The objectives of the National Survey of American Life (NSAL) are to investigate the nature, severity, and impairment of mental disorders among national samples of the black and non-Hispanic white (n = 1,006) populations in the US. Special emphasis in the study is given to the nature of race and ethnicity within the black population by selecting and interviewing national samples of African-American (N = 3,570), and Afro-Caribbean (N = 1,623) immigrant and second and older generation populations. National multi-stage probability methods were used in generating the samples and race/ethnic matching of interviewers and respondents were used in the largely face-to-face interview, which lasted on average 2 hours and 20 minutes. The Diagnostic and Statistical Manual (DSM) IV World Mental Health Composite Interview (WHO-CIDI) was used to assess a wide range of serious mental disorders, potential risk and resilience factors, and help seeking and service use patterns. This paper provides an overview of the design of the NSAL, sample selection procedures, recruitment and training of the national interviewing team, and some of the special problems faced in interviewing ethnically and racially diverse national samples. Unique features of sample design, including special screening and listing procedures, interviewer training and supervision, and response rate outcomes are described. Copyright © 2004 Whurr Publishers Ltd.