Most people don’t give much weight to the influence that television shows have on politics. After all, there’s dozens of channels, outlets, and shows available no matter what your political leaning may be. People tune into the political shows that generally align with their political views.
It may then come as a shock that a single television show can actually have a much bigger impact than anyone realized. So big an impact, in fact, that a set of researchers have concluded that it was one of the deciding factors in the 2016 United States presidential election.
We know from past social psychology research that television shows can and do have an impact on voter attitudes. Research conducted in 2005 by Matthew Baum and another study conducted in 2010 (Parkin, 2010) demonstrated that “when presidential candidates appear on televised comedy shows, they are able to have an unusual impact on otherwise disengaged voters” (Porter & Wood, 2019).
Despite the conventional wisdom, most people don’t turn to these shows primarily for news updates, like they do a mainstream news outlet like Fox News or CNN. Instead, most people watch a show like The Daily Show on Comedy Central to be entertained. And it’s easy to understand how a liberal show appeals mostly to other liberals, just as a conservative show appeals mostly to other conservatives.
What is less well-known is these shows’ impact on undecided voters. And it could be much larger than most people imagine:
However, as previous studies have shown, the viewers Stewart and Colbert most affected were otherwise politically disengaged and uninformed. As Zaller (2004) demonstrates, such voters are precisely the most likely voters to shift partisan allegiances between presidential elections.
What Does the New Research Show?
During 2016, two Comedy Central shows — The Daily Show and The Colbert Report — transitioned hosts, providing a rare opportunity for researchers to study and model the effects that this change had on viewers and potential voters’ attitudes. When both shows transitioned hosts — Trevor Noah taking over for the popular Jon Stewart and Larry Wilmore taking over for Stephen Colbert — they both lost ratings. As a control group for the study, the researchers also looked at other popular Comedy Central shows running at the same time and none of them suffered similar ratings declines.
And when ratings declined, so did a lot of the viewers who were either politically neutral or unengaged with politics:
We find that the transition in hosts at The Daily Show, and the subsequent ratings decline, had a positive effect on Donald Trump’s 2016 vote share. We do not find that the replacement of Colbert with Wilmore had similar effects on vote share, nor do we find that ratings of other Comedy Central shows affected vote share.
How bad was it?
If the vote share differences we attribute to The Daily Show’s ratings decline had not occurred, would the 2016 presidential election have turned out differently?
Our evidence suggests that the answer is yes. In a world in which Trump does not gain 1.1% of the vote share over Romney at the county level, Clinton wins the electoral college in 69% of our simulations.
These results should be interpreted cautiously; they do not mean that Stewart, and Stewart alone, paved Donald Trump’s path to the White House.
While winning 69 percent of the simulations is not overwhelming evidence for the impact of this single TV show, it is strong data that is suggestive that the show wielded more political influence than most people gave it credit for. In fact, even Jon Stewart downplayed the impact of The Daily Show when he hosted it:
And when asked to describe his importance, he said: “On a scale of zero to 10, I’d go with a zero, not very important” (Cooper and Bailey, 2008).
Yet as is well-known, many viewers counted The Daily Show as a primary news source during Stewart’s tenure (Pew, 2004).
Few television shows could be said to have the impact that The Daily Show once had. It’s a shame the show’s political impact appears to have declined when Stewart departed.
Limitations of the present study include the fact that it did not use a randomized design and used an econometric tool in order to make causal inferences — something that isn’t as powerful or robust as using an actual experimental design meant to study causality. And, of course, the study is only relevant to a single presidential election; a different election between different candidates may have resulted in very different findings.
This research is an interesting reminder that sometimes the things we think don’t matter might actually matter more than we realized. A political comedy show that seemed like it was mostly just for fun and comedy may have had a much bigger impact on the 2016 presidential election than anyone realized at the time.
Baum, M. (2005). Talking the vote: why presidential candidates hit the talk show circuit. Am. J. Pol. Sci., 49 (2), 213-234.
Cooper, CA & Bailey, MB. (2008). Homer Simpson Goes to Washington: American Politics through Popular Culture. University Press of Kentucky, Lexington, Kentucky.
Parkin, M. (2010). Taking late night comedy seriously: how candidate appearances on late night television can engage viewers. Publ. Opin. Q., 63, 3-15.
Porter, E. & Wood, TJ. (2019). Did Jon Stewart elect Donald Trump? Evidence from television ratings data. Electoral Studies, in press.
Zaller, J. (2004). Floating voters in U.S. Presidential elections, 1948-2000. Willem E. Saris, Paul M. Sniderman (Eds.), Studies in Public Opinion: Attitudes, Nonattitudes, Measurement Error, and Change, Princeton University Press: Princeton, NJ.