The NCS was based on multistage area probability sample of people age 15–54 in the non-institutionalized civilian population of the coterminous U.S., with a supplemental sample of students living in campus group housing (Kessler et al., 1994). The survey was carried out face-to-face in the homes of respondents by professional interviewers employed by the Survey Research Center (SRC) at the University of Michigan. The survey was carried out between September 1990 and February 1992. The response rate was 82.6%, with a total of 8,098 respondents participating in the survey. A supplemental non-response survey was carried out to adjust for non-response bias, with a random sample of initial non-respondents offered a financial incentive to complete a short form of the diagnostic interview. A non-response adjustment weight was constructed for the main survey data to compensate for elevated rates of disorders found among the initial non-respondents in this non-response survey. Significance tests were made using design-based methods because of this weighting and clustering of the data (Kish & Frankel, 1970; Koch & Lemeshow, 1972; Woodruff & Causey, 1976). More details about design and weighting procedures are reported elsewhere (Kessler, Little et al., 1995).
Kessler, R.C. (2004), "GENDER DIFFERENCES IN MENTAL DISORDERS IN THE U.S. NATIONAL COMORBIDITY SURVEY", Marcotte, D.E. and Wilcox, V. (Ed.) The Economics of Gender and Mental Illness (Research in Human Capital and Development, Vol. 15), Emerald Group Publishing Limited, Bingley, pp. 7-25. https://doi.org/10.1016/S0194-3960(04)15002-6
Emerald Group Publishing Limited
Copyright © 2004, Emerald Group Publishing Limited