Search Results

The second essay asks whether the effect of father presence varies by maternal age in two adolescent outcomes, math achievement scores and early cigarette smoking. The increased incidence of first births among women age 30 or over suggests study of this population as researchers have investigated births to younger women. Positive effects of father presence are found for a range of child and adolescent outcomes among children at all birth timings, but theories of parenting and empirical differences in older and younger mothers suggest that the father presence effect may vary by maternal age. Using data from the National Longitudinal Survey of Youth, 1997 Cohort, I find that the effect of father presence increases with maternal age for both math achievement and deferred smoking. The change is attributable to observable parental characteristics that are correlated with maternal age rather than a pure effect of maternal age.

Respondents of the NLSY97 were born in the years 1980-84 and were 16 to 20 years old in the fourth round of annual interviews. Although the response rates for the first three rounds of interviewing were above 90 percent, the response rate for Round 3 was slightly lower than that of Round 2. The payment incentive during the first three rounds was constant at $10 and was paid by the field interviewer at the completion of the interview. Concerned over the decline in responses rates and its implication for future rounds, the sponsor of the study, the Bureau of Labor Statistics (BLS), applied for and received approval from the Office of Management and Budget (OMB) to conduct an experiment manipulating the timing of payment and the level of respondent fee paid. The interviews for the NLSY97 are conducted by the National Opinion Research Center (NORC) of the University of Chicago under contract to the BLS. The experiment had two treatments: (1) payment of the fee prior to scheduling the in-person interview; and (2) increases in respondent fees.