The high cost of mental health surveys of the general population has sparked interest in less costly research methods. Two low-cost mental health survey strategies (mail and telephone) were compared in terms of cost, response rate and quality of data obtained. A total of 1,074 persons agreed to participate in the study as a sample, one-half by telephone and the other half by mail. They completed the Diagnostic Interview Schedule Self-Administered, a questionnaire designed to be self-administered, which was used to assess specific mental disorders and to evaluate risk factors. In addition, 239 respondents who were selected according to the presence or absence of specific diagnoses were reinterviewed face-to-face using the Diagnostic Interview Schedule as an external criterion. The telephone method yielded a better response rate (15% higher) and better control over answers (for example, less missing data). The mail strategy was less expensive and appeared to yield data of slightly better quality, particularly for respondents suffering from anxiety disorders.