Regression Discontinuity World Bank SIEF / APHRC Impact ...

1 Regression DiscontinuityWorld bank SIEF / APHRCI mpact evaluation Training2015 Owen OzierDevelopment Research GroupThe World Bank6 May 2015 Owen Ozier (The World bank ) Regression Discontinuity6 May 20151 / 9 Regression Discontinuity ( RD ) Stata sectionWe will use a new dataset (.dta file) separate .do files are provided for this module:1. Getting the RD package Exploring a new dataset: a (fictional) training Basics of Regression Discontinuity : therdcommand and its manual Fuzzy RD and manipulation of the running variable: density and covariate Ozier (The World bank ) Regression Discontinuity6 May 20152 / 9 Regression Discontinuity ( RD ) Stata sectionWe will use a new dataset (.dta file) separate .do files are provided for this module:1. Getting the RD package Exploring a new dataset: a (fictional) training Basics of Regression Discontinuity : therdcommand and its manual Fuzzy RD and manipulation of the running variable: density and covariate Ozier (The World bank ) Regression Discontinuity6 May 20152 / 9 Regression Discontinuity ( RD ) Stata sectionWe will use a new dataset (.)

2 Dta file) separate .do files are provided for this module:1. Getting the RD package Exploring a new dataset: a (fictional) training Basics of Regression Discontinuity : therdcommand and its manual Fuzzy RD and manipulation of the running variable: density and covariate Ozier (The World bank ) Regression Discontinuity6 May 20152 / 9 Regression Discontinuity ( RD ) Stata sectionWe will use a new dataset (.dta file) separate .do files are provided for this module:1. Getting the RD package Exploring a new dataset: a (fictional) training Basics of Regression Discontinuity : therdcommand and its manual Fuzzy RD and manipulation of the running variable: density and covariate Ozier (The World bank ) Regression Discontinuity6 May 20152 / 9 Regression Discontinuity ( RD ) Stata sectionWe will use a new dataset (.

3 Dta file) separate .do files are provided for this module:1. Getting the RD package Exploring a new dataset: a (fictional) training Basics of Regression Discontinuity : therdcommand and its manual Fuzzy RD and manipulation of the running variable: density and covariate Ozier (The World bank ) Regression Discontinuity6 May 20152 / 9 Regression Discontinuity ( RD ) Stata sectionWe will use a new dataset (.dta file) separate .do files are provided for this module:1. Getting the RD package Exploring a new dataset: a (fictional) training Basics of Regression Discontinuity : therdcommand and its manual Fuzzy RD and manipulation of the running variable: density and covariate Ozier (The World bank ) Regression Discontinuity6 May 20152 / 9A fictional computer and typing skills training programPhoto credits: Wikipedia / Robin Taylor, flickr.

4 Wikipedia / AbelkazzahOwen Ozier (The World bank ) Regression Discontinuity6 May 20153 / 9 Quickly reviewing Stata syntax[command] [variables or other "arguments"], [options]as insummarize motheryoborsummarize motheryob, detailand some commands, options, and even variables can be abbreviated:su mothery, din general, to learn more about a command and its syntax:help [command]Owen Ozier (The World bank ) Regression Discontinuity6 May 20154 / 9 Quickly reviewing Stata syntax[command] [variables or other "arguments"], [options]as insummarize motheryoborsummarize motheryob, detailand some commands, options, and even variables can be abbreviated:su mothery, din general, to learn more about a command and its syntax:help [command]Owen Ozier (The World bank ) Regression Discontinuity6 May 20154 / 9 Quickly reviewing Stata syntax[command] [variables or other "arguments"], [options]as insummarize motheryoborsummarize motheryob, detailand some commands, options, and even variables can be abbreviated:su mothery, din general, to learn more about a command and its syntax:help [command]Owen Ozier (The World bank ) Regression Discontinuity6 May 20154 / 9 Quickly reviewing Stata syntax[command] [variables or other "arguments"], [options]as insummarize motheryoborsummarize motheryob, detailand some commands, options, and even variables can be abbreviated:su mothery, din general, to learn more about a command and its syntax.

