THE GERMAN VERSION OF THE CENTER FOR EPIDEMIOLOGIC STUDIES DEPRESSIONSCALE AS AN INSTRUMENT IN GERONTOLOGICAL RESEARCH

Citation
M. Riediger et al., THE GERMAN VERSION OF THE CENTER FOR EPIDEMIOLOGIC STUDIES DEPRESSIONSCALE AS AN INSTRUMENT IN GERONTOLOGICAL RESEARCH, Zeitschrift fur klinische Psychologie, Psychiatrie und Psychotherapie, 46(4), 1998, pp. 344-364
Citations number
44
Categorie Soggetti
Psycology, Clinical
ISSN journal
14318172
Volume
46
Issue
4
Year of publication
1998
Pages
344 - 364
Database
ISI
SICI code
1431-8172(1998)46:4<344:TGVOTC>2.0.ZU;2-B
Abstract
The applicability of the German version of the Center for Epidemiologi cal Studies Depression Scale (CES-D) in gerontological research was in vestigated. The German CES-D shows satisfactory psychometric propertie s for the use as research instrument. It has a high internal consisten cy. The German CES-D is applicable as screening instrument for detecti on of subjects with ''Depressive Disorders'' and ''Major Depression''. It is, however, concluded that the so far common practice of using a unitary cutting point in different age-by-sex groups is not warranted for many gerontological research questions. Goodness indices of severa l potential cutting points are reported for the total sample and sever al sub-samples. These results are based on data of a representative sa mple of the at least 70 years old West-Berlin population and can, ther efore, be used as a guideline for choosing relevant cutting points in investigations of old subjects. Confirmatory factor analysis was condu cted testing the four-factor-structure repeatedly replicated in invest igations of the English CES-D. The four-factor-solution fits the data in our sample satisfactorily well. This result indicates measurement e quivalence of the English and German CES-D. It is concluded that both versions are applicable in cross-cultural comparisons of elderly perso ns. Although the current German CES-D exhibits a sufficient quality as research instrument, several items showed unsatisfactory item propert ies in our sample.