Value Added Estimation in the Presence of Missing Data
Niu Gao () and
Anastasia Semykina ()
Additional contact information
Niu Gao: Public Policy Institute of California
No wp2017_08_02, Working Papers from Department of Economics, Florida State University
Ignoring missing data may introduce biases into value-added estimation. We consider a model where both correlated student effects and idiosyncratic factors may cause selection biases. Moreover, we model selection as a function of school-level factors (e.g. charter proximity) that may have distinct effects on teachers in different traditional public schools. We discuss a correction procedure and study its performance using simulations. We find that correction tends to produce noisier estimates of teacher productivity, but can help to substantially reduce the bias. Using observational data from North Carolina we find that corrected estimates of teacher productivity are very similar to those produced by OLS, which appear to be due to small partial effects of school-level covariates on the probability of selection.
Keywords: value-added estimation; missing data; selection correction (search for similar items in EconPapers)
JEL-codes: C33 I20 (search for similar items in EconPapers)
References: Add references at CitEc
Citations: Track citations by RSS feed
Downloads: (external link)
https://coss.fsu.edu/econpapers/wpaper/wp2017_08_02.pdf First version, 2017-08 (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:fsu:wpaper:wp2017_08_02
Ordering information: This working paper can be ordered from
Access Statistics for this paper
More papers in Working Papers from Department of Economics, Florida State University Contact information at EDIRC.
Bibliographic data for series maintained by Dmitry Ryvkin ().