Discovering semantic structures in psychological scales: A response-free front-end to efficient simplification

Psychological scale refinement traditionally relies on response-based methods such as factor analysis, item response theory, and network psychometrics to uncover latent constructs and optimize item composition. Although methodologically rigorous, these approaches require large respondent samples and may be limited by data availability, cross-cultural comparability, and replication constraints. Recent advances in natural language processing suggest that the wording of questionnaire items contains construct-relevant semantic organization, offering a complementary response-free perspective. We introduce a topic-modeling framework that operationalizes item-level semantic structure to systematically simplify scales. We encode items using contextual sentence embeddings and group them through density-based clustering to identify semantic topics without predefining their number. Within each cluster, class-based term weighting identifies representative keywords that summarize each topic and support an interpretable semantic pre-structure of the item pool. The framework then hierarchically merges semantically related clusters and selects representative items using structure-aware membership criteria within an integrated reduction pipeline. We evaluated the framework on three widely used instruments: DASS, IPIP, and EPOCH, covering negative affect, personality traits, and well-being. Results showed that the identified semantic topics often formed factor-like groupings aligned with established constructs. Representative items selected through semantic membership probabilities reduced item counts by 60.5% on average while maintaining acceptable psychometric properties and preserving factor correlation patterns. These findings support semantic topic-structure modeling as an interpretable front-end for theory-consistent scale construction and simplification, guiding rather than replacing subsequent psychometric validation. An integrated visualization-supported tool further enables one-click semantic analysis and structured item selection.