Psychometric properties of Geriatric Depression Scale (GDS) among elderlies in Tehran using multidimensional Rasch model

The purpose of this study was to examine the psychometric properties of the Geriatric Depression Scale (GDS) when applied to the elderly of Tehran. This research is applied-developmental, descriptive and quantitative.

The research population was Tehrani elderlies, among which 400 people responded to the Geriatric Depression Scale voluntarily and conveniently. The gathered data on sample’s demographic characteristics and the findings regarding the research items were analyzed by SPSS 18 and R-3.3.3 software.

The results of exploratory factor analysis and confirmatory factor analysis yielded four factors in the questionnaire. Having determined the unidimensionality of each dimension and examined the item fit—applying Rasch model—and removed misfit items, the psychometric properties of the remained items were estimated by application of the multidimensional Rasch model. The reliability of the questionnaire was estimated by Cronbach's alpha and the reliability coefficient of the multidimensional Rasch model, both of which showed the questionnaire is of adequate reliability. The validity of the Geriatric Depression Scale was examined twice investigating the convergent validity (using the Beck Anxiety Inventory and Beck Depression Inventory) and the construct validity, both of which showed adequate and acceptable validity.The results of examining differential item functioning in men and women groups led to the removal of two items, and in groups of different education levels led to the removal of another two items. The results of reexamining the item fit using the Rasch model showed excellent item fit when the 4-dimensional model was applied. Finally, 21 items out of 30 items of the questionnaire were kept.

The results of this study showed that the Geriatric Depression Scale, after eliminating inappropriate questions, is a valid and reliable tool which can be used as a acceptable test in Tehran.