A study examined the project methodology used by the Bureau of Labor Statistics and the related projections made by the state employment security agencies. Findings from a literature review indicated that the system has steadily improved since 1979. Projections made from the Occupational Employment Statistics Surveys were remarkably accurate. Suggestions from the literature for the improvement of the projection methodology were: (1) follow-up studies for collecting information to use in developing allocation ratios; (2) improvement of projection information on replacement needs; (3) improvement of substate projections; (4) inclusion of all employment sectors in projections; (5) collection of information on entry-level occupations; (6) inclusion of a statistical indication of the accuracy of projections; (7) consideration of earnings when planning programs; and (8) improvement of communications between producers and users of employment projection information. Semistructured interviews were conducted with individuals within Arizona who had an intimate knowledge of employment projection methodology information. They indicated that confidentiality restrictions limited the availability of certain information at the county level and the lack of information on replacement needs due to occupational mobility limited demand data. The major problem was users who did not understand how to use the labor market information for decision making. (55 references.) (YLB)