Measurement Invariance Analysis on the Revised Cheek and Buss Shyness Scale in a U.S. and Japanese College Sample

Abstract

In the present study, we tested the factor structure of the 20-item Revised Cheek and Buss Shyness Scale (RCBS; Cheek & Melichor, 1985) in a U.S. sample and used measurement invariance analysis to explore its cross-cultural validity in a previously collected Japanese sample (Sato et al., 2018). We additionally measured related personality traits (Big Five; Locus of Control, LOC; and Behavioral Inhibition System/Behavioral Activation System, BIS/BAS) to examine whether each predicted shyness. Participants were undergraduate students in the United States (N = 525) and Japan (N = 508). Exploratory factor analysis (n = 261) revealed a unidimensional structure for shyness to be the best fit. Measurement invariance analysis (n = 256 for United States) showed that the RCBS did not establish configural invariance, the weakest level in measurement invariance, suggesting that it may not be a cross-culturally valid and reliable tool to assess shyness. We thus did not perform any cross-cultural comparisons of variables, as initially intended. Results from a hierarchical multiple regression in the U.S. sample foundextraversion to negatively predict shyness, β = −.59, p < .001, and neuroticism, β = .18, p < .001, LOC (external), β = .14, p < .001, and BIS, β = .19, p < .001, to positively predict shyness. Gender was not a significant predictor, β = .09, p = .075. Future work may focus on the development of a Japanese shyness measure independent from U.S. conceptualizations to improve the cross-cultural validity and reliability of shyness measures.