RETROSPECTIVE BIAS IN THE DISPLACED WORKER SURVEYS

The Displaced Worker Survey of the U.S. Bureau of Labor Statistics understates the number of workers displaced over the preceding five years covered by the survey by approximately one-third. The bias results from the fact that respondents' memory of displacement erodes over time. Memory loss is not randomly distributed across demographic groups, thereby making analyses based on these survey data suspect. The note reports memory-loss adjusted estimates of the number of displaced workers between 1979 and 1989 based on Displaced Worker Surveys for 1984, 1986, 1988, and 1990.