Development of a brief form of the Chinese version of the Odor Awareness Scale(OAS-B): A confirmatory factor analysis approach to item reduction

Abstract

Objective The Odor Awareness Scale is a questionnaire to assess individual differences in awareness of odors in the environment, the Chinese version of which has a long length with 27 items, and its psychometric properties do not meet an excellent level. Hence, this study aimed to develop a brief form of the Chinese version of the Odor Awareness Scale(OAS-B) with sufficient psychometric properties. Materials and methods A total of 978 participants from a college were recruited, who were randomly allocated into two groups for item reduction(N = 492) and scale validation(N = 486). Confirmatory factor analysis was utilized to remove items identified as information overlaps. After removing overlapping items, the brief scale was validated in the other subgroup for reliability and validity. The criterion-related validity was measured by correlation analysis between OAS-B and the Social Odor Awareness Scale(SOS), the Body Odor Sniffing Questionnaire(BOSQ), and the Generalized Anxiety Disorder Scale (GAD-7). Results Removing 12 overlapping items, OAS-B holds 15 items with three subscales. The model fit indices for the OAS-B were: χ2/df = 2.594, RMSEA = 0.057, CFI = 0.938, TLI = 0.925, SRMR = 0.046. The Cronbach’s α coefficient for OAS-B was 0. 868. The Pearson correlations between OAS-B and SOS, BOSQ, and GAD-7 were 0.518 (p < 0.01), 0.410 (p < 0.01) and 0.165 (p < 0.01). Conclusion OAS-B has a short length with sufficient psychometric properties, which could be applied to measure odor awareness in general population.