Evaluation of Development Programs : Randomized Controlled Trials or Regressions?

Elbers, Chris ; Gunning, Jan Willem

World Bank, Washington, DC

关键词: BENEFICIARIES;    CALCULATION;    CAPACITY BUILDING;    CONTROL GROUPS;    CORRELATIONS;   

DOI  :  10.1596/1813-9450-6587
RP-ID  :  WPS6587

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

Can project evaluation methods be usedto evaluate programs: complex interventions involvingmultiple activities? A program evaluation cannot be basedsimply on separate evaluations of its components ifinteractions between the activities are important. In thispaper a measure is proposed, the total program effect (TPE),which is an extension of the average treatment effect on thetreated (ATET). It explicitly takes into account that in thereal world (with heterogeneous treatment effects) individualtreatment effects and program assignment are oftencorrelated. The TPE can also deal with the common situationin which such a correlation is the result of decisions on(intended) program participation not being taken centrally.In this context RCTs are less suitable even for the simplestinterventions. The TPE can be estimated by applyingregression techniques to observational data from arepresentative sample from the targeted population. Theapproach is illustrated with an evaluation of a healthinsurance program in Vietnam.