Pointing out who stayed silent made students 17% more willing to voice unpopular views

In an experiment with 383 UC Berkeley students, those holding socially unpopular views self-censored more. Showing students how many peers had stayed silent corrected their beliefs about public opinion and made holders of unpopular views about 17% more likely to speak.

Illustration of a student discussion circle where the quiet students are highlighted and one begins to speak.
Share this:

College students with unpopular opinions keep quiet, and their silence makes the popular view look even more dominant than it is. But a simple nudge, reminding people how many others said nothing, can break that cycle.

That is the finding of an experiment with 383 UC Berkeley undergraduates by Yuen Ho of UC Berkeley’s Department of Agricultural and Resource Economics and Yihong Huang, now at Peking University’s Guanghua School of Management. Their working paper, “Breaking the Spiral of Silence,” is dated October 5, 2024.

Testing an old theory

The “spiral of silence” idea dates to the 1970s. People who hold views seen as socially inappropriate stay quiet, so the views that get aired skew toward the acceptable side. If listeners ignore the silence and treat what they hear as representative, they overestimate support for the acceptable view, which pushes even more people into silence.

In the experiment, students discussed four contentious topics in groups: renaming public schools named after controversial historical figures, considering race in college admissions, abolishing the death penalty, and requiring vaccinations on campus. Before the discussions, a survey recorded their private views and their guesses about what other participants thought.

A first group of students discussed the topics with no extra information. The rest then saw a summary of those first discussions. A control group saw how many first-round students had publicly agreed or disagreed. A treatment group saw the same numbers plus how many had stayed silent, for example 7 agreeing, 5 disagreeing and 13 saying nothing.

Both groups were told the total number of first-round participants, so the silent count could be worked out either way. The difference was only how visible the silence was.

Silence, made visible

The spiral showed up clearly. First-round students holding socially appropriate views were about 15% more likely to speak up than those holding inappropriate views. Fewer than 2% said something in public that contradicted their private view, suggesting people choose silence over lying.

Making silence visible changed both beliefs and behavior. Right after the summary, treatment students guessed the socially appropriate view was about 6% less common than control students did. In their own discussions, treatment students with socially inappropriate views were about 17% more likely to express them. Students who held the appropriate view were unaffected.

The gap then widened. By the end, treatment students estimated socially appropriate views to be 10% less prevalent than control students did. Control students drifted further toward overestimating the popular view, while treatment students’ beliefs became more accurate.

Beyond the lab

A structural model fitted to the data suggests the intervention raised attention to silence by around 40 percentage points. In a society that ignored silence entirely, the model predicts people would overestimate the popularity of the socially appropriate view by 50 percentage points.

The authors also ran a version using real public comments on proposed building renamings at Berkeley. Students told how many private comments had been sent alongside the public ones guessed that fewer peers backed renaming. Those who privately opposed renaming were 20% more likely to sign a public petition against renaming other campus buildings.

Study Details: