Patrick Dattalo
- Published in print:
- 2009
- Published Online:
- February 2010
- ISBN:
- 9780195378351
- eISBN:
- 9780199864645
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195378351.003.0003
- Subject:
- Social Work, Research and Evaluation
This chapter describes the following alternatives and complements to RS in terms of their assumptions, implementations, strengths, and weaknesses: (1) randomization tests; (2) multiple imputation; ...
More
This chapter describes the following alternatives and complements to RS in terms of their assumptions, implementations, strengths, and weaknesses: (1) randomization tests; (2) multiple imputation; and (3) mean-score logistic regression. Randomization tests are statistical alternatives to RS. Multiple imputation is a statistical supplement RS. Mean-score logistic regression is a statistical alternative or supplement to RS.Less
This chapter describes the following alternatives and complements to RS in terms of their assumptions, implementations, strengths, and weaknesses: (1) randomization tests; (2) multiple imputation; and (3) mean-score logistic regression. Randomization tests are statistical alternatives to RS. Multiple imputation is a statistical supplement RS. Mean-score logistic regression is a statistical alternative or supplement to RS.
Myoung-jae Lee
- Published in print:
- 2016
- Published Online:
- August 2016
- ISBN:
- 9780190258733
- eISBN:
- 9780190258771
- Item type:
- book
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780190258733.001.0001
- Subject:
- Economics and Finance, Econometrics
This monograph reviews the three most popular methods in applied economics and other social sciences: matching, regression discontinuity, and difference in differences. This book also examines ...
More
This monograph reviews the three most popular methods in applied economics and other social sciences: matching, regression discontinuity, and difference in differences. This book also examines related methods such as weighting, regression imputation, complete pairing, and triple difference. The book introduces the underlying econometric/statistical ideas for these methods, shows what is identified and how the identified parameters are estimated, and then illustrates how they are applied with empirical examples. The volume emphasizes how to implement the methods with data; many data and programs are provided in the online appendix that will be updated over time. The minimum requirement for the book is basic econometrics and statistics including least squares estimator, and the target audience is upper‐level undergraduate students, graduate students and researchers on self‐study. Despite this, all readers—theoretical econometricians/statisticians, applied economists/social scientists, and researchers/students—will find something useful in the book from different perspectives.Less
This monograph reviews the three most popular methods in applied economics and other social sciences: matching, regression discontinuity, and difference in differences. This book also examines related methods such as weighting, regression imputation, complete pairing, and triple difference. The book introduces the underlying econometric/statistical ideas for these methods, shows what is identified and how the identified parameters are estimated, and then illustrates how they are applied with empirical examples. The volume emphasizes how to implement the methods with data; many data and programs are provided in the online appendix that will be updated over time. The minimum requirement for the book is basic econometrics and statistics including least squares estimator, and the target audience is upper‐level undergraduate students, graduate students and researchers on self‐study. Despite this, all readers—theoretical econometricians/statisticians, applied economists/social scientists, and researchers/students—will find something useful in the book from different perspectives.
Myoung-jae Lee
- Published in print:
- 2016
- Published Online:
- August 2016
- ISBN:
- 9780190258733
- eISBN:
- 9780190258771
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780190258733.003.0003
- Subject:
- Economics and Finance, Econometrics
Matching is not the only way to control for covariates. Weighting divides each group response by the covariate-conditioned group choice probability to identify the treatment and control group means ...
More
Matching is not the only way to control for covariates. Weighting divides each group response by the covariate-conditioned group choice probability to identify the treatment and control group means separately. Regression imputation integrates out the covariates in each group’s regression function to get the two group means separately. Complete pairing forms a weighted average of all possible pair differences across the two groups with the weight depending on the covariate difference. Treatment effects for sample selection models arise when the treatment can affect participation in an activity and the performance there; separating these effects has interesting policy implications.Less
Matching is not the only way to control for covariates. Weighting divides each group response by the covariate-conditioned group choice probability to identify the treatment and control group means separately. Regression imputation integrates out the covariates in each group’s regression function to get the two group means separately. Complete pairing forms a weighted average of all possible pair differences across the two groups with the weight depending on the covariate difference. Treatment effects for sample selection models arise when the treatment can affect participation in an activity and the performance there; separating these effects has interesting policy implications.