The current study presents a Rasch-derived short form of the Center for Epidemiologic Studies-Depression scale (CES-D) for use as a depression screening tool in the general population. In contrast to short forms developed with reliance on classical measurement techniques, those developed using techniques based on item response theory produce a measure that offers true interval scaling, provide enhanced information about responders with extreme scores, and expand understanding of the underlying latent structure. Cross-validation of the Rasch-derived CES-D short form supported its utility and structural validity across samples. Tests of structural validity using latent variable modeling methodology indicated that a hierarchical, single-factor model of depression had the best fit for the original full form and the Rasch-derived short form of the CES-D. This finding challenges depression researchers and theorists to reconsider the interfactor relationships in the study and assessment of depression.

One-hundred-four women who met Research Diagnostic Criteria for nonpsychotic, nonbipolar Major Affective Disorder were treated in 21 therapy groups for 10 weekly 1 j hr sessions. Three versions of the self-control therapy program for depression, one with a behavioral target, one with a cognitive target, and one with a combined target, were used to treat 7 groups each. AH conditions improved significantly and equally on self-report and clinician rating scales of depression. All conditions improved equally on measures of both behavioral and cognitive target variables, and initial level on these variables was not related to outcome.The self-control therapy program for depression is based on a model (Rehm, 1977(Rehm, , 1981a(Rehm, , 1982 that assumes that depression can be characterized as a series of specific deficits in selfmanagement behavior. This highly structured, group-format therapy program has been developed and evaluated as a research tool. Two studies (Fuchs & Rehm, 1977;Rehm, Fuchs, Roth, Kornblith, & Romano, 1979) validated the program against control conditions, and two additional studies disassembled the program package to evaluate its separate self-management components (Kornblith, Rehm, O'Hara, & Lamparski, 1983;Rehm et al., 1981). The program has been used by other researchers in disassembly (Tressler & Tucker, 1980) and comparison studies (Fleming & Thornton, 1980;Roth, Bielshi, Jones, Parker, & Osborn, 1982). The program has also been extended to other populations, including community mentalhealth-center patients (Glanz & Dietz, 1980), psychiatric inpatients (Kornblith & Greenwald, 1982

No abstract

No abstract

No abstract