UM18-06: Validity and Reliability of Machine Coding of Occupation & Industry in the Health and Retirement Study
Many longitudinal studies, such as the Health and Retirement Study (HRS), use census definitions to code the occupation and industry of the jobs held by respondents. The codes change every 10 years with each new census. Researchers interested in tracing occupational characteristics or exposures over time, or generally understanding job ladders and transitions, create crosswalks to harmonize the differences in codes over time. A more accurate approach would apply the current codes to historical information. Hand-coding of occupation and industry, as has been done in the past at the HRS, is time-consuming and expensive. This project proposes to explore the feasibility of using the National Institute for Occupational Safety and Health (NIOSH) Industry & Occupation Computerized Coding System (NIOCCS) to code historical occupation and industry data in the HRS.
- Feasibility and Reliability of Automated Coding of Occupation in the Health and Retirement Study (Research Brief)
- Feasibility and Reliability of Automated Coding of Occupation in the Health and Retirement Study (Working Paper)