Comprehensive Food Security and Vulnerability Analysis and Nutrition Survey 2012 - Rwanda

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

The objective of this Comprehensive Food Security and Vulnerability Analysis & Nutrition Survey (CFSVANS) 2012 is to measure the extent and depth of food and nutrition insecurity in Rwanda, analyze trends over time, and integrate the findings with those from the recent 'Third Integrated Household Living Conditions Survey' (EICV 3) and 'Rwanda Demographic Health Survey 2010' (DHS 2010). The aim of the survey is to identify the population and location of people vulnerable to food insecurity and malnutrition, highlight causes and determine necessary interventions and priority areas that can reduce their vulnerability and support their livelihood.

The survey made use of a two stage cluster sampling method. In total, 7,500 households were sampled from 30 districts. In addition, 748 key informant interviews were carried out. Primary data analysis from both quantitative survey and qualitative interviews was complemented by secondary data analysis - through a review of literature; to produce a detailed report.

The report of the 2012 CFSVANS presents the analyses of the state of food security in Rwanda through the three distinct, but interrelated, dimensions of food availability, food access and food utilization. It also describes the nutritional status of children under the ages of 5; by analysing results from the additional nutritional survey.

Lastly, the report suggests recommendations to tackle the issues and reduce the vulnerability of households to food insecurity and malnutrition. Such strategies include improving coverage and targeting of assistance; improving household living conditions and livelihoods; diversifying agricultural production; building household resilience; and improving child food consumption.

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

National coverage

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

Households

Universe
---------------------------

Urban & Rural household members

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

Sample survey data [ssd]

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

Rwanda is administratively divided into four provinces (Northern Province, Southern Province, Eastern Province and Western Province) plus Kigali City and a total of 30 districts. Districts are further divided in sectors and cells. The sampling frame was based on the data from the recent EICV 3 (2010/2011). To facilitate comparison with existing studies, the CFSVANS 2012 was designed to provide statistically representative and precise information at the district level. In addition, it was decided to include both urban and rural households and not to exclude the capital province Kigali. The sampling frame was organized according to 30 districts. Subsequently, a two-stage cluster sample procedure was applied.

In the first stage, 25 villages per district were randomly selected with probability proportional to population size. In the second stage, ten households in each of the 25 villages in the 30 provinces were selected for participation in the survey. A systematic random sampling technique was chosen for this stage. The team leader, together with the village head, listed all households in the village. Based on this list, a systematic random sample was utilized to pick ten households to be interviewed and three reserve households should any of the first ten households be missing at the time of the interview. Households were eligible for participation in the assessment if living in the selected villages at the time of the interviews. Thus ten households, from 25 villages, from 30 provinces were chosen to participate in the survey, amounting up to 7,500 households.

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

Face-to-face paper [f2f]

Research instrument
---------------------------

Two instruments were used to collect primary data: a key informant questionnaire administered to the village head and other key informants in each of the sampled villages, and a household questionnaire administered to sample households, including an anthropometric section for women of reproductive age (15-49), children under five years, and a section on infant and young child feeding practices intended only for children between six months and two years.

Household survey:
The study gathered information through household questionnaires that included sections on demographics, housing and facilities, assets and access to credit, agriculture, livelihoods, expenditures, food consumption and sources, shocks, and women and child health and nutrition. Some questions in the housing facilities section were replicated from the recent EICV 3 and DHS. This was done to be able to compare the results with those two studies. Out of the 7500 households sampled for the survey, 7498 households actually participated in the survey.

Community questionnaire:
For each visited village, the head of the village was interviewed as key informant with a structured questionnaire. 748 Key informant interviews were conducted. Topics covered included community infrastructure, market information, agricultural crop calendar, shocks and received assistance. This information was then used to contextualize the results from the household questionnaire.

The questionnaires were developed in English and administered in Kinyarwanda. Careful training was conducted to reduce individual variations on how enumerators interpreted the questionnaire and understood the questions.

Cleaning operations
---------------------------

Data was downloaded directly from the PDA to an access database and exported to SPSS for analysis. Data cleaning consisted of examining frequency distributions for all variables in order to detect those values which are not logical or possible. Each participating household, child, and woman had a unique identification number made up of the cluster number and household number and, for individuals, an individual number. For some variables, specifically anthropometric z-scores, standard criteria were applied to delete z-scores which were judged to be impossible and most likely due to error in measurement.

Data appraisal
---------------------------

A series of data quality tables and graphs were available to review the quality of the data.

