Populist Wave: How Japan's First Major Anti-Establishment Party Found Its Voters

What We Found

As the July 2025 Upper House election approached, many supporters of the Democratic Party for the People (DPP), a centrist party, moved to Sanseito, a new right-leaning party running against the political establishment.

On the surface, these voters looked pretty “mainstream”—they came from a centrist party, not the fringes. But their attitudes set them apart: low trust in institutions, heavy reliance on YouTube and social media, and openness to conspiratorial claims.

Why It Matters

The data suggest that the DPP had quietly acted as a buffer for voters unhappy with existing parties. When the DPP mishandled its candidate selection for the 2025 race, that buffer broke down. The result was a sudden, measurable flow of voters to Sanseito, which went on to win 14 seats and gain enough strength to influence the governing coalition.

Similar stories—where centrist parties lose frustrated supporters to newly organized populist forces—have been seen in Europe and the United States. This project offers one of the first real-time, individual-level views of that dynamic in Japan.

Visual Highlight

The Sankey diagram below turns voter flows into a concrete picture—showing how the same people switched parties over time, which snapshot polls can’t capture.

The chart highlights the 1,200 respondents who answered all three key waves.

How We Did It

High-frequency panel design

Working with Prof. Asako Miura, a social psychologist and professor at the University of Osaka with expertise in political psychology, we ran 10 online survey waves from early spring through election day. The questionnaire design relied heavily on her lab’s expertise. Respondents were drawn from an online panel and re-contacted each wave, creating a panel-like dataset that let us follow the same individuals over time.

Measuring attitude profiles

Each wave asked not only about party support and vote intention, but also media habits, trust in institutions, views on COVID-19 measures and vaccines, and openness to anti-elite messages. These variables were then used to compare the underlying profiles of DPP, Sanseito, and other party supporters.

Analysis and visualization

Using R, I reshaped the data into a person-by-wave format and counted how many people moved from one party to another at key moments. These counts were turned into a Sankey diagram to highlight flows out of the DPP toward Sanseito.

My Role

• Built the full R pipeline for cleaning, analyzing, and visualizing the panel survey data.
• Developed the Sankey diagram and other charts used in newsroom discussions and in the final series.
• Worked with politics reporters to turn the statistical patterns into narratives that readers could follow easily.
• Reported and wrote the full article series, integrating quantitative analysis with interviews and political context.
• Documented the workflow so that colleagues could review and reuse the methods in later elections.

Impact & Publication

• Findings ran as a 9-part series in The Asahi Shimbun, explaining who Sanseito’s voters were and how they emerged.
• The analysis informed election-night coverage and post-election explainers on shifts in Japan’s party system.
• The project helped establish a template for running similar panel surveys in later elections.