Making Words Count: Computational Linguistics in Management Research

In recent years, a rapidly growing number of management scholars are using computational linguistics (CL) to analyze vast corpora of naturally occurring organizational language. Yet the rapid adoption of CL methods has outpaced the development of shared standards for linking theoretical constructs, linguistic data, and computational measurement. We synthesize 353 articles published in leading management journals for 2013–2025, identifying four recurring challenges shaping the credibility of CL-based research: misalignment between constructs and textual corpora, unnecessary model complexity, the opacity of large language model workflows, and insufficient transparency in reporting. Building on these patterns, we develop four integrative design principles—construct-language fit, minimal sufficient complexity, LLM sensitivity and auditability, and disclosure-as-assessability—and identify key research implications that reposition CL as a theoretically consequential lens for management scholarship.