Improving sampling and response rates in children's health research through participatory methods.

BACKGROUND: Children's health is an important indicator of community health because children are especially vulnerable to disease. The school setting is ideal for assessing these vulnerabilities and prevalence of disease, yet the methods that produce high participation among students and their families are not usually described or evaluated. This is important because high participation rates increase reliability of data and interpretation of results.
METHODS: In order to determine which methods produce the highest response rates in a parent-report school-based survey, a child-focused protocol, consisting of a brief classroom presentation and incentives for participation, was compared to a parent/teacher outreach method and a teacher-only outreach protocol. Methods were implemented in schools matched on student demographics and neighborhood socioeconomic level. The child-focused method was found to yield the highest response rate; thus, it was deployed in 26 schools to determine its effectiveness in schools of varying socioeconomic levels.
RESULTS: Response rates were highest using the child-focused protocol when compared to those resulting from the parent/teacher outreach or the teacher-only protocols (80.5% vs 74.7%, 83.4% vs 45.8%, respectively). After full deployment, the response rate of the child-focused protocol was 76.9% (70.8-82.8%), with 46.9% of classrooms achieving a response rate of 80% or greater. The sample obtained was highly representative in racial/ethnic distribution, as evidenced by its strong correlation with the individual school enrollment, the 5- to 12-year-old population of surrounding ZIP codes, and the New York City public school population.
CONCLUSION: Participatory research methods that engage schoolchildren and school staff directly in the research process yield the highest response rates for school-based studies and provide educational benefits for the school community.