Weighting Nonprobability and Probability Sample Surveys in Describing Cancer Catchment Areas.

BACKGROUND: The Population Health Assessment initiative by NCI sought to enhance cancer centers' capacity to acquire, aggregate, and integrate data from multiple sources, as well as to plan, coordinate, and enhance catchment area analysis activities. METHODS: Key objectives of this initiative are pooling data and comparing local data with national data. A novel aspect of analyzing data from this initiative is the methodology used to weight datasets from sites that collected both probability and nonprobability samples. This article describes the methods developed to weight data, which cancer centers collected with combinations of probability, and nonprobability sampling designs. RESULTS: We compare alternative weighting methods in particular for the hybrid probability and nonprobability sampling designs employed by different cancer centers. We also include comparisons of local center data with national survey data from large probability samples. CONCLUSIONS: This hybrid approach to calculating statistical weights can be implemented both within cancer centers that collect both probability and nonprobability samples with common measures. Aggregation can also apply to cancer centers that share common data elements, and target similar populations, but differ in survey sampling designs. IMPACT: Researchers interested in local versus national comparisons for cancer surveillance and control outcomes should consider various weighting approaches, including hybrid approaches, when analyzing their data.

Computer availability and principals' perceptions of online surveys(*).

BACKGROUND: School-based risk behavior surveys traditionally have been administered via paper-and-pencil. This study assessed the feasibility of conducting in-class online surveys in US high schools. METHODS: A paper-and-pencil questionnaire assessing computer availability and perceptions of online surveys was mailed to a nationally representative sample of public and private high school principals in fall 2008. Completed surveys were returned by principals from 580 of 704 selected schools. Prevalence and 95% confidence intervals were computed. RESULTS: Most high schools have at least 1 computer lab, most computers in computer labs are connected to the Internet, and just under half of schools with a stationary computer lab had a sufficient number of computers to accommodate an in-class online survey administration. The 2 most common problems associated with online surveys were logistics of providing enough computers for an entire class and rotating classes into computer labs. Nearly two thirds of principals preferred online to paper-and-pencil surveys when administered to 4 randomly selected classes that met at different times during the school day, but less than half reported this preference when administered to 4 randomly selected classes that met at the same time during the school day. CONCLUSION: Many schools do not have sufficient computer capacity to participate in a voluntary in-class online survey. An online survey could impose significant perceived and actual burden on schools and therefore could result in unacceptably low school participation rates. Online administration for in-class surveys of students in US high schools are not recommended.

Comparison of paper-and-pencil versus Web administration of the Youth Risk Behavior Survey (YRBS): risk behavior prevalence estimates.

The authors examined whether paper-and-pencil and Web surveys administered in the school setting yield equivalent risk behavior prevalence estimates. Data were from a methods study conducted by the Centers for Disease Control and Prevention (CDC) in spring 2008. Intact classes of 9th- or 10th-grade students were assigned randomly to complete a survey via paper-and-pencil or Web. Data from 5,227 students were analyzed using logistic regression to identify associations of mode with reporting of 74 risk behaviors. Mode was associated with reporting of only 7 of the 74 risk behaviors. Results indicate prevalence estimates from paper-and-pencil and Web school-based surveys are generally equivalent.

Methods: School Health Policies and Programs Study 2006.

BACKGROUND: The School Health Policies and Programs Study (SHPPS) 2006 examined 8 components of school health programs: health education, physical education and activity, health services, mental health and social services, nutrition services, healthy and safe school environment, faculty and staff health promotion, and family and community involvement. All 8 components were assessed at the state, district, and school levels. Two components, health education and physical education and activity, also were assessed at the classroom level. METHODS: Computer-assisted telephone interviews or self-administered mail questionnaires were completed by state education agency personnel in all 50 states plus the District of Columbia and among a nationally representative sample of school districts (n=538). Computer-assisted personal interviews were conducted with personnel in a nationally representative sample of elementary, middle, and high schools (n=1103), with a nationally representative sample of teachers of required health education classes or courses (n=912), and with a nationally representative sample of teachers of required physical education classes or courses (n=1194). RESULTS: This article provides a detailed description of the development of the questionnaires; sampling; data collection; and data cleaning, weighting, and analysis. CONCLUSIONS: SHPPS 2006 is the largest and most comprehensive study of school health programs ever conducted. Fielding a study of this magnitude provides many challenges, and several recommendations for future studies emerged from the experience.