Survey Statistics

Survey Statistics - People with maps

RTI is a leader in survey research and survey statistics. Augmenting our extensive program in survey research with unmatched statistical science capabilities, our survey statisticians are able to consult with clients on virtually every aspect of survey design and analysis and offer solutions tailored to specific client needs.

Capabilities

  • Sampling
    Frame construction, including stratification, clustering, and composite measures of size; optimum allocation, controlling for differential costs and variances; probability proportional to size sampling; probability minimum replacement sampling; controlled selection; multiphase (or double) sampling; multiframe sampling; adaptive sampling
  • Weighting
    Sensitivity analyses of potential survey biases; reduction of nonresponse and coverage biases; possible sampling-error reductions
  • Imputation
    Regression imputation; predictive mean neighborhoods; logical imputation; recursive partitioning to form imputation classes; multiple imputation
  • Nonsampling error analysis
    Nonresponse reduction and compensation; evaluation and reduction of survey error; measurement error modeling; reduction of data processing error
  • Small area estimation
    Benchmarked SAEs for large sample areas formed as aggregates of lower-level estimates; use of unit- and/or person-level and area-level predictors; geographic information mapping; Fay-Harriot SAE methodology; robust Hybrid Hierarchical Bayes (HHB) methodology for general mixed models with additive random coefficients
  • Statistical disclosure control
    Analyzing risk levels of variables; identifying and removing personally identifiable information; applying data coarsening techniques to continuous or categorical variables; data swapping by exchanging data between randomly selected individuals; assessing and limiting disclosure risk per HIPAA Safe Harbor and Expert Opinion options; implementing MASSC™ (RTI software that de-identifies data for public use files)

Projects

  • National Surveys on Drug Use and Health (NSDUH)
  • National Survey of Child and Adolescent Well-Being (NSCAW)
  • High School Longitudinal Study of 2009 (HSLS:09). HSLS:09 will follow 9th graders through their high school, postsecondary, and early career experiences, focusing on college decision-making and math learning using a new algebra assessment. Data are collected from students, administrators, math and science teachers, school counselors, parents, and administrative records.
  • Beginning Postsecondary Students Longitudinal Study (BPS). BPS follows students from when they first begin their postsecondary education. Initially, students in the NPSAS surveys are identified as being first-time beginners of undergraduate studies. These students are asked questions about their experiences during and transitions through postsecondary education and into the labor force, as well as family formation.
  • National Inmate Survey of Sexual Assaults (NIS). The National Inmate Survey, a congressionally mandated study funded by the Bureau of Justice Statistics, is measuring the incidence and effects of sexual assault in correctional facilities and determining characteristics of victims and perpetrators. RTI is responsible for developing a touch-screen ACASI instrument, sampling about 100,000 inmates in 450 facilities, collecting data, record abstraction, data processing, data analysis, and reporting.
  • Early Childhood Longitudinal Study-Birth Cohort (ECLS-B). ECLS-B is a longitudinal study intended to provide information that will inform policy regarding young children, their families and schools, and early care and education. The primary goal is to provide a comprehensive and reliable set of data that may be used to describe and better understand children's early development; their preparation for school; key transitions during the early childhood years; children's experience in early care and education programs, kindergarten, and the primary and elementary grades; and how their early experiences relate to their later development, learning, and experiences in school.
  • Baccalaureate and Beyond (B&B). This longitudinal study collects data from students who originally participated in NPSAS, asking questions about their job search activities, education, employment experiences after graduation, and the impact of postsecondary education on subsequent life experiences (employment, family formation, lifestyles).
  • National Study of Postsecondary Faculty (NSOPF). NSOPF provides data about faculty and instructional staff at public and private not-for-profit 2- and 4-year postsecondary institutions in the United States to postsecondary education researchers, planners, and policymakers.
  • Education Longitudinal Study: 2002 (ELS:2002). ELS:2002 is a longitudinal survey that monitors the transitions of a national sample of young people as they progress from tenth grade to the world of work. ELS:2002 initially obtained information from students, school records, parents, teachers, librarians, and school administrators and continues to collect data from the original students.
  • Occupational Information Network (O*NET). O*NET, is a comprehensive database of worker attributes and job characteristics. As the replacement for the Dictionary of Occupational Titles (DOT), O*NET is the nation's primary source of occupational information.