Event History Data and Survey Recall: An Analysis of the National Longitudinal Survey of Youth 1979 Recall Experiment
Prior to its switch from an annual interviewing format to a biennial one, the National Longitudinal Survey of Youth 1979 (NLSY79) conducted an experiment to see how less frequent interviews would affect data quality. This paper analyzes this experiment with respect to data collected using event history techniques, namely AFDC and food stamp recipiency and employment history. Respondents faced with the longer recall period failed to report short spells of recipiency, employment, and nonemployment. The pattern of the coefficients in econometric models of the type often used to study event history did not change greatly, though tests reject the equality of the coefficients between the recall sample and the control group in half of the models estimated.