text analysis
Definition
Text analysis refers to the computational examination of words, syntactic structures, and semantic relationships embedded within written material, used here to extract psychological meaning from psychometric scale items. In the study of depression, anxiety, and stress, item texts are processed through Textual Forma Mentis Networks, which reconstruct syntactic links and synonym-based connections among concepts without relying on simple word co-occurrence counts. This approach combines AI-based syntactic parsing with psychological valence data to identify which emotion-conveying words cluster around key ideas across an entire questionnaire. Applied to the 42-item Depression Anxiety and Stress Scale with responses from 39,775 individuals, text analysis of this kind yielded semantic communities whose distribution across psychometric factors was non-random, matching EGA-derived factor structures at significance levels up to p < 0.001.
Sources: Stanghellini et al. (2024)
Related Terms
- network psychometrics (1 shared article)
- cognitive network science (1 shared article)
- network clustering (1 shared article)
- psychometric measurements (1 shared article)
- semantic framing (1 shared article)
Applications
Text Analysis and Psychometric Factor Analysis
Text analysis of questionnaire items produces semantic communities, clusters of syntactically or semantically related concepts, whose overlap with statistically derived psychometric factors can be quantified through semantic loadings. When applied to the DASS, these semantic loadings identified specific aspects of emotional dysregulation, emotional exhaustion, physical distress, and tension states within EGA-based factors, demonstrating that the verbal content of items and the numerical ratings they elicit are systematically linked.
Sources: Stanghellini et al. (2024)
Text Analysis and Cognitive Network Science
Textual Forma Mentis Networks situate text analysis within the broader framework of cognitive network science, where nodes represent individual concepts expressed as words and edges encode syntactic specification or semantic overlap. This framing allows the structure of knowledge embedded in psychometric item texts to be visualised and measured in the same graph-theoretic terms used to model human associative memory, connecting the formal analysis of written items to the study of semantic and autobiographical memory.
Sources: Stanghellini et al. (2024)
