Document Type

Paper

Publication Date

4-2026

Abstract

This follow-up study from Montclair State University researchers in the Joetta Di Bella and Fred C. Sautter III Center for Strategic Communication in the College of Communication and Media shows (again) that the volume of political posts on X (formerly Twitter) may not reflect actual voter behavior. The gap between post volume about candidates and real election votes was 420.7% in the analyzed case of the April 12, 2026, election in Hungary. This reflects an even greater gap than the Center’s previous study of the New Jersey gubernatorial election where X posts favored the losing candidate by a margin of nearly 40%. Similarly, contributing factors to this particular gap may include the volume of specific ideological accounts, bot-based activity, and overall platform infrastructure/algorithms. Results identified 798,038 posts in the month preceding the election mentioning Péter Magyar with 4,155,106 posts that mentioned Viktor Orbán in the period studied. Sentiment analysis showed similar levels of favorability between the two candidates in X posts. Despite this, final vote totals for the election favored Magyar’s Tisza party by more than double. Findings of this study and the previous study of the New Jersey election indicate that the generation of posts for a candidate on X (formerly Twitter) may no longer correlate expected electoral outcomes.

Publisher

Joetta Di Bella and Fred C. Sautter III Center for Strategic Communication

Download

Included in

Communication Commons

Share

COinS