This article examines the association between male circumcision and HIV infection in a national sample. The analysis is based on the 2003 Kenya Demographic and Health Survey (KDHS), a nationally representative household-based population survey of adults, in which male respondents self-reported their circumcision status. In addition, in some households eligible for individual interview, blood samples were subsequently anonymously obtained for HIV testing, making this the first study linking socio-demographic information to HIV status at the national level. The study sample is limited to 3 413 men aged 15–54 years who gave valid information on their circumcision and HIV statuses. Nearly 5% of the men were HIV-positive, and 86% had been circumcised. HIV prevalence was significantly higher among the uncircumcised men (12%) than among the circumcised men (3%). This indication of the protective effect of male circumcision on HIV infection remained statistically significant (OR 0.15; 95%CI: 0.09–0.23) even after controlling for the effects of socio-demographic variables, age at first sexual intercourse, and use of paid sex. Based on these results, we recommend that HIV-prevention advocates and activists, scholars, bio-medical communities and political leaders find ways