Communicative coherence in management accounting: theory and practice in an automated pipeline

Resumo

In light of the relevance, text genre, and signaling theories, this study investigates the extent to which communicative coherence (CC) between the titles and abstracts of management accounting (MA) articles varies by subfield and journal tier. There is a lack of theoretical and methodological consensus on how to measure CC in MA and link it to impact metrics in a comparable way. The evidence is fragmented and heterogeneous in terms of definitions and procedures, which prevents cumulative syntheses. To address this issue, we developed an automated metric and analyzed 13,614 MA articles indexed in Scopus from 1995 to 2025. We applied natural language processing techniques in an automated pipeline. Methodologically, the study constructed a thematic dictionary of MA and developed a replicable protocol for semantic analysis to ensure transparency, reproducibility, and adherence to open science principles. Our analysis revealed that 8.7% of articles exhibited high CC, 62.9% exhibited medium CC, and 28.4% exhibited low CC. Articles from established subfields and those published in Q1–Q2 journals tended to have a higher number of citations associated with high CC. However, the differences between the medium and high levels were not statistically significant. These results suggest that textual clarity is a sign of the field’s discursive maturity, establishing CC as an indicator of integrity and readability in academic writing. Our findings contribute to improving editorial and educational practices by demonstrating that consistency between the title↔abstract tends to facilitate understanding of the study’s scope and guides screening processes without directly affecting interest, visibility, or citations. This practical contribution strengthens Sustainable Development Goal 4 (SDG 4) (Quality Education) by promoting the clear, accessible, and socially engaged dissemination of scientific knowledge.