Comparison of factor structure and psychometric properties of original and abbreviated version of the Penn State Worry Questionnaire in a nonclinical sample: a cross-sectional psychometric study

Abstract

Background and objectives: Worry as a main symptom of generalized anxiety disorder is a chain of repetitive and uncontrollable thoughts about possible negative events in the future. The Penn State Worry Questionnaire is one of the most widely used measures for assessing pathological worry. The purpose of the present study was to investigate the psychometric properties of the abbreviated version of the Penn State Worry Questionnaire-Abbreviated (PSWQ-A) and to compare it with the original version of the Penn State Worry Questionnaire (PSWQ) in a nonclinical sample. Participants and methods: The present study is a cross-sectional psychometric study. The statistical population of this study consisted of all students studying at Zanjan University of Medical Sciences (n = 3500) from January to October 2018. A sample of 350 people was selected for the study. Research measures were the PSWQ-A, PSWQ, and the Generalized Anxiety Disorder 7-item (GAD-7) Scale. The present study was carried out after approval of Social Determinants of Health Research Center of Zanjan University of Medical Science with the project code of A-12-924-5 on October 7, 2017. The project was also approved by the Ethics Committee of Zanjan University of Medical Sciences on October 17, 2017 and the approval ID was IR.ZUMS.REC.1396.187. Results: Exploratory factor analysis, scree plot, and parallel analysis supported the single factor structure PSWQ-A. The total variance explained by the single-factor model of PSWQ-A was higher (53.1% versus 49.1%). Generally, fit indices for the PSWQ-A was better fitted than the PSWQ. Both measures had acceptable convergent validity (r=0.52 for both questionnaires) and satisfactory internal consistency (α=0.87 for both questionnaires). Conclusion: PSWQ-A has better psychometric properties compared to PSWQ, and it can be used for faster and more accurate assessment of worry in psychological studies and therapeutic settings.