Psychometric Evaluation of the Social Phobia Inventory (SPIN) in Algeria: A Comprehensive Approach Utilizing Network Analysis, Confirmatory Factor Analysis, and the Polytomous Rasch Model.

Abstract

This study aimed to evaluate the psychometric characteristics of the Social Phobia Inventory (SPIN) by employing network analysis, confirmatory factor analysis, and the Polytomous Rasch Model. A cross-sectional data set was collected comprising 1,530 participants, with 959 being women and 571 being men. The Bootstrap Exploratory Graph Analysis unveiled the presence of two dimensions, with Items 17, 15, 5, 14, 6, and 9 exhibiting the highest strength centrality index. Notably, the Network Comparison Test indicated no differences in Network Invariance and global strength between the networks of women and men. Furthermore, the confirmatory factor analysis results demonstrated that the two extracted dimensions displayed an acceptable goodness of fit. In addition, the reliability coefficient values were acceptable, exceeding the threshold of 0.70. The Rasch analysis results suggested an overall fit, but some items exhibited overlap, suggesting their potential removal. Furthermore, it was recommended to develop new items to address gaps between existing items, particularly for measuring the lower levels of Social Anxiety Disorder. In conclusion, these findings provide robust evidence supporting the reliability and validity of the SPIN as a tool for measuring Social Anxiety Disorder in Algeria.