Telephone surveys were conducted in 1988 on random samples of persons ages 40-74 years in two populations with relatively high rates of colorectal cancer, i.e., in Long Island (NY) and in Connecticut. Correct responses to questions concerning warning signs for and the frequency of colorectal cancer relative to stomach cancer were not related to age but were positively associated with educational level and were greater among females than males. Awareness of the possible role of dietary factors in cancer (i.e., regarding both cured meat and fiber-rich foods) was high, while daily consumption of bran cereals was infrequent. Proportions of respondents who reported having colorectal screening tests increased with age and were higher among college graduates and among those having annual medical check-ups (especially for college graduates). In multivariate analyses age (i.e., 60-74 vs 40-59 years) and education (i.e., college graduates vs all others), along with frequency of medical check-ups, were significant independent predictors of use of colorectal screening tests (both fecal occult blood and proctoscopy), while gender (female) was negatively associated with proctoscopy. Implications for planning intervention programs are discussed.