Let’s start with what may be the understatement of the year: Donald Trump’s presidential campaign was often surprising, occasionally shocking. He routinely made gaffes that would have disqualified a “politics as usual” candidate. He had a skeletal campaign organization. He spent almost nothing on television advertising until the very end of the campaign. And, he routinely claimed the process was rigged against him, implicating not just the Republican and Democratic political establishments but the traditional news media as well.
In October, Harvard political scientist Pippa Norris explained in the Monkey Cage why such rigged election talk would likely backfire. Her conclusion, supported by a wealth of empirical evidence, was that casting doubt on the integrity of electoral process typically drives down voter participation. After all, why participate in an election if the votes aren’t going to be counted fairly?
Our evidence is based on data from a GfK Knowledge Panel conducted from November 4 – November 8, 2016. The data include measures of voting intention, partisan affiliation, standard demographic variables (education, income, age, and race), as well as items gauging perceptions of voter fraud and media bias. The final data are weighted to match the most recent available census estimates (March 2016) and include 2,367 responses (and 1,887 likely voters).
Specific question wording for the voter fraud and media bias questions are listed below:
1. Do you think that voter fraud will affect the results of the 2016 presidential election?
Yes
No
2. IF YES: Which of the following comes closest to your view when it comes to how fraud might affect the 2016 presidential election?
Hillary Clinton will receive a lot more votes than she should because of vote fraud
Hillary Clinton will receive somewhat more votes than she should because of vote fraud
There will be no difference in votes due to voter fraud
Donald Trump will receive a lot more votes than she should because of vote fraud
Hillary Clinton will receive a lot more votes than she should because of vote fraud
3. Thinking now about news coverage of the candidates over the course of the campaign, would you say that, on balance, news coverage has been…
Very biased in favor of Hillary Clinton
Somewhat biased in favor of Hillary Clinton
Fair and Balanced
Somewhat biased in favor of Donald Trump
Very biased in favor of Donald Trump
As you might expect, Republicans were much more inclined to believe that vote fraud would affect the election results, that voter fraud would yield a lot more votes for Hillary Clinton, and that the media were very biased in favor of Clinton. Democrats were not only less inclined to believe any of these propositions but were also not particularly inclined to believe the media were favoring Trump OR that Trump would benefit from voter fraud. Independents fell somewhere in-between.
How did these perceptions influence self-reported intent to vote in the November 8 election? The answer depended on partisanship. Republicans who believed that voter fraud would affect the election – and especially Republicans who thought Clinton would benefit from fraud – were much more likely to say they would definitely vote. Democrats hearing these same arguments became less inclined to vote.
A similar story emerges when we consider perceptions of media bias. Republicans who believed the media were very biased in favor of Clinton were more likely to say they would definitely vote. Democrats who believed the media were biased in favor of Clinton were less likely to vote. Trump’s delegitimizing rhetoric, perhaps because it specifically blamed Democrats and the liberal media for trying to rig the campaign, fell into a mobilization sweet spot. It simultaneously mobilized Republicans, while demobilizing Democrats.
At the end of a long campaign decided by fewer than 110,000 votes (slightly larger than the size of a major college football stadium), Trump’s incessant complaining about the fairness of the process sounded like a candidate looking for an excuse for an inevitable defeat. In hindsight, it may have been more strategic and effective than it initially appeared. Rather than accepting the results of a rigged campaign or letting their loss of faith in the political system keep them home, Trump’s supporters (we suspect especially in rural communities) turned out to express their anger by voting against the establishment. They may have been motivated, at least in part, by Trump’s expression of anger and frustration that the system was rigged against them.