683 – Testing for Data Quality: Challenges in Practice
Identifying Patterns in Employer Reporting Errors in the BlS Survey of Occupational Injuries and Illnesses
Sara Wuellner
Washington State Department of Labor & Industries
Polly Phipps
Bureau of Labor Statistics
The Bureau of Labor Statistics (BLS) estimates of non-fatal work-related injuries and illnesses are based on employer reported data. Collected through the Survey of Occupational Injuries and Illnesses (SOII), these data are used to estimate and characterize work injury and illness for the US and most states. Increasingly, research suggests that employers underreport injuries in SOII, and that underreporting varies by establishment and injury characteristics, as well as by state. A follow-back telephone survey was conducted with SOII participants in four states to gather data on workplace injury and illness recordkeeping and reporting practices. Using the multi-state telephone survey data, we sought to identify within and between state differences in: compliance with injury recordkeeping requirements; knowledge of recordkeeping requirements; and sources of employer injury data (e.g., workers' compensation claims data and OSHA injury and illness records). Patterns in occupational injury and illness recordkeeping were explored through regression models. We identified common recordkeeping errors and establishment characteristics associated with such errors.