July 25, 2011
RTI International Statisticians Will Present New Research Findings to Guide Policy at International Conference
RESEARCH TRIANGLE PARK, N.C.—RTI International statisticians will present new research findings that enhance statistical methodology and guide public policy decisions at the annual American Statistical Association's Joint Statistical Meetings.
The conference will take place July 30–August 4 in Miami Beach, Fla., with more than 6,000 attendees. At the world's largest gathering of statisticians, RTI's leading researchers will present 15 papers and poster presentations.
These presentations will highlight RTI's efforts to inform sound public policy through the use of objective, independent statistical analysis.
RTI presentations at the conference will include:
- Comparison of 95% Confidence Interval for Intra-Cluster Correlation Estimates for Clustered Binary Data
- Nonresponse and Measurement Error: Relationship, Relative Magnitude, and Correction
- Additive Random Coefficient Models for Robust Small-Area Estimation
- Comparing Mixed Models, Pattern Mixture Models, and Selection Models in Estimating Treatment Effects of Quality of Life Data with Nonignorable Missing Data in Clinical Trial Study
- Variance Inflation Factors in the Analysis of Complex Survey Data
- Seeing Is Believing: Graphical Depictions of Low-Probability Outcomes for Eliciting Treatment-Related, Benefit-Risk Tradeoff Preferences
- WTADJX Is Coming: Calibration Weighting in SUDAAN When Unit Nonrespondents Are Not Missing at Random and Other Applications
- Uncertainty and Standard Errors in Agent-Based Disease Projections
- Using the DAG Jackknife to Measure the Variance of an Estimator in the Presence of Item Nonresponse
- A Markov Latent Class Analysis of the National Crime Victimization Survey
- A Comparative Assessment of Disclosure Risk and Data Quality Between MASSC and Other Statistical Disclosure Limitation Methods
RTI also has six statisticians recognized as American Statistical Association Fellows.
In the exhibit hall, RTI will showcase the SUDAAN® statistical software package developed by RTI researchers, which provides both descriptive and modeling procedures for analyzing survey and other cluster-correlated data encountered in epidemiological studies, clinical trials, experimental studies and longitudinal studies.