Use of employee interviews to estimate worker populations: comparison of multiple recapture methods.
Fischbach-T; Wittes-J; Dacey-E; Shilling-S
Proceedings of the Survey Research Methods Section of the American Statistical Association. Alexandria, VA: American Statistical Association, 1983 Jan; :1-7
Information about the size of worker populations is very useful for occupational health surveillance. For example, when the number of workers exposed to a suspected hazard can be reliably estimated, rates of illness, injury, and/or death can be estimated to obtain epidemiological evidence of the associated occupational risk. Plant records can be used to determine the number of workers when these have been maintained. However, very often such records are either not available for the entire period of interest or else they are too incomplete to use to estimate the desired population at risk. An overlooked source of information for estimating worker population size is the current workforce. Currently employed workers, when interviewed individually or in groups, can provide lists of both current and former co-workers which might be useful for estimation of the desired or target population. Recognizing the potential of this source of information, the National Institute for Occupational Safety and Health (NIOSH) has attempted to develop and test multiple recapture methods to analyze worker generated data for population at risk estimation.
Information-systems; Information-retrieval-systems; Surveillance-programs; Epidemiology
Proceedings of the Survey Research Methods Section of the American Statistical Association