Sponsored by the National Institute of Statistical Sciences
in conjunction with the SAMSI program on Latent Variable Models in the Social Sciences
Organizers
Paul Biemer
919-541-6056
ppb@rti.org
Jerry Reiter
919-668-5227
jerry@stat.duke.edu
The goal of the workshop is to bring together researchers from federal agencies, academia, and survey organizations to discuss methods for measuring nonsampling errors. The workshop will include invited presentations by distinguished researchers in survey methodology, particularly those specializing in the estimation of components of total survey error. These researchers will present a wide range of methods for quantifying components of total survey error. In addition, workshop participants will invited to discuss the types of nonsampling errors they experience in collecting and analyzing survey data and, therefore, seek to quantify. Note that, in the interest of preserving the focus of the workshop on estimation methods and issues, no discussion of the methods for reducing survey errors is planned.
We expect lively, open floor discussions to follow each presentation. Speakers will be given 30 minutes to present their ideas leaving 30 minutes for discussion and comment from the other participants. On Day 2 of the workshop following the last presentation, conference participants will be encouraged to join several discussion groups charged with identifying an agenda for future research for total survey error. Discussion leaders in each break-out group will then be given an opportunity to summarize the ideas of their groups and recommend what next steps should be taken to advance this field of research.
The workshop was held at the Bureau of Labor Statistics in Washington, DC, on March 17 and 18, 2005 (beginning at 8:30 AM on Thursday and ending by 5:00 PM on Friday).
Presentations
Total Survey Error: Past Present and Future, Robert Groves, University of Michigan
Why Should the Federal Government Care?, Clyde Tucker, Bureau of Labor Statistics
Nonsampling Error Research in Practice, J. Michael Brick and Graham Kalton, Westat
A Review of (Total) Survey Error Models, William Kalsbeek, University of North Carolina
Total Error in Surveys, Forecasts, and Randomized Social Experiments: Modeling Approaches, Bruce Spencer, Northwestern University
Seven Model Motivated Rules of Thumb or Equations, Fritz Scheuren, NORC
The Structural Equation Modeling approach, Willem Saris, University of Amsterdam
Discussion: Play-Acting at Science?, Robert Fay, U.S. Census Bureau
Bayesian Modeling of nonsampling error, Alan Zaslavsky, Harvard Medical School
Multilevel Models in Survey Error Estimation, Joop Hox, Utrecht University
Analyzing Survey Error with Latent Class Models, Paul Biemer, RTI International and University of North Carolina
Comments on "Analyzing Survey Error with Latent Class Models", Michael Larsen, Iowa State University
Other material contributed by workshop participants:
Some topics for research in TSE
Summary of research topics in TSE that were generated from breakout sessions at the conference.Suggested policy/practical goals for agencies related to TSE
Summary of policy/practical goals that were generated from breakout sessions at the conference.Example of a Total Survey Error study: Contributed by Bruce Spencer
This is an example of a total survey error study proposed for the 2000 A.C.E. A more general yet simpler version of the model will appear in chapter 10 of the forthcoming book by Alho and Spencer (2005).Statistical data editing bibliography: Contributed by Bill Winkler
Reference list related to statistical data editing. See also:
http://www.dis.uniroma1.it/~dq/dqcis/ -ICDT Workshop on Data Quality in Cooperative Information Systems 2003http://csaa.byu.edu/kdd03cleaning.html - ACM KDD Workshop on Data Cleaning, Record Linkage, and Object Identification 2003
http://cimic.rutgers.edu/~sigmod05/ and http://iqis.irisa.fr/ ACM SIGMOD Workshop on Information Quality in Information Systems 2005
Statistical record linkage bibliography: Contributed by Bill Winkler
Reference list related to record linkage.Data confidentiality bibliography: Contributed by Bill Winkler
Reference list related to data confidentiality.