Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Using event history data, this paper analyzes the distribution of reemployment spell durations conditional on the completed length of the preceding (contiguous) spell of unemployment. The model is used to infer how unemployment insurance, through unemployment duration, may affect the likelihood of reentering unemployment. Special attention is paid to the endogeneity between unemployment and reemployment duration and the sample information contained in the fact that some individuals obtain reemployment by recall as opposed to acceptance of a new job. Copyright 1995 by MIT Press.