Two Become One: Improving the Targeting of Conditional Cash Transfers with a Predictive Model of School Dropout.
This paper offers a methodology to improve targeting design and assessment when two or more groups need to be considered, and trade-offs exist between using different targeting mechanisms. The paper builds from the multidimensional targeting challenge facing conditional cash transfers (CCTs). I anal...
| Publicado en: | Economía Vol. 21; no. 1; pp. 1 - 46 |
|---|---|
| Autor principal: | |
| Formato: | Artículo |
| Publicado: |
London School of Economics & Political Science
Fall2020
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=153590667&site=ehost-live header: @attributes: shortDbName: ssf uiTerm: 153590667 longDbName: Social Sciences Full Text (H.W. Wilson) uiTag: AN controlInfo: bkinfo: jinfo: jid: 15297470 N8H jtl: Economía issn: 15297470 maglogo: N pubinfo: dt: Fall2020 vid: 21 iid: 1 pid: 66142 pub: London School of Economics & Political Science artinfo: ui: 153590667 10.1353/eco.2020.0011 ppf: 1 ppct: 45 formats: fmt: @attributes: type: P size: 2MB tig: atl: Two Become One: Improving the Targeting of Conditional Cash Transfers with a Predictive Model of School Dropout. aug: au: CRESPO, CRISTIAN affil: London School of Economics and Political Science, London, UK su: Chile School dropouts College dropouts Public officers Conditional cash transfer programs Prediction models Machine learning National school lunch program sug: subj: School dropouts College dropouts Public officers Chile Conditional cash transfer programs Prediction models Machine learning National school lunch program keyword: conditional cash transfers machine learning Multidimensional targeting school dropout prediction conditional cash transfers machine learning Multidimensional targeting school dropout prediction ab: This paper offers a methodology to improve targeting design and assessment when two or more groups need to be considered, and trade-offs exist between using different targeting mechanisms. The paper builds from the multidimensional targeting challenge facing conditional cash transfers (CCTs). I analyze whether a common CCT targeting mechanism, namely, a proxy means test (PMT), can identify the poor and future school dropouts effectively. Despite both being key target groups for CCTs, students at risk of dropping out are rarely considered for CCT allocation or in targeting assessments. Using rich administrative data sets from Chile to simulate different targeting mechanisms, I compare the targeting effectiveness of a PMT and other mechanisms based on a predictive model of school dropout. I build this model using machine learning algorithms. Using two novel metrics, I show that combining the outputs of the predictive model with the PMT increases targeting effectiveness except when the social valuation of the poor and future school dropouts differs to a large extent. More generally, public officials who value their key target groups equally may improve policy targeting by modifying their allocation procedures. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: N holdings: @attributes: islocal: N |
|---|