5 Help [command]Owen Ozier (The World bank ) Regression Discontinuity6 May 20154 / 9 Quickly reviewing Stata syntax[command] [variables or other "arguments"], [options]as insummarize motheryoborsummarize motheryob, detailand some commands, options, and even variables can be abbreviated:su mothery, din general, to learn more about a command and its syntax:help [command]Owen Ozier (The World bank ) Regression Discontinuity6 May 20154 / 9 Basic characteristics of the data. describeContains data from : 80,000 Example Regression Discontinuity Data - Owen Ozier 2015vars: 17 5 May 2015 07:57size: 2,960,000------------------------------- ---------------------------------------- ---------------------------------------- ---------------------------------------- ----storage display valuevariable name type format label variable label----------------------------------- ---------------------------------------- ---------------------------------------- ---------------------------------------- id float % Unique identifierdatacollected byte % Full data available for this individualselfskills float % Pre-baseline: Self-assessed labor market skills score 2011skills float % Baseline.

6 Labor market skills score in 2012el byte % Baseline: Eligibility for 2013 training programelsk float % Baseline: Eligibility for program X skills scorepri byte % Baseline: Completed primary schooling by 2012intcomp byte % intlev Baseline: Interest in gaining computer skills (2012)motheryob int % Baseline: Mother s year of birthmothermath float % Baseline: Mother s mathematics test score in 2012pasttrainings byte % Baseline: Participated in job trainings by 2012friendtrain byte % Baseline: Any close friends participated in job training by 2012finishtraining byte % Administrative data: Finished 2013 training programtsk byte % Endline: Typing skill (WP30s), end of 2013 training programempdayafter byte % Endline: Has formal job, one day after 2013 training programempyearafter byte % Endline: Has formal job, one year after training program (2014)earningsafter int % Endline: Earnings, one year after training program (2014)---------------------------------- ---------------------------------------- ---------------------------------------- ---------------------------------------- -Sorted by.

7 IdOwen Ozier (The World bank ) Regression Discontinuity6 May 20155 / 9 Basic characteristics of the data. codebook pri------------------------------------- ---------------------------------------- -----------pri Baseline: Completed primary schooling by 2012------------------------------------ ---------------------------------------- ------------type: numeric (byte)range: [0,1] units: 1unique values: 2 missing .: 20000/80000tabulation: Freq. Value30326 029674 120000 .Owen Ozier (The World bank ) Regression Discontinuity6 May 20156 / 9 Basic characteristics of the data. codebook skills---------------------------------- ---------------------------------------- --------------skills Baseline: Labor market skills score in 2012------------------------------------ ---------------------------------------- ------------type: numeric (float)range: [-1.]

8 99992049] units: values: 58982 missing .: 20000/80000mean: dev: .409222percentiles: 10% 25% 50% 75% 90% .270603 .513644 Owen Ozier (The World bank ) Regression Discontinuity6 May 20157 / 9 Basic characteristics of the dataWhich of these would make a better running variable for a Regression Discontinuity ?-------------------------- ---------------------------------------- ----------------------pri Baseline: Completed primary schooling by 2012------------------------------------ ---------------------------------------- ------------type: numeric (byte)range: [0,1] units: 1unique values: 2tabulation: Freq. Value30326 029674 1--------------------------------------- ---------------------------------------- ---------skills Baseline: Labor market skills score in 2012------------------------------------ ---------------------------------------- ------------type: numeric (float)unique values: 58982percentiles: 10% 25% 50% 75% 90%.

9 270603 .513644 Owen Ozier (The World bank ) Regression Discontinuity6 May 20158 / 9 Basic characteristics of the dataWhich of these would make a better running variable for a Regression Discontinuity ?0510152025 Density01 Baseline: Completed primary schooling by 1 . : Labor market skills score in 2012 Owen Ozier (The World bank ) Regression Discontinuity6 May 20159 / 9