Rainer W Alexandrowicz1, Stefan Fritzsche, Ferdinand Keller. 1. Abteilung für Angewandte Psychologie und Methodenforschung, Institut für Psychologie, Alpen-Adria-Universität Klagenfurt, Universitätsstraße 65, 9020, Klagenfurt, Österreich, rainer.alexandrowicz@aau.at.
Abstract
BACKGROUND: The revised version of the Beck Depression Inventory (BDI-II) is a broadly used instrument for assessing the severity of depression of adolescents of at least 13 years of age and adults. The self-assessment questionnaire contains 21 polytomous items and follows the criteria for a major depression specified in the DSM-IV. Clinical samples have often been used to analyze the psychometric properties of the instrument primarily with factor analytic methods. METHODS: The present study performs a psychometric analysis in a non-clinical sample in order to ascertain, whether the instrument performs equally well with the different kinds of samples. A clinical sample and a sample of students filled in the questionnaire. A partial credit model was applied and parameter estimates and model fit of the two samples were compared. RESULTS: Threshold parameters and model fit largely agreed, however some items exhibited characteristic deviations. Nevertheless, person parameter estimates notably agreed in both samples. CONCLUSIONS: These results indicate that the BDI-II performs in clinical and non-clinical samples comparably well, only some items show characteristic deviations in the non-clinical sample.
BACKGROUND: The revised version of the Beck Depression Inventory (BDI-II) is a broadly used instrument for assessing the severity of depression of adolescents of at least 13 years of age and adults. The self-assessment questionnaire contains 21 polytomous items and follows the criteria for a major depression specified in the DSM-IV. Clinical samples have often been used to analyze the psychometric properties of the instrument primarily with factor analytic methods. METHODS: The present study performs a psychometric analysis in a non-clinical sample in order to ascertain, whether the instrument performs equally well with the different kinds of samples. A clinical sample and a sample of students filled in the questionnaire. A partial credit model was applied and parameter estimates and model fit of the two samples were compared. RESULTS: Threshold parameters and model fit largely agreed, however some items exhibited characteristic deviations. Nevertheless, person parameter estimates notably agreed in both samples. CONCLUSIONS: These results indicate that the BDI-II performs in clinical and non-clinical samples comparably well, only some items show characteristic deviations in the non-clinical sample.
Authors: Neusa Sica da Rocha; Eduardo Chachamovich; Marcelo Pio de Almeida Fleck; Alan Tennant Journal: J Psychiatr Res Date: 2012-10-13 Impact factor: 4.791
Authors: Adam B Smith; Robert Rush; Lesley J Fallowfield; Galina Velikova; Michael Sharpe Journal: BMC Med Res Methodol Date: 2008-05-29 Impact factor: 4.615
Authors: Mary Princip; Miriam Koemeda; Rebecca E Meister; Jürgen Barth; Ulrich Schnyder; Hansjörg Znoj; Jean-Paul Schmid; Roland von Känel Journal: Health Psychol Open Date: 2015-06-29