AI TrackerFrontier desk · AEST
Back to news
normaltech

semanticfa: An R Package for Response-Free Semantic Factor Analysis of Psychometric Scales

arXiv cs.CL · 28 Oct 2026, 02:18 AEST

The R package semanticfa facilitates response-free semantic factor analysis of psychometric scales by utilizing item embeddings. It interprets factors, identifies redundant items, and screens new items, enhancing the analysis of constructs like personality traits.

Key facts

Read aloud

The R package semanticfa facilitates response-free semantic factor analysis of psychometric scales by utilizing item embeddings. It interprets factors, identifies redundant items, and screens new items, enhancing the analysis of constructs like personality traits. The items of a reflective psychometric scale can be read as a proposal about a construct's structure. We introduce semanticfa, an R package for semantic factor analysis, which factors item-embedding similarities. A worked example analyzes the fifty-item International Personality Item Pool Big-Five Factor Markers, embedded with Qwen3-Embedding-8B, and compares the results with a factor analysis of eight hundred seventy four thousand four hundred thirty four respondents. With five factors, a twenty five-item short form matched the domains better than all two hundred random five-per-domain sets, but parallel analysis kept only three. We propose semantic-behavioral coherence, the agreement between semantic and response structures, as a measurable property of a scale, given an embedding model.

Sources in this cluster