Does the media influence people’s political views in a divided country? A recent MIT study reveals that it depends on what media people like and how we measure those preferences.
The researchers combined a large online survey with data tracking which news websites participants visited a month before the study. They found that what people said in the survey generally matched their actual news consumption. But there were some important differences.
Varied news habits
First, people who reported the same media preferences actually had quite varied news habits, which means that survey-based measures might not tell the whole story. Second, people with different media preferences in the survey often visited the same online news sites. These findings challenge common ideas about how polarized Americans’ media habits are and make us question the use of survey data when studying the impact of political media.
“There’s good reason to think that the information people report in surveys may not be a perfect representation of their actual media habits,” the researchers explain.
The study aimed to address a divide in academic research. Some scholars argue that existing polarization leads to people consuming highly partisan media, while others believe that partisan media sources shape individuals’ more polarized views. However, very few studies have simultaneously measured both the self-selection of media and its persuasive effects, using a combination of survey and behavioral data.
Understanding the news
To carry out the experiment, the researchers partnered with the media analytics company comScore in 2018 to enlist a diverse group of American adults. ComScore then merged survey responses from over 3,300 participants with detailed information about their web browsing habits in the month leading up to the study.
In the first part of the study, participants were asked about their media preferences, including how much and what type of news they liked to read. In the second part, participants were divided into two groups. The first group could choose whether they wanted to read media from Fox News, MSNBC, or an entertainment option, while the second group had to read articles from one of these three sources.
This approach allowed the researchers to compare what people said they liked in the survey with what they actually read online, and to understand how persuasive partisan media could be to different groups of readers.
Variations in persuasiveness
The study found variations in how persuasive partisan media was among different news audiences. When looking at how much news participants consumed, the researchers noticed that those who visited fewer news sites compared to entertainment sites were more likely to be influenced by partisan media.
However, when examining the political leanings of participants’ news consumption, there was a small but significant difference between what they reported in the survey and what their web browsing behavior revealed. Survey data suggested that people might be open to information from sources with opposing ideologies. On the other hand, the web-browsing data indicated that individuals with more extreme media preferences were primarily influenced by outlets that aligned with their existing views.
“Together, these results suggest that inferences about media polarization may depend heavily on how individuals’ media preferences are measured,” the authors conclude.
