The National Occupational Hazard Survey (NOHS) was conducted from 1972 to 1974 to provide estimates of the number of workers potentially exposed to job hazards, to describe industries and occupations where these potential hazards were observed, and to identify medical, industrial hygiene, and safety practices used to protect workers. One of the major results of the NOHS was a list of chemical ingredients found in 65,000 out of 80,000 trade name products found in a probability sample of all United States workplaces. The NOHS data are used as an input for setting NIOSH research priorities. The data are also sent to OSHA where they are used for part of the regulatory priority setting mechanism for controlling workplace hazards. The NOHS also contributes to identifying high risk industries, occupations, or workers exposed to such classes of chemicals or carcinogens, mutagens, and teratogens. The NOHS data have been used to prepare computer generated county maps that depict potential workplace hazards. A national profile was developed by extrapolating from the NOHS sample of 5000 workplaces and a commercial file of 4.7 million worksites in the United States coded according to the Standard Industrial Classification (SIC) code. A new survey is now underway called the National Occupational Exposure Survey (NOES). The authors discuss problems that have been encountered in their attempts to make occupationally related exposure assessments: exposures were not measured in either the NOHS or the NOES; there is a great deal of variability within any SIC group; linking of health effects data to occupational exposure that may have occurred many years ago in a highly mobile population; and the surveys are extremely costly to conduct.