摘要
---------------------------
本综合粮食安全与脆弱性分析及营养调查（CFSVANS）2012的目标在于衡量卢旺达粮食和营养不安全的程度与深度，分析其随时间的发展趋势，并将调查结果与近期第三次综合家庭生活状况调查（EICV 3）以及2010年卢旺达人口与健康调查（DHS 2010）的结果相结合。调查旨在识别易受粮食不安全和营养不良影响的人群及其地理位置，揭示其成因，并确定必要的干预措施和优先领域，以降低其脆弱性并支持其生计。

调查采用了两阶段聚类抽样方法。总计从30个区中抽取了7,500户家庭作为样本。此外，还进行了748次关键信息提供者访谈。定量调查和定性访谈的初级数据分析得到了次级数据分析的补充——通过文献综述；以产出一份详细的报告。

2012年CFSVANS的报告通过三个相互关联但又独立的维度——食物可获得性、食物可及性和食物利用率，分析了卢旺达粮食安全状况。它还描述了5岁以下儿童的营养状况；通过分析额外的营养调查结果。

最后，报告提出了应对问题并减少家庭对粮食不安全和营养不良脆弱性的建议。这些策略包括改善援助的覆盖范围和针对性；改善家庭生活条件和生计；多样化农业生产；构建家庭韧性；以及改善儿童食物消费。

地理覆盖范围
---------------------------
全国覆盖

分析单元
---------------------------
家庭

总体
---------------------------
城市与农村家庭成员

数据类型
---------------------------
样本调查数据 [ssd]

抽样程序
---------------------------
卢旺达行政上分为四个省（北部省、南部省、东部省和西部省）以及基加利市，共30个区。区进一步分为部门和服务站。抽样框架基于近期EICV 3（2010/2011）的数据。为了便于与现有研究进行比较，CFSVANS 2012被设计成能够在区级提供统计上具有代表性和精确的信息。此外，决定包括城市和农村家庭，并且不排除首都省基加利。抽样框架根据30个区组织。

随后，应用了两阶段聚类抽样程序。

在第一阶段，每个区随机选取了25个村庄，选取概率与人口规模成正比。在第二阶段，从每个省的25个村庄中各选取了10户家庭参与调查。这一阶段选择了系统随机抽样技术。团队领导和村长共同列出了村庄中的所有家庭。根据此名单，利用系统随机抽样选取了10户家庭进行访谈，并预留了3户备选家庭以防前10户家庭在访谈时缺失。如果被调查的家庭在访谈时居住在所选村庄，则具有参与评估的资格。因此，从30个省的25个村庄中选择了10户家庭参与调查，共计7,500户。

数据收集方式
---------------------------
面对面纸质调查 [f2f]

研究工具
---------------------------
收集初级数据使用了两种工具：一种是对每个样本村庄的村长和其他关键信息提供者进行的关键信息提供者问卷，另一种是对样本家庭进行的家庭问卷，包括针对育龄妇女（15-49岁）、5岁以下儿童以及仅针对6个月至2岁儿童设计的婴儿和幼儿喂养实践部分的问卷。

家庭调查：通过家庭问卷收集信息，问卷包括人口统计学、住房与设施、资产与信贷获取、农业、生计、支出、食物消费与来源、冲击以及妇女和儿童健康与营养等部分。住房设施部分的一些问题是从近期EICV 3和DHS复制的，以便能够与这两项研究的结果进行比较。在为调查抽取的7,500户家庭中，有7,498户实际参与了调查。

社区问卷：对每个访问的村庄，村长作为关键信息提供者接受了结构化问卷的访谈。进行了748次关键信息提供者访谈。涵盖的主题包括社区基础设施、市场信息、农业作物日历、冲击和获得的援助。然后，使用这些信息来解释家庭问卷的结果。

问卷以英语开发，并在卢旺达语中实施。进行了仔细的培训，以减少调查员对问卷的解释和问题理解的个人差异。

数据清理
---------------------------
数据直接从PDA下载到Access数据库，并导出到SPSS进行分析。数据清理包括检查所有变量的频率分布，以检测那些不合理或不可能的值。每个参与的家庭、儿童和妇女都有一个独特的识别号码，由聚类编号和家庭编号组成，对于个人，有一个个体编号。对于某些变量，特别是人体测量学Z分数，应用了标准标准来删除被认为是不可行且最可能因测量错误而导致的Z分数。

数据评估
---------------------------
提供了一系列数据质量表格和图表以审查数据质量。