|
Activity Number:
|
511
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 7, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301149 |
|
Title:
|
Using Regression Discontinuity Design for Program Evaluation
|
|
Author(s):
|
Hyunshik Lee*+ and Tom Munk
|
|
Companies:
|
Westat, Inc. and Westat, Inc.
|
|
Address:
|
1650 Research Blvd., Rockville, MD, 20850,
|
|
Keywords:
|
Effect size ; power ; sample size ; random control trial ; relative efficiency
|
|
Abstract:
|
In many education and social intervention programs it is thought efficacious and ethical to offer a treatment to individuals based on a measure such as need or worthiness. This entails assigning subjects to the treatment group if they are at or above some cutoff score on the measure. The outcome of the treatment is then analyzed using the regression discontinuity design technique assuming that the treatment effect can be detected at the assignment cutoff through regression analysis. In this study the power of such a design is analyzed to determine the sample size necessary for implementing the parametric regression model under some common assumptions. Violations of those assumptions threaten the validity of the RDD study. Ways to address such violations are also discussed.
|
