Targeting poor students in Thailand with proxy means test

Proxy Means Test (PMT) is one of the most efficient ways to target the poor by measuring a wealth or poverty without using income, consumption or expenditure. Procedure of PMT is using household characteristic variables which have relation with income as a proxy for poverty. In this study, we created the PMT poverty scorecard to identify poor students in 10 provinces of Thailand, including Mae Hong Son, Nan, Nakhon Ratchasimi, Udon Thani, Nakhon Phanom, Chiang Rai, Trung, Kanchanaburi, Chanthaburi and Phuket, by using Least Squares Regression to estimate relationships between household characteristic variables and (log of) student’s average monthly household income per capita separately by province. The result indicated that PMT poverty targeting works well in terms of low undercoverage rate around 12 percent and high targeting accuracy rate in both poverty and total accuracy around 88 and 72 percent respectively. Therefore, we suggest using PMT as a main poverty targeting for any social welfare program in Thailand to transfer a targeted subsidy to those who are truly poor.

代理均值测试（Proxy Means Test，PMT）是无需借助收入、消费或支出数据，通过测算财富或贫困状况以精准瞄准贫困人口的高效方法之一。PMT的实施逻辑为：选取与收入存在相关性的家庭特征变量，作为贫困状况的代理指标。本研究构建了PMT贫困评分卡（poverty scorecard），用于识别泰国10个省份的贫困学生，覆盖夜丰颂府、南府、呵叻府、乌隆他尼府、那空帕农府、清莱府、Trung府、北碧府、尖竹汶府以及普吉府；研究采用最小二乘回归（Least Squares Regression），按省份分别估计家庭特征变量与（对数化的）学生家庭人均月平均收入之间的关联关系。结果显示，PMT贫困瞄准效果优异：漏评率约为12%，贫困识别准确率与总体准确率分别约为88%和72%。据此，我们建议将PMT作为泰国各类社会福利项目的核心贫困瞄准工具，向真正的贫困群体发放定向补贴。

搜集汇总

背景与挑战

背景概述

该数据集基于泰国10个省份的贫困学生目标定位研究，使用代理均值测试（PMT）方法，通过家庭特征变量替代收入测量贫困，并应用最小二乘回归进行分析。结果显示PMT具有低漏覆盖率（约12%）和高目标定位准确率（贫困准确率约88%，总准确率约72%），建议将其作为泰国社会福利项目的主要贫困目标定位工具。