Objectives. To determine the effect of stage and health locus of control (HLC) on willingness to accept local and systemic therapy and willingness to participate in clinical trials. Methodology. This was a cross-sectional survey of female breast cancer patients at the Greenebaum Cancer Center. Respondents had the option of completing the questionnaire using a computer or a paper version. Willingness to accept therapy was determined by the minimum number of cancer-free years respondents required to accept therapy. Clinical data were abstracted from the respondents' medical records; stage was classified as either "early" (stages 0--2) or late (stages 3 and 4). HLC was assessed using the Multidimensional Health Locus of Control, Form C. Tobit models were used to determine the association between stage and HLC on willingness to accept therapy and to participate in clinical trials. Results. Among 79 respondents, the mean age was 56.0 years (SD +/- 9.43); 58.2% identified themselves as white; 75.6% had early stage cancer. Respondents required the most cancer-free years, a median of 4.0, to accept mastectomy, 3.5 to participate in a clinical trial, and 1.0 to accept breast conserving therapy, chemotherapy, or tamoxifen RTM. Late stage patients were more willing to participate in clinical trials than early stage patients. Late stage patients were less willing to take tamoxifen; late stage patients in good physical health were less willing to have mastectomies. Patients with a higher internal locus of control (IHLC) were more willing to accept mastectomy or chemotherapy, and more willing to participate in clinical trials. Respondents with less education, those with a higher chance health locus of control (CHLC) were more willing to participate in clinical trials. Conclusions. Late stage patients were less willing to accept therapy which may be perceived as a less aggressive or effective form of therapy such as tamoxifen, but are more willing to accept experimental therapy. Patients with a higher IHLC appear to prefer the therapies that are shorter in course and require less follow-up. Respondents with a higher CHLC accept therapies that "leave nothing to chance" and appear to have a greater understanding of the concept of randomization.