Construct validity of the Depression and Somatic Symptoms Scale: evaluation by Mokken scale analysis

Abstract

Ya-Hsin Chou,1 Chin-Pang Lee,1,2 Chia-Yih Liu,1,2 Ching-I Hung1,2 1Department of Psychiatry, Chang-Gung Memorial Hospital at Linkou, 2School of Medicine, Chang Gung University College of Medicine, Taoyuan, Taiwan Objective: Previous studies of the Depression and Somatic Symptoms Scale (DSSS), a free scale, have been based on the classical test theory, and the construct validity and dimensionality of the DSSS are as yet uncertain. The aim of this study was to use Mokken scale analysis (MSA) to assess the dimensionality of the DSSS.Methods: A sample of 214 psychiatric outpatients with mood and anxiety disorders were enrolled at a medical center in Taiwan (age: mean [SD] =38.3 [10.5] years; 63.1% female) and asked to complete the DSSS. MSA was used to assess the dimensionality of the DSSS.Results: All 22 items of the DSSS formed a moderate unidimensional scale (Hs=0.403), supporting its construct validity. The DSSS was divided into 4 subscales (Hs ranged from 0.35 to 0.67), including a general somatic scale (GSS), melancholic scale (MS), muscular pain scale (MPS), and chest symptom scale (CSS). The GSS is a weak reliable Mokken scale; the other 3 scales are strong reliable Mokken scales.Conclusion: The DSSS is a psychometrically sound measure of depression and somatic symptoms in adult psychiatric outpatients with depression or anxiety. The summed score of the DSSS and its 4 subscales are valid statistics. Further research is required for replication of the 4 subscales of the DSSS. Keywords: depression, somatization, Mokken scale analysis, item response theory, construct validity

Keywords

• depression
• somatization
• Mokken scale analysis
• item response theory
• construct validity.