Framing Unionization on Facebook: Communication around Representation Elections in the United States

TL;DR

This study investigates how labor unions’ Facebook communications around representation elections relate to election outcomes by integrating NLRB election data (2015–2024) with 158,238 union posts. It introduces a RoBERTa-based multi-label classifier to detect five discourse frames (Diagnostic, Prognostic, Motivational, Community, Engagement) and analyzes pre- and post-election framing dynamics, with distinctions between won and lost cases. The authors provide an annotated dataset and publicly available classifier, revealing systematic framing differences by union type (industrial vs craft) and outcome, and showing a shift from mobilization toward consolidation after victories. The findings contribute to theories of framing and union revitalization in the digital era and offer practical guidance for tailored online communication strategies, alongside open data that enables further micro-level analysis of digital labor movements.

Abstract

Digital media have become central to how labor unions communicate, organize, and sustain collective action. Yet little is known about how unions' online discourse relates to concrete outcomes such as representation elections. This study addresses the gap by combining National Labor Relations Board (NLRB) election data with 158k Facebook posts published by U.S. labor unions between 2015 and 2024. We focused on five discourse frames widely recognized in labor and social movement communication research: diagnostic (identifying problems), prognostic (proposing solutions), motivational (mobilizing action), community (emphasizing solidarity), and engagement (promoting interaction). Using a fine-tuned RoBERTa classifier, we systematically annotated unions' posts and analyzed patterns of frame usage around election events. Our findings showed that diagnostic and community frames dominated union communication overall, but that frame usage varied substantially across organizations. In election cases that unions won, communication leading up to the vote showed an increased use of diagnostic, prognostic, and community frames, followed by a reduction in prognostic and motivational framing after the event--patterns consistent with strategic preparation. By contrast, in lost election cases unions showed little adjustment in their communication, suggesting an absence of tailored communication strategies. By examining variation in message-level framing, the study highlights how communication strategies adapt to organizational contexts, contributing open tools and data and complementing prior research in understanding digital communication of unions and social movements.

Figures (8)

• Figure 1: Relative usage of discourse frames by labor unions compared to the baseline median (percentage difference over baseline). Red color indicates a usage higher than the baseline, blue color indicates a usage lower than the baseline. The dendrogram shows the clustering of unions based on their framing patterns across the five frame types. Union acronyms are colored according to their organizational structure.
• Figure 2: Usage of frames leading to the election, comparing lost and won cases, in terms of mean and 95% confidence interval. Values on the y-axis refer to detrended scores. The $\star$ sign refers to the significance of the test for difference in means between losing and winning: $\star$ indicates significance at $\alpha=0.05$, $\star\star$ at $\alpha=0.01$, and $\star\star\star$ at $\alpha=0.001$.
• Figure 3: Changes in frame usage before and after elections, comparing lost and won cases. Bold values in the cells refer to the difference between proportion of pattern presence for a given frame between losses and wins. Raw counts of union-case instances assigned to each cell are reported for losses (L) and wins (W). The $\star$ sign indicates a significant difference between losing and winning at $\alpha=0.05$.
• Figure C1: Usage of frames leading to elections, for lost and won cases. Results are reported in terms of mean and 95% confidence interval. Values on the y-axis refer to detrended scores. The $\star$ sign refers to the significance of the test for difference in means between the losing and winning: $\star$ indicates significance at $\alpha=0.05$, $\star\star$ at $\alpha=0.01$, and $\star\star\star$ at $\alpha=0.001$.
• Figure C2: Usage of frames post-elections, for lost and won cases. Results are reported in terms of mean and 95% confidence interval. Values on the y-axis refer to detrended scores. The $\star$ sign refers to the significance of the test for difference in means between the losing and winning: $\star$ indicates significance at $\alpha=0.05$, $\star\star$ at $\alpha=0.01$, and $\star\star\star$ at $\alpha=0.001$.