Network analysis of the foreign language classroom anxiety scale

The Foreign Language Classroom Anxiety Scale (FLCAS) is among the most widely used instruments for assessing anxiety in the foreign language classroom, yet prior factor-analytic studies have produced inconsistent solutions ranging from two to seven components across different samples and target languages. The present study applies network psychometrics to examine the structure of the FLCAS. This approach has been used productively to examine anxiety constructs in clinical and educational psychology, but not yet in the context of foreign language learning. Using responses from 481 undergraduate students enrolled in Spanish courses at a U.S. university, we constructed a network in which nodes represented individual FLCAS items and edges reflected the strength of partial correlations between item pairs. We address three research questions: (1) Which FLCAS items are most central to the anxiety network as measured by closeness centrality? (2) Do FLCAS items cluster into theoretically meaningful communities? (3) What does the resulting network structure reveal that conventional psychometric approaches do not? Micro-level analysis identified item 26 (general tense feelings in class) as the most central node, consistent with prior item-total correlation findings, and highlighted several speaking-related items as high-centrality targets for educational intervention. Meso-level community detection identified three communities corresponding to psychophysiological and behavioral responses, negative beliefs and outcomes, and language-learning confidence. These findings illustrate how network psychometrics can complement conventional approaches to scale analysis and offer practical guidance for educators seeking to reduce foreign language classroom anxiety and improve student outcomes.