Academic journal article Journal of School Health

Strategies to Maximize Retention of a Sample of Young Adolescents in a Longitudinal Evaluation of Healthy & Alive!

Academic journal article Journal of School Health

Strategies to Maximize Retention of a Sample of Young Adolescents in a Longitudinal Evaluation of Healthy & Alive!

Article excerpt

Over the past decade, educators and public health professionals sought to increase education to reduce behaviors that threaten students' health.[1-3] Behaviors targeted for school-based interventions include sexual activity, smoking, alcohol and drug use, and violence.[4-7] While one advantage of school settings is that it provides a "captive audience," sample retention in longitudinal studies can prove challenging. Retaining the cohort can be especially difficult in public schools in low-income communities, with high rates of student mobility, troubled infrastructure, overburdened staff, and insufficient resources.

Loss of data from subjects who drop out can threaten the internal validity of longitudinal studies if sample loss differentially affects experimental and control conditions.[8-10] The external validity also may be seriously threatened when lost subjects possess characteristics closely related to the outcomes being studied.[8,11-13]

Progress has been made in developing statistical methods to adjust for loss of students.[14-16] However, because completely satisfactory methods are not available to account for underlying mechanisms that cause missing data, researchers often concede that it is optimal to minimize attrition in the first place.[14,17,18]

Evaluation literature provides information about expected attrition rates and strategies for retaining subjects in longitudinal samples.[12,13,19,20] Most references involve studies with adult populations. However, such studies may not be appropriate for younger populations who are more difficult to track because they usually move with their primary contacts (ie, parents), they often cannot provide reliable information about other contacts, and they cannot be tracked through motor vehicles departments.[11] Published intervention studies rarely describe reasons for attrition. Overall retention rates of longitudinal school-based evaluation studies with adolescent samples vary considerably. A meta-analysis of 85 cohorts in school-based studies found the mean retention rate of students was 81% at three months after baseline, 78% at six months, 73% at 12 months, and 72% at 24 months.[12] Time between measurements explained less than 5% of the variance in attrition. Retention declined most steeply at the earliest follow-up survey and gradually declined thereafter.

A review of seven longitudinal studies (1-9 years) on adolescent drug use revealed retention rates of 66% to 94%.[21] In more recent school-based prevention studies, Siegel et al[22] retained only 61% of students in grades 7-9 at a three-month follow-up. Pirie et al[23] retained 83% of a baseline sample of middle-grade students at both five-year and six-year follow-ups. The 1988 National Educational Longitudinal Study, NELS:88, which offered monetary incentives to teachers for assisting with student tracking, kept an extraordinary 94% and 91% of the baseline sample at the two-year and four-year follow-up surveys, respectively.[24,25] NELS:88 researchers estimated the cost per student of conducting comprehensive tracking and retention strategies was about 10 times the cost of surveying students without these procedures.[25]

Assessment of threats to external validity by comparing retained and lost students is infrequently reported. Ellickson, Bianca, and Schoff[11] found that students who were chronically absent or had dropped out of school typically possessed characteristics closely associated with risk-related outcomes. Students lost to follow-up in an evaluation involving students in grades 7-9 had higher rates of tobacco and alcohol use, and sexual activity than did retained students.[26]

This paper presents strategies implemented to maximize student retention during a longitudinal cohort study of Healthy & Alive!, a two-year intervention implemented in grades 7 - 8 to help prevent infection by the human immunodeficiency virus and other sexually transmitted diseases (HIV/STD). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.