Uttar Pradesh and Bihar Survey of Living Conditions 1997-1998 - India

Abstract
---------------------------

A two-part study of rural poverty was carried out in 1997-98 in south and eastern Uttar Pradesh and north and central Bihar. This study utilized both qualitative methods - rapid rural appraisal (RRA) & participatory rural appraisal (PRA) methodologies, and semi-structured interviews - as well as quantitative methods drawing on data collected from household and community surveys modelled after the World Bank's Living Standards Measurement Study (LSMS) surveys. The data being distributed are from the quantitative component of the study, field work for which was carried out between December 1997 and March 1998. Data were collected through household and village-level questionnaires in 120 villages drawn from a sample of 25 districts in UP and Bihar states; a total of 2,250 households were interviewed during the course of the survey (more details on distribution of the sample are provided in the sampling section of this note). Of the sample of 120 villages where the household and village surveys were conducted, 30 had been visited in the earlier qualitative component of the study, while the remaining 90 were drawn at random from the sample districts.

Geographic coverage
---------------------------

Regional

Analysis unit
---------------------------

Households

Kind of data
---------------------------

Sample survey data [ssd]

Sampling procedure
---------------------------

Sampling Information: Uttar Pradesh and Bihar, the two states selected for the study, are divided into 8 statistical regions: 5 in Uttar Pradesh (Himalayan, Western, Central, Eastern, and Southern) and 3 in Bihar(Southern, Northern, and Central).

Sampling Universe: The universe for the study comprised 4 statistical regions: 2 in Uttar Pradesh (Eastern and Southern), and 2 in Bihar (Northern and Central). Altogether, there were 55 districts in the area covered by the study: 24 districts in the 2 statistical regions in Uttar Pradesh, and 31 districts in the 2 statistical regions covered in Bihar. In the first phase of the project, qualitative field work was carried out in 30 villages: 3 villages each from 4 districts in Bihar (Mungher, Jehanabad, Saharsa, and Vaishali), and 6 villages each from 3 districts in Uttar Pradesh (Banda, Allahabad, and Gorakhpur).

Sampling Strategy: The sampling strategy followed for the quantitative study basically involved dividing the sample population into four main strata: 1) districts that were covered in the qualitative study in Bihar (i.e. 4 districts) 2) districts that were covered in the qualitative study in Uttar Pradesh (i.e. 3 districts) 3) remaining districts in the 2 selected regions of Bihar (i.e. 27 districts) 4) remaining districts in the 2 selected regions of Uttar Pradesh (i.e. 21 districts) All 12 villages in Stratum 1 that were covered in the qualitative study were included in the sample. Similarly, all 18 villages in Stratum 2 that were covered in the qualitative study were included in the sample. In each of these 30 villages, 30 households each were picked at random for the survey. In stratums 3 and 4, 45 villages each were selected for the survey. A two-step procedure was used to select villages in these two strata: first, 9 districts were selected in each stratum using PPS. In each of the 9 districts, 5 villages were then selected at the second stage, again using PPS. In each of these 90 villages altogether, 15 households each were selected for the survey.

Mode of data collection
---------------------------

Face-to-face [f2f]

摘要
---------------------------
本项研究于1997-1998年在乌塔尔普拉德邦东南部与比哈尔邦北部及中部地区进行了关于农村贫困问题的两项分阶段研究。研究采用了定性与定量相结合的方法，包括快速农村评估（RRA）与参与式农村评估（PRA）等定性研究方法，以及借鉴世界银行生活标准测量研究（LSMS）调查数据的半结构化访谈。所提供数据为研究定量部分，实地调查工作于1997年12月至1998年3月间进行。数据通过120个村庄的家庭和村级问卷收集，这些村庄均选自乌塔尔普拉德邦和比哈尔邦的25个区中的样本；在调查过程中共访谈了2,250户家庭（样本分布的详细信息见本笔记的抽样部分）。在120个村庄的样本中，其中30个村庄已在研究的早期定性阶段进行过访问，而其余90个村庄则从样本区随机选取。
地理覆盖范围
---------------------------
区域

分析单位
---------------------------
家庭

数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
抽样信息：本研究选取的乌塔尔普拉德邦和比哈尔邦被划分为8个统计区域：乌塔尔普拉德邦5个（喜马拉雅、西部、中部、东部和南部）和比哈尔邦3个（南部、北部和中部）。
抽样总体：研究总体包括4个统计区域：乌塔尔普拉德邦2个（东部和南部），比哈尔邦2个（北部和中部）。总体覆盖区域内共有55个区：乌塔尔普拉德邦2个统计区域的24个区，比哈尔邦2个统计区域的31个区。在项目第一阶段，对30个村庄进行了定性实地工作：比哈尔邦4个区各3个村庄（Mungher、Jehanabad、Saharsa和Vaishali），乌塔尔普拉德邦3个区各6个村庄（Banda、Allahabad和Gorakhpur）。
抽样策略：定量研究的抽样策略主要包括将样本人口划分为四个主要层级：1）比哈尔邦定性研究覆盖的区（即4个区）2）乌塔尔普拉德邦定性研究覆盖的区（即3个区）3）比哈尔邦两个选定区域剩余的区（即27个区）4）乌塔尔普拉德邦两个选定区域剩余的区（即21个区）。所有在定性研究中覆盖的层级1中的12个村庄均包含在样本中。同样，所有在定性研究中覆盖的层级2中的18个村庄也包含在样本中。在每个这些30个村庄中，随机选取30户家庭进行调查。在层级3和4中，每个层级选取45个村庄进行调查。在这些两个层级中，采用两步法选取村庄：首先，在每个层级中，使用概率比例抽样（PPS）选取9个区。在每个区中，第二阶段再选取5个村庄，同样使用PPS。在这些总共90个村庄中，每个村庄选取15户家庭进行调查。
数据收集方式
---------------------------
面对面 [f2f]