Technical and allocative efficiencies of rice farmers in Bangladesh: Data and STATA codes

This dataset contains household-level data collected from the two greater Rangpur and Mymensingh regions (Lalmonirhat, Kurigram, Rangpur, Gaibandha, Jamalpur, and Sherpur districts) of northwest Bangladesh in 2016. To collect data, a multi-stage stratified random sampling technique was employed: (i) each district mentioned above was divided into two strata, flood-prone and not flood-prone, based on historical flood information from the Bangladesh Bureau of Statistics (BBS), making 12 sub-districts; (ii) five local administrative units from each sub-district were selected, which are called Unions; (iii) from each Union, several villages were randomly selected; (iv) an equal number of rice farmers (twenty-five) were randomly selected and interviewed from each village with a structured questionnaire; (v) finally, a total of 1,500 farmers were face-to-face interviewed. In this study, a total of 998 samples were used, excluding observations that feature missing information and outliers.

本数据集收录了2016年从孟加拉国西北部大朗布尔（Greater Rangpur）与大迈门辛（Greater Mymensingh）两大区域（包含拉尔莫尼哈德、库里格拉姆、朗布尔、盖班达、贾马尔普尔及谢布尔县）采集的家庭层面数据。本次数据采集采用多阶段分层随机抽样方法：(i) 依据孟加拉国统计局（Bangladesh Bureau of Statistics, BBS）的历史洪水数据，将上述各县划分为易涝与非易涝两个层，共得到12个次级区域；(ii) 从每个次级区域中选取5个地方行政单位，即联盟（Union）；(iii) 从每个联盟（Union）中随机选取若干村落；(iv) 采用结构化问卷，从每个村落中随机选取25名水稻种植户开展访谈；(v) 最终累计完成1500名农户的面对面访谈。本研究最终纳入998份有效样本，剔除了存在信息缺失与异常值的观测样本。