As part of the Honolulu Heart Program studies, 4,581 men of Japanese ancestry living in Hawaii completed a psychosocial questionnaire in 1971. By 1980, 280 new cancer cases had been recorded. Among 12 items related to stressful life situations, there were no consistent associations with total or site-specific cancer incidence. Most of the associations which were found were in the direction opposite to that predicted by the stress hypothesis. Among eight items related to social networks, two were significantly associated with total cancer incidence in multivariate analysis, but one of these associations was in the direction opposite to that of the social support hypothesis. Among the different measures of acculturation, there was a general pattern indicating that ties to traditional Japanese culture and diet were associated with an increased risk of cancer. Cancer incidence was associated with low levels of socioeconomic status. In general, there was little evidence to support the idea that either stressful or supportive life situations were associated with cancer incidence.