THE CDS-2014 SAMPLE

The CDS-2014 sample—and the PSID sample more generally—is designed to be representative of the corresponding U.S. population of children and families. By design, PSID and CDS-2014 have certain gaps in coverage. In particular, PSID and CDS-2014 do not cover families in which both partners migrated to the U.S. after 1997 (when PSID last added a new immigrant refresher sample) and do not cover children in these post-1997 immigrant families. Note, however, that PSID is currently undertaking a new immigrant screening and recruitment operation in order to add a sample of post-1997 immigrant families and children to the sample in 2017. Children in these post-1997 new immigrant families will be eligible for participation in future waves of CDS.

The CDS-2014 sample eligibility criteria were defined as follows:

• Family participated in the 2013 Core PSID survey.

• Child’s reported birth year was 1997–2013.

• Child was classified as belonging to the PSID sample (i.e., has the “PSID gene”).

• Child was not classified as a household head or the spouse/partner of a household head.

• Child was not deemed to be eligible for the original CDS study (ER33418 ≠ 1).

A total of 5,816 children were deemed to have been eligible for a CDS-2014 interview based on these criteria. An additional 71 cases were released to the field but were later determined to have not been eligible for CDS-2014 and have been coded as ineligible.

The CDS-2014 fieldwork proceeded in several stages. Interviewers began by attempting to contact families with eligible children and completing a “Coverscreen” that collected information about the household composition and the identity of the CDS-2014 sample children’s primary caregiver (PCG; typically the mother). As shown in Table 3.1, below, for a number of cases (interviewers were unable to locate the family (68 children), exhausted allowed number of contact attempts to reach the family (244 children), or reached the end of the field period without contacting the family (567 children). After families were successfully contacted, respondents could refuse to participate (390 children) or, for one child, had a language barrier that prevented any interview from being conducted. Towards the end of the fieldwork period, a random subsample of eligible families were deselected for interviews (180 children), and these cases are classified as non-sample. Finally, there were 15 children not released to the field due to an office error and a final set of 16 children have not received a final classification yet.

In CDS-2014, information was collected on a total of 4,333 children from an eligible sample of (5,816 – 180 =) 5,636 children. The overall, unconditional response rate at the child level for CDS-2014 was 4,333 / 5,636 = 77 percent. Note that this response rate is not directly comparable to the CDS-I response rate because in CDS-I interviews were conducted immediately following the 1997 Core PSID interview and hence there was no initial screening non-response as occurred in CDS-2014. In a future revision of this User Guide, we will recalculate the CDS-2014 response rate to be comparable to the rate for CDS-I.

Children in CDS-2014 ranged in age from 0 to 17 years, as shown in Table 3.2 based on their year of birth and the dates of the CDS-2014 fieldwork. Note that children born in 2013 comprise a smaller share of the sample than births in other years. This occurred because eligibility for CDS-2014 was established based on each family’s completion of the Core PSID interview in 2013. Because the average date for the Core PSID interviews in 2013 was about half-way through the calendar year, only about half the births that occurred in 2013 were identified for CDS-2014. The remaining births in 2013 will be eligible for the next wave of CDS. Table 3.2 also shows that the CDS-2014 sample was divided approximately evenly between males and females.