Following an Adolescent Cohort into Adulthood:
Reestablishing a Longitudinal Study for an Unanticipated Add-on Wave of Data Collection
Abstract for the 2006 Methodology of Longitudinal Surveys International Conference
Marjorie Hinsdale and Brian J. Burke, RTI International
Implementing a third wave of a longitudinal study, originally designed to track adolescents over a two-year period, created unique challenges. With no plans for a third wave, no attempts were made to gather future contacting information at the conclusion of the second wave for the cohort of over 20,000 sample members. Implementing a tracing plan to track the cohort five to seven years after the last contact proved to be a dynamic undertaking. The contact information that was available was for parents and other relatives of the sample members who had been ages 11 to 18, living at home and attending secondary school during the previous waves. At the time of the third follow-up, the sample members were 18 to 25 and primarily not living with parents at the last known address. Another hurdle to be overcome was the impact of a study design, which involved beginning each follow-up wave with the sample from the initial wave rather than only following the respondents from the previous wave. For Wave III, this meant going back to the Wave I respondents including those who had not been interviewed in Wave II. While this protocol helped maintain the large scale of the panel, it meant that contact information for part of the sample was older than it was for participants in the second wave.
Thesefindings are based on experience from the National Longitudinal Study of Adolescent Health, Wave III, which was a government-sponsored national survey conducted in the United States in 2001 to 2002. This paper documents the most significant issues and provides insight into the most effective solutions to the challenge of locating and interviewing a large sample of young adults. A wide variety of tracing activities were implemented before the beginning of the third follow-up wave, including: post card mailings with requests for change of address, batch data base searches, and centralized interactive database searches. Once data collection began, additional tracing was still required for a significant percentage of cases. Traditional means of tracing a cohort were found to be less effective for the young adults in the study sample. Uniquely designed tracing protocols based on the sample population itself were required to reach the targeted locating and response rates.
In particular, there were two innovative tracing protocols that proved to be effective. Tapping into the design of the initial school-based sample, in-field tracing protocols were developed that targeted the sample members themselves as a tracing resource. The Respondent-Assisted Tracing System (RATS) was developed as an interactive database that the sample members could scroll through at the conclusion of the interview to identify schoolmates and provide fresh tracing leads when known. Based on a similar concept, but using a different protocol, sibling cases were linked in such a way that even when traced to different locations across the United States, field interviewers would have a way to obtain updated locating information from siblings that had been located. Surprising findings related to the tracing success rates for the cohort that had been lost in the second wave demonstrate that the demographics of the study population may be more important than the age of the locating information itself in some cases.
With plans for a fourth wave of follow-up data collection well underway, a final section of this paper is devoted to a discussion of additional plans to be implemented for Wave IV that will further ensure successful tracing and high response rates for this complex and ambitious longitudinal study.
Marjorie Hinsdale
RTI International
3040 Cornwallis Road
Research Triangle Park, N.C.U.S.A.
Phone: (919)541-7368 / Fax: (919)541-7250
Brian J. Burke
RTI International
3040 Cornwallis Road
Research Triangle Park, N.C.U.S.A.
Phone: (919)541-6851 / Fax: (919)361-3867