|
Activity Number:
|
108
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Social Statistics Section
|
| Abstract - #303662 |
|
Title:
|
Model versus Randomization-Based Inference from Group Randomized Trials
|
|
Author(s):
|
Lisa M. LaVange*+ and David M. Murray*+ and Jennifer Hill*+ and Keith E. Muller*+ and Stuart Kerachsky*+
|
|
Companies:
|
The Ohio State University and The University of North Carolina at Chapel Hill and New York University and University of Florida and U.S. Department of Education
|
|
Address:
|
137 East Franklin St., Chapel Hill, NC, 27514, B222 Starling Loving Hall, Columbus, OH, 43210, 246 Greene St., 3rd Floor, New York, NY, 10003, 1329 SW 16th Street Room 5125, Gainesville, FL, 32610-0177, Institute of Education Sciences, Washington, DC, 20006,
|
|
Keywords:
|
Confidence intervals ; Generalizability
|
|
Abstract:
|
Consider a group-randomized school trial of schools, where either schools or classrooms are randomized, there is a strong set of covariates at multiple levels, and where the outcomes are a mixture of binary items, Likert agreement scale items, and standardized cognitive assessment scores. The panel will discuss alternative approaches to the analysis. Particular attention will be paid to the implications of different approaches for the communication of study results to broad audiences. Specific questions for the panelists: What language would you recommend for the press release and/or executive summary on how to interpret any error bands that you would generate in the analysis? How would you describe the generalizability of the results (again to a nonstatistical audience)? An education policy official will comment on the communications aspects of the solutions suggested.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
