Comparing data quality of fertility and first sexual intercourse histories

This paper evaluates the data quality of two demographic variables in light of hypotheses on respondent recall from the literature on survey methodolo gy. An emerging consensus in this literature is that recall of the timing o f an event declines with recall duration unless the dating of an event is f requently ''rehearsed.'' We provide empirical evidence consistent With this hypothesis by assessing the quality of demographic data on two event histo ry variables as supplied by female respondents. A first outcome concerns th e interval between a first and second birth. We assess examine birth interv als using birth registration data from the Viral Statistics on Natality (VS N) and individual-level survey data from the 1990 June Current Population S urvey (CPS), the 1979-93 waves of the National Longitudinal Survey of Youth (NLSY), and the 1988 National Survey of Family Growth (NSFG). Overall, We find relatively little variation in the quality of birth interval data acro ss these four surveys, with one exception-CPS data in which responses have been allocated. A second demographic variable is age at first sexual interc ourse. We engage in several analyses of this variable. First, we use NLSY d ata to analyze discrepancies between successive reports on age (to the near est year) at first inter course. Second, we analyze a form of partially mis sing data (respondent inability to recall the calendar month of intercourse ) that occurs in both the NLSY and IVSFG. Third, we identify NLSY responden ts who. in successive interviews, give contradictory reports about whether or not sexual ac tivity had been initiated Our findings suggest that data q uality varies significantly with duration of recall and with measures of re spondent ability related to arithmetic facility and memory. Observed differ ences by race and ethnicity narrow substantially, when controlling for thes e and other background factors. We find evidence for a nonlinear associatio n between duration of recall and data quality, with similar patterns occurr ing in both the NLSY and NSFG. Finally, our NLSY results are suggestive of a pattern in which recent initiation of sexual activity may be concealed by respondents.