Bounding the Impact of a Gifted Program on Student Retention Using a Modified Regression Discontinuity Design
Published in: Annals of Economics and Statistics, no. 111/112, Special Issue on Education, July/Dec. 2013, p. 10-34
Posted on RAND.org on December 29, 2015
This paper examines whether gifted programs can help urban districts retain students. Gifted programs often employ IQ thresholds for admission, which creates strong incentives to manipulate the IQ score of students to increase access to the program. In the presence of local manipulation, the standard regression discontinuity estimator does not identify the local average treatment effect. However, we can construct a lower bound for the effectiveness of the program by exploiting a weak monotonicity assumption regarding the mean outcome in the untreated state. Our point estimates suggest that the gifted program has a strong positive effect on retention for higher income students.