Political protests do not just show changing political preferences, but can actually cause political views and behaviors to change, according to a new research paper co-authored by assistant professors at the Harvard Kennedy School.

The researchers collated data on the first major Tea Party protests in April 2009 to conclude that initial attendance at the rallies impacted how conservatively citizens and elected officials voted afterwards.

David Yanagizawa-Drott, one of the paper’s co-authors and an assistant professor at the Kennedy School, said that in the past, political and policy changes have been associated with protests, such as during the civil rights movement. However, no one had studied whether the protests themselves contributed to the changes.

“The alternative hypothesis is that policy preferences change over time, and political protests are just a reflection of that,” he said. Yanagizawa-Drott also noted that there was too little data to empirically study the effect of protests in the past, but the abundance of data on the Tea Party has made this analysis possible.

The researchers found that there was a large multiplier effect, so that a growth from 440,000 to 810,000 protesters at the rallies translated to 3.2 to 5.8 million additional votes for the Republican Party in the 2010 House of Representatives midterm elections. Elected officials also voted more conservatively in response to the size of the protests.

Stan Veuger, a co-author of the paper and an economist at the American Enterprise Institute, called the research the first natural experiment quantifying the influence of political protests on policy and political attitudes.

“We picked areas of the country that had very similar demographic and local dimensions,” Veuger said. “We saw that in places where it did not rain, the protests were bigger, and there were greater impacts in things like local voting, in spite of the two counties’ starting point being the same.”

It is unclear how exactly the Tea Party protests created such change, though Veuger said that personal interactions and information handed out at these rallies may partially explain the results.

The paper’s findings have several implications for political science.

“It shows that grassroots movements can certainly be effective, and that it doesn’t have to be about SuperPACs and national politics,” Veuger said.

Yanagizawa-Drott also added that the results helped explain the current political polarization in Washington, D.C., which he said can be explained as a consequence of the rise in the Tea Party’s prominence in the Republican Party after the 2009 protests.

Although the researchers focused on the 2009 Tax Day Tea Party protests, they said they believe their findings can be generalized and that they plan to continue their research on the impact of political protests.

“I can see how similar mechanisms can be applied to movements on the liberal end of the spectrum, and my colleagues and I have begun looking into labor movements in Europe,” Yanagizawa-Drott said.