ahead of his own victory influence voters’ perception of the election’s integrity? Using national survey data, Betsy Sinclair, Steven S. Smith and Patrick Tucker find that Republicans gained confidence in the vote after Trump’s election win, while Democrats’ confidence fell. This effect was even more pronounced among better-educated Republicans and Democrats who identified most strongly with their party.
The major presidential party campaigns in the 2016 election discussed the issue of election integrity in a way that was unprecedented in the modern era. Most obviously, Donald Trump made high profile claims that the system was “rigged.” In the primaries Trump complained that the GOP nomination process could present unfair obstacles. These accusations of unfairness were extended to the general election, as well. During an August campaign stop, the Republican nominee told that crowd that he was “afraid the election’s going to be rigged,” pointing to seemingly unbelievable rumors that entire precincts reported no votes for Mitt Romney in 2012.
At the same time, the Democratic party brought attention to election fairness. During the primaries, Clinton’s overwhelming support from superdelegates led many supporters of Bernie Sanders to suggest the eventual nominee had an unfair advantage. The general election also highlighted the issue of election integrity. In response to many of Trump’s claims of rigging and his unwillingness to accept the election’s outcome should he lose Hillary Clinton strongly rebutted these accusations during the third presidential debate, suggesting that Trump was making excuses for losing and that electoral institutions were strong. Similarly, election experts dismissed Trump’s suggestions.
This heightened importance of electoral integrity led some many to wonder about voters’ confidence with the election’s outcome. While this level of attention to American voters’ perceptions may have appeared novel, political scientists have studied what they term the “winner effect.” That is, those voters who support the winner in the election are more likely to believe their own ballots and the ballots of others were counted as intended, while those who supported the losing candidate are more likely to believe their votes were counted incorrectly. In their study of voter confidence in national elections 2000 through 2012, Michael Sances and Charles Stewart found that while most supporters of both the winner and loser of an election are very confident in that their own vote was counted correctly, the confidence of those on the losing side slightly decreases. Moreover, the confidence of those on the winning side increases significantly from its pre-election level.
“Reading the ballot” by Dan Kennedy is licensed under CC BY NC SA 2.0
To investigate if this winner effect existed in an election where the eventual winner was sending warnings of possible fraud while the eventual loser was endorsing institutions’ legitimacy, we took advantage of The American Panel Survey (TAPS), an online, nationally representative survey. Respondents were asked to report how confident they were that their confidence in vote counting in October 2014, October 2016, November 2016, December 2016, and January 2016. Responses ranged from 1 (“not at all confident”) to 4 (“very confident”).
Figure 1 shows the average level of confidence on the four-point scale for Trump and Clinton voters from 2014 through early 2017. Interestingly, in 2014 the confidence of supporters for both candidates was roughly the same level. Yet, by the fall of 2016 a significant gap in confidence has occurred. Most interestingly, this gap was not necessarily due to Republicans adopting Trump’s message of rigging. Rather, Republicans remained essentially unchanged, on average, from their 2014 levels. Instead, we see Clinton voters significantly increasing their level of confidence before the 2016 presidential election. Like the findings of Sances and Stewart, we find strong evidence for a winner effect following the election. The average Trump voter significantly increased their level of confidence that their vote was counted correctly, while the average Clinton voter had less confidence. The shifts appeared to be durable, as the gap remained between the two groups, and if anything, widened as the Trump administration began.
Figure 1 – Voter confidence in vote counting 2014-2017
Our data also allows us to identify if any individual characteristics were associated with change for both candidates’ supporters. For Trump voters, we found that education level was highly predictive of change. As Figure 2 illustrates, a Trump supporter with a 12th grade education (8 on the x-axis), was predicted to increase their level of confidence by roughly ¾ of a category on the 4-point scale. This predicted effect was not significantly distinct from zero for Trump voters with post-graduate education (a 13 or above on the x-axis) and also not predicted to be distinct from Clinton voters’ change. At the same time, we found that among Clinton voters, those who identified as strong Democrats were predicted to change their level of confidence much more negatively than their “weak” Democratic counterparts.
Figure 2 – Education and party identification and support for Trump or Clinton
Finally, because of the way we collected our data, we were able to investigate whether news of Russian hacking may have influenced voters’ perceptions of electoral integrity. The first reporting that Russians may have attacked state election voting systems occurred on November 16th. When we compared those who answered the survey before that date, we found that Clinton voters expressed significantly weaker levels of confidence that their vote was counted accurately compared to those Clinton voters who answered closer to election day. Surprisingly, Trump voters displayed similar drops in confidence compared to their counterparts who answered closer to election day.
Our findings are important: they confirm that the winner effect existed in an election in which electoral integrity was highly questioned, and our survey results suggest that these changes in confidence persisted even after the election. Our findings should encourage those who look at voter confidence to examine the long-term effects of elections on voters’ beliefs about democracy.
- This article is based on the paper, ‘“It’s Largely a Rigged System”: Voter Confidence and the Winner Effect in 2016’, in Political Research Quarterly.
Note: This article gives the views of the author, and not the position of USAPP– American Politics and Policy, nor of the London School of Economics.
Shortened URL for this post: http://bit.ly/2QlNRow
Betsy Sinclair – Washington University in St. Louis
Betsy Sinclair is a Professor of Political Science at Washington University in St. Louis.
Steven S. Smith – Washington University in St. Louis
Steven S. Smith is the Kate M. Gregg Distinguished Professor of Social Science and a Professor of Political Science at Washington University in St. Louis. He is the Director of the Weidenbaum Center on the Economy, Government, and Public Policy.
Patrick Tucker – Yale University
Patrick Tucker is a postdoctoral associate in the Institution for Social and Policy Studies (ISPS) and the Center for the Study of American Politics (CSAP) at Yale University.