We evaluated the newly available National Institute for Occupational Safety and Health (NIOSH) National Occupational Hazard Survey (NOHS) job exposure matrix (JEM) by considering mesothelioma risk from asbestos exposure. We applied this system (NOHS-JEM) to the Los Angeles County Cancer Surveillance Program (CSP) to see how many cancer cases could be assigned asbestos exposure and how asbestos exposure affected mesothelioma risk. Using the same CSP data, our "experts" classified asbestos exposure simply by occupation and industry. Both exposure classifications were divided into low and high; the NOHS-JEM by the number of exposed people per couplet, and ours by judgements of intensity. Odds ratios (OR) for mesothelioma risk for low and high asbestos exposure for the NOHS-JEM were 2.0 (95% C.I. 1.2-3.4) and 2.5 (95% C.I. 1.2-4.8). For ours, corresponding risks were 1.6 (95% C.I. 1.1-2.4) and 6.3 (95% C.I. 2.5-15.1). Our system was able to assign more cases to couplets then the NOHS-JEM (35,895 to 22,369). Three limitations of the NOHS-JEM were that many occupation-industry couplets were not classified at all, many couplets associated with past asbestos exposure (before the 1972-1974 NOHS survey) were not classified as asbestos exposure, and no assessment of intensity was made. These limitations may apply to other exposures and should be carefully considered before the NOHS-JEM is applied to other case-control studies.