People might be more likely than ever to protest in reaction to a social problem or geopolitical crisis. But do such activist events, even large-scale demonstrations, change public opinion?
New research shows that protests rarely change views or alter voting decisions, findings that could have important implications in an election year marked by widespread protests.
Harvard Business School Professor Vincent Pons teamed up with Amory Gethin, a fellow economist at the World Bank Development Research Group, to study 14 major social movements in the United States within the past decade. The researchers measured changes in people’s attitudes before, during, and after protests on a variety of topics, including environmental protection, gender equality, gun control, immigration, national and international politics, and racial issues.
In a working paper published in April
by the National Bureau of Economic Research, Pons and Gethin say “protests generate substantial internet activity but have limited effects on political attitudes.”
One exception: Black Lives Matter
The Black Lives Matter protests in the wake of George Floyd’s murder by police on May 25, 2020 “were followed by a sharp increase in liberal attitudes on racial issues and vote intentions for the Democrats” in the 2020 US presidential election, the paper says. While it’s unclear why the Black Lives Matter movement shifted more views than other social protests, Pons says the intense media coverage over a long period of time and the large number of protesters who took to the streets may have been factors.
“Clearly the scale of the protest matters, but it cannot just be that.”
“If you ask anyone to cite important [protests] that took place in the US in the last six years, they will cite Black Lives Matter,” says Pons. “And so, one dimension of this is just the number of participants.” At the same time, however, he notes that the Women’s March of 2017 had even more participants yet did not sway as many opinions. “So clearly the scale of the protest matters, but it cannot just be that. These are hypotheses that are difficult to test.”
A study of several protests
While most previous studies of nonviolent social movements often focused on just one protest, Gethin and Pons set out to use a wider lens, researching 14 protest movements between 2017 and 2022. To gauge personal attitudes, previous projects often relied on surveys taken at two points: before the protest and after. Pons and Gethin also wanted to measure attitudes weeks before and weeks after the event, since they wondered: Would opinions be more likely to change while emotions were high during a protest versus over the longer haul?
The researchers gathered data from Twitter (the platform now called X), Google searches, and high-frequency surveys like The Gallup Poll Social Series, Nationscape, and Cooperative Congressional Election Study to chart the evolution of online interest, policy views, and vote intentions before and after the outset of each social movement.
“Our approach significantly improves upon existing work by providing a comprehensive view on the impact of protests on attitudes and on the underlying channels,” the researchers write.
Most protests don’t change long-term views
The results showed that 13 out of 14 protests didn’t have a large, sustained impact. Specifically, the researchers found:
- Protests generated significant online activity initially, but it declined relatively quickly. At the beginning of a social movement, social posts and Google searches related to the issues raised by the protesters increased dramatically. However, this flurry of activity was relatively short-lived, reducing to pre-movement levels within 10 days.
- The majority of protests didn’t alter people’s views. Most movements did not lead to significant changes in public opinion on policies or alter people’s voting intentions.
- The George Floyd protests were the one exception to the rule. The Black Lives Matter protests created a large and sustained change in views, including a 4 percentage-point increase in liberal views. After the protests, the percentage of respondents who intended to vote for Donald Trump decreased by 2 to 3 points. The percentage of people who planned to vote in the presidential election also ticked up.
Breaking down voter views
This research is just a slice of Pons’ larger research agenda aimed at understanding what makes people vote a certain way, how democratic systems function, and how they can improve. For instance, what factors, such as parents’ beliefs and geographic location, influence voters? And do stricter voting requirements keep people away from the voting booth on election day?
“The hope is that in a democracy, not only do you have strong participation but also an informed electorate. ”
“One question I want to know the answer to is: How exactly do people make up their minds?” Pons says. “And the second is, how informed are voters about the candidate’s quality? The propositions of the candidates? Because the hope is that in a democracy, not only do you have strong participation but also an informed electorate. So, it’s an open question to understand how informed exactly voters are.”
You Might Also Like:
Feedback or ideas to share? Email the Working Knowledge team at hbswk@hbs.edu.
Image: Image created by HBSWK using asset from AdobeStock/Spaxiiax
“Harvard Business School is the graduate business school of Harvard University, a private research university in Boston, Massachusetts. It is consistently ranked among the top business schools in the world and offers a large full-time MBA program, management-related doctoral programs, and executive education programs.”
