Bureau of Justice Statistics, USA, December 28, 2015
Research and Development Papers
Assessing the Coverage and Reliability of Subnational Geographic Identifiers in the NCVS Public-Use File (NCJ 249467) and Evaluation of Direct Variance Estimation, Estimate Reliability,
and
Confidence Intervals for the National Crime Victimization Survey (NCJ 249242)
These 2 reports are now available on BJS.gov. Our new Research and Development (R&D) Papers series details statistical methods that will be applied to analyzing and reporting official findings from BJS's data collection programs.
Assessing the Coverage and Reliability of Subnational Geographic Identifiers in the NCVS Public-Use File examines the coverage and reliability of the NCVS sample in the subnational geographic areas that can be created from the public-use files by combining Census region, population size, and urbanicity.
Evaluation of Direct Variance Estimation, Estimate Reliability, and Confidence Intervals for the National Crime Victimization Survey examines the feasibility of using direct variance estimation for the National Crime Victimization Survey. It compares generalized variance function estimates to two direct variance estimation methods (Taylor Series Linearization and Balanced Repeated Replication).