Digital propaganda via social media and perceived democratic debate in the digital public sphere: evidence from Jordanian journalists during the 2024 Jordanian parliamentary elections

Authors:

Abstract
This study examines how platform-specific digital election propaganda is associated with Jordanian journalists’ evaluations of democratic debate in the digital public sphere concerning the 2024 Jordanian parliamentary election. The framework integrates digital-public-sphere theory, platform affordances and network media logic, connective action, and affective-publics theory. Perceived richness of democratic debate is represented by three related evaluative components: rational-deliberative quality, affective-inclusive communication, and a global criterion, each measured with seven-point multi-item scales informed by the Discourse Quality Index and online-deliberation research. Construct validation included exploratory and confirmatory factor analyses, reliability assessment, convergent and discriminant validity, common-method diagnostics, ordinal bivariate tests, multivariate decomposition with demographic controls, subgroup comparisons, and platform-specific models. The analytic dataset comprised 340 complete cases. Facebook and WhatsApp had the highest exposure levels. The seven-factor measurement model demonstrated excellent fit (CFI = .992, TLI = .991, RMSEA = .014, SRMR = .039). Dialogic affordances were positively associated with rational-deliberative quality (Spearman rho = .498, p < .001); active engagement and affective-inclusive communication were positively associated with the global richness criterion (rho = .454 and.309, respectively, p < .001); and active engagement was associated with campaign influence (rho = .437, p < .001). The rational-deliberative by affective-inclusive interaction was not significant. A hierarchical decomposition model accounted for 46.4% of the variance in the global criterion. Because the criterion and the dimension-level scores belong to the same perceptual construct family, this value is interpreted descriptively rather than as evidence of an independent causal mechanism.

This publication has 21 references indexed in Scilit: