A questionnaire dealing with sex role related household task performance was administered to 284 individuals who were randomly selected from the Memphis (Tennessee) City Directory. The questionnaire was designed to explore both attitudes regarding who "should" carry out 29 different household tasks, and who actually carried out each task in the households of the individuals interviewed. The worker categories included: woman, man, either sex, shared by both sexes, boy, girl, and all. Data were also collected on race, church affiliation, education and, in some cases, income. Sixty-three percent of the respondents were black and 37 percent were white. The results confirm that assumptions made by sociologists about sex roles are essentially valid: twenty out of 29 tasks have a modal response that assigns the task to one sex or another. There is some indication, however, that the women's movement has caused changed in sex role expectations, with both a smaller proportion of tasks being performed by only women (40 percent, as opposed to a 1968 estimate of 75-85 percent), and a significant proportion of tasks being shared by both sexes or all family members. (GC)