Votes against voting: Recall escalation as behavioural evidence of electoral (mis)confidence in Taiwan’s “Great Recall”

This article examines public confidence in the results of democratic elections through Taiwan’s “Great Recall” in 2025, during which over half of the legislators were targeted by recall campaigns. Rather than relying on self-reported attitudes, the article conceptualises electoral (mis)confidence as a latent disposition whose behavioural implications can be inferred from institutionalised electoral contestation. The escalation of recall efforts is treated as a behavioural proxy for this latent construct: the further a recall advances, the more clearly it suggests a withdrawal of specific electoral support from an existing representative outcome. Methodologically, the analysis integrates supervised machine learning with Bayesian ordinal estimation. A LASSO regression identifies salient predictors among thirteen variables categorised as structural, behavioural, and discursive, while a Bayesian ordinal model estimates their probabilistic associations with recall advancement. The dependent variable is a four-level ordinal measure of recall progress. Explanatory variables include legislators’ gender, constituency type, party alignment, electoral margins, turnout volatility, and the discursive factors of volume and sentiment in traditional news coverage ( China Times and Liberty Times ) and new-media discussions (PTT and Threads). Results indicate that recall escalation is more likely in urban constituencies and when traditional media coverage is extensive and negative, but less likely for female and ruling-party legislators, and where media sentiment is more positive. The article argues that urban contexts and negative media environments are associated with higher levels of recall escalation, whereas ruling-party affiliation and gendered representational quality are associated with lower levels. The findings suggest how patterned recall escalation can function as a behavioural proxy for analysing electoral confidence under conditions of heightened political contestation.

Authors

Institutions

Publication Details

Journal
Electoral Studies
Published
2026-08-26
DOI
https://doi.org/10.1016/j.electstud.2026.103141
Primary Topic
Electoral Systems and Political Participation
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Votes against voting: Recall escalation as behavioural evidence of electoral (mis)confidence in Taiwan’s “Great Recall”

Jiru Zeng
Electoral Studies
Electoral Systems and Political Participation
article

Votes against voting: Recall escalation as behavioural evidence of electoral (mis)confidence in Taiwan’s “Great Recall”

Jiru Zeng
article en

Abstract

This article examines public confidence in the results of democratic elections through Taiwan’s “Great Recall” in 2025, during which over half of the legislators were targeted by recall campaigns. Rather than relying on self-reported attitudes, the article conceptualises electoral (mis)confidence as a latent disposition whose behavioural implications can be inferred from institutionalised electoral contestation. The escalation of recall efforts is treated as a behavioural proxy for this latent construct: the further a recall advances, the more clearly it suggests a withdrawal of specific electoral support from an existing representative outcome. Methodologically, the analysis integrates supervised machine learning with Bayesian ordinal estimation. A LASSO regression identifies salient predictors among thirteen variables categorised as structural, behavioural, and discursive, while a Bayesian ordinal model estimates their probabilistic associations with recall advancement. The dependent variable is a four-level ordinal measure of recall progress. Explanatory variables include legislators’ gender, constituency type, party alignment, electoral margins, turnout volatility, and the discursive factors of volume and sentiment in traditional news coverage ( China Times and Liberty Times ) and new-media discussions (PTT and Threads). Results indicate that recall escalation is more likely in urban constituencies and when traditional media coverage is extensive and negative, but less likely for female and ruling-party legislators, and where media sentiment is more positive. The article argues that urban contexts and negative media environments are associated with higher levels of recall escalation, whereas ruling-party affiliation and gendered representational quality are associated with lower levels. The findings suggest how patterned recall escalation can function as a behavioural proxy for analysing electoral confidence under conditions of heightened political contestation.

Electoral StudiesVol. 103
Durham University (GB)
Economic and Social Research Council
Gender equality
Openalex Percentile: Top 3%
Electoral Systems and Political Participation
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Votes against voting: Recall escalation as behavioural evidence of electoral (mis)confidence in Taiwan’s “Great Recall” — Jiru Zeng · Electoral Studies (2026) | TGRS Research Map | TGRS