The 2014 Child Development Supplement (CDS) is a new, nationally-representative survey of U.S. children. It is part of the Panel Study of Income Dynamics (PSID), the world’s longest running nationally-representative household panel that began in 1968. Until 1997, when the original CDS was launched, PSID collected limited information on children. The original CDS collected information on a cohort of 3,600 children aged 0–12 years in 1997. Follow-up rounds of data collection on the same children were completed in 2002/03 and 2007/08.
By 2014, children from the original CDS cohort had reached adulthood, and PSID launched a successor survey that was designed to: collect information on all children in PSID households; convert the study from a cohort model to a panel of repeated cross-sections (of PSID children); and update the measures and modules. CDS-2014 fieldwork was conducted between October 2014 and April 2015. The target sample comprised approximately 5,700 children in 3,300 households. Information was collected in interviews with the children’s primary caregiver (typically the mother) and with older children (aged 12–17) who also are asked to complete a computerized interactive voice response interview covering sensitive topics (such as sexual behavior and substance use). The survey obtains anthropometric measurements, saliva samples, and consent for linkage with children’s school and vital statistics birth records. For a subsample, the survey also conducted cognitive assessments of PCGs and children aged 3+ years and collected time diaries.
We describe the rationale for the major design decisions that shaped the survey. For example, we explain the rationale for interviewing a random half of the households in-person, and explain the sampling scheme used to select these households. We then describe innovative components of the household instruments and their development, including the IVR interviews and the mail-out / mail-back protocols. Next, we describe the operational fieldwork plans, including interviewer selection and training, respondent incentives, informed consent, end-game strategies, and various problems and challenges that were encountered. We also briefly describe an experiment in which a random sample of 200 hard-to-reach respondents were offered a supplementary incentive to complete all survey components of a study during a three-week winter holiday period. Finally, we describe the overall response rates, differences in response rates by respondent characteristics, and variation in response across the different study components. We end by summarizing some of the major lessons we learned for future rounds of CDS and other major longitudinal studies of children.