Crime videos push liberal voters toward stricter laws but not the political right

When exposed to videos of retail crime, voters in a highly liberal city adopted tougher, more conservative stances on local criminal justice policies. The participants did not, however, alter their broader political identities or overall ideological leanings. These findings were published in the The Journal of Social Psychology.

In psychology, a concept called terror management theory explores how people manage their deep-seated awareness of their own mortality. The theory suggests that reminders of death or existential threats create intense anxiety. To cope with this discomfort, individuals often seek comfort in structured, predictable, and historically conservative worldviews.

Decades of research on this topic have yielded mixed results regarding how people vote under stress. Some experiments show that threats push people toward the political right. Other studies suggest that threats cause polarization, driving both liberals and conservatives further toward their respective extremes.

Recent political shifts in the United States provided a real-world testing ground for these ideas. In California, voters recently considered revising a state law known as Proposition 47. Originally designed to reduce prison populations by classifying minor thefts as misdemeanors, the law has faced intense debate in recent years. Critics have argued that the $950 threshold for a felony charge encouraged organized retail theft, as offenders faced minimal legal consequences.

University of San Francisco psychologist Violet Cheung-Blunden and her colleagues wanted to see if fear-inducing media could sway voters in a deeply liberal area. They designed two experiments to test whether videos about local crime could alter decisions on Proposition 47. The researchers also wanted to know if this exposure would change the participants’ general political identities.

Free daily newsletter

In the first experiment, the research team recruited 206 adults from the San Francisco area during the week leading up to the election. The researchers asked the participants about their preexisting political beliefs, including their views on authority, immigration, and economics. The participants were then randomly assigned to watch one of two short videos.

Half of the group watched a video showing local tourist attractions, like the construction of the Golden Gate Bridge, to serve as a baseline comparison. The other half watched a compilation of news clips and online videos depicting organized looting and retail theft in the city. After watching the videos, the participants completed a mock exercise to revise Proposition 47.

The participants were asked to propose the ideal dollar amount for classifying retail theft as a felony. They also had to provide the minimum and maximum amounts they would be willing to accept on a ballot. Finally, the participants rated their current overall political orientation on a scale ranging from ultra-liberal to ultra-conservative.

The participants who watched the crime video proposed a markedly lower minimum and ideal dollar threshold for felony charges compared to the group that watched the tourist video. In practical terms, exposure to the crime footage prompted these voters to adopt a stricter, more punitive stance on retail theft. They became more conservative on the specific issue of criminal justice.

This shift toward a conservative policy position did not translate to a broader ideological change. The participants’ self-reported political orientation remained anchored to their initial beliefs, and they did not shift toward the political right overall. A single exposure to threatening media was not enough to uproot their overarching political identities.

The researchers also looked for evidence of political polarization. They checked whether the crime video caused initially liberal participants to become more liberal and initially conservative participants to become more conservative. The data did not support this idea, as the participants largely maintained their general ideological footing.

To ensure the reliability of these initial patterns, Cheung-Blunden and her team conducted a second experiment with 98 college students in San Francisco. This study followed a nearly identical procedure to the first one, but it took place in a controlled laboratory setting. The researchers also had all participants watch a five-minute breathing and relaxation video before viewing their assigned footage.

This relaxation step was intended to reduce baseline mood variations among the students. By calming the participants first, the researchers hoped to make any emotional reactions to the subsequent tourist or crime videos easier to measure. The results of the second experiment ended up mirroring the first.

The students exposed to the crime video proposed substantially lower dollar thresholds for felony theft across all three categories. Just like the community adults, the student participants embraced a tougher stance on crime after viewing the threatening footage. Once again, this shift toward a stricter crime policy did not alter the students’ broader political identities.

The crime video failed to move the student participants’ overall political orientation toward the right. The researchers also found no indication that the threatening footage caused the students’ general political views to become more polarized. The students’ overarching political ideologies remained stable, even as their position on a specific local policy shifted.

While the experiments show how media exposure can alter specific policy preferences, the results come with a few boundaries. The studies took place in a highly liberal city, meaning the participants likely sat further to the political left than the average American voter. The second experiment also included a high proportion of female participants, who generally reported more liberal leanings than male participants in the study.

These demographic skews mean the findings might not neatly apply to voters in more conservative or politically divided regions. Additionally, the researchers point out that their experiments measured political shifts using a mock policy threshold and a general ideology scale. Future research could incorporate a wider range of political choices, such as preferences for specific political figures in an actual election.

It is also difficult to know if the crime videos specifically triggered a deep fear of mortality, as terror management theory suggests. The footage might have simply provoked general anxiety about safety and social order. Future studies will need to directly measure the specific types of emotional discomfort that drive voters to change their minds on public policy.

The study, “Threat-Induced Conservatism in a Liberal Electorate: Issue-Based Shifts without Ideological Transformation,” was authored by Violet Cheung-Blunden, Serena Rajkumar, Yuka Hirata, Caelyn A. McDonough, and Chonhei Liao.