Living Standards Survey, 2018-2019 - Nigeria

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

The main objectives of the 2018/19 NLSS are: i) to provide critical information for production of a wide range of socio-economic and demographic indicators, including for benchmarking and monitoring of SDGs; ii) to monitor progress in population's welfare; iii) to provide statistical evidence and measure the impact on households of current and anticipated government policies. In addition, the 2018/19 NLSS could be utilized to improve other non-survey statistical information, e.g. to determine and calibrate the contribution of final consumption expenditures of households to GDP; to update the weights and determine the basket for the national Consumer Price Index (CPI); to improve the methodology and dissemination of micro-economic and welfare statistics in Nigeria.

The 2018/19 NLSS collected a comprehensive and diverse set of socio-economic and demographic data pertaining to the basic needs and conditions under which households live on a day to day basis. The 2018/19 NLSS questionnaire includes wide-ranging modules, covering demographic indicators, education, health, labour, expenditures on food and non-food goods, non-farm enterprises, household assets and durables, access to safety nets, housing conditions, economic shocks, exposure to crime and farm production indicators.

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

National coverage

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

Households

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

The survey covered all de jure households excluding prisons, hospitals, military barracks, and school dormitories.

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

Sample survey data [ssd]

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

SAMPLING PROCEDURE
The 2018/19 NLSS sample is designed to provide representative estimates for the 36 states and the Federal Capital Territory (FCT), Abuja. By extension. The sample is also representative at the national and zonal levels. Although the sample is not explicitly stratified by urban and rural areas, it is possible to obtain urban and rural estimates from the NLSS data at the national level. At all stages, the relative proportion of urban and rural EAs as has been maintained. Before designing the sample for the 2018/19 NLSS, the results from the 2009/10 HNLSS were analysed to extract the sampling properties (variance, design effect, etc.) and estimate the required sample size to reach a desired precision for poverty estimates in the 2018/19 NLSS.

EA SELECTION: The sampling frame for the 2018/19 NLSS was based on the national master sample developed by the NBS, referred to as the NISH2 (Nigeria Integrated Survey of Households 2). This master sample was based on the enumeration areas (EAs) defined for the 2006 Nigeria Census Housing and Population conducted by National Population Commission (NPopC). The NISH2 was developed by the NBS to use as a frame for surveys with state-level domains. NISH2 EAs were drawn from another master sample that NBS developed for surveys with LGA-level domains (referred to as the “LGA master sample”). The NISH2 contains 200 EAs per state composed of 20 replicates of 10 sample EAs for each state, selected systematically from the full LGA master sample. Since the 2018/19 NLSS required domains at the state-level, the NISH2 served as the sampling frame for the survey. Since the NISH2 is composed of state-level replicates of 10 sample EAs, a total of 6 replicates were selected from the NISH2 for each state to provide a total sample of 60 EAs per state. The 6 replicates selected for the 2018/19 NLSS in each state were selected using random systematic sampling. This sampling procedure provides a similar distribution of the sample EAs within each state as if one systematic sample of 60 EAs had been selected directly from the census frame of EAs.

A fresh listing of households was conducted in the EAs selected for the 2018/19 NLSS. Throughout the course of the listing, 139 of the selected EAs (or about 6%) were not able to be listed by the field teams. The primary reason the teams were not able to conduct the listing in these EAs was due to security issues in the country. The fieldwork period of the 2018/19 NLSS saw events related to the insurgency in the north east of the country, clashes between farmers and herdsman, and roving groups of bandits. These events made it impossible for the interviewers to visit the EAs in the villages and areas affected by these conflict events. In addition to security issues, some EAs had been demolished or abandoned since the 2006 census was conducted. In order to not compromise the sample size and thus the statistical power of the estimates, it was decided to replace these 139 EAs. Additional EAs from the same state and sector were randomly selected from the remaining NISH2 EAs to replace each EA that could not be listed by the field teams. This necessary exclusion of conflict affected areas implies that the sample is representative of areas of Nigeria that were accessible during the 2018/19 NLSS fieldwork period. The sample will not reflect conditions in areas that were undergoing conflict at that time. This compromise was necessary to ensure the safety of interviewers.

HOUSEHOLD SELECTION: Following the listing, the 10 households to be interviewed were selected from the listed households. These households were selected systemically after sorting by the order in which the households were listed. This systematic sampling helped to ensure that the selected households were well dispersed across the EA and thereby limit the potential for clustering of the selected households within an EA. Occasionally, interviewers would encounter selected households that were not able to be interviewed (e.g. due to migration, refusal, etc.). In order to preserve the sample size and statistical power, households that could not be interviewed were replaced with an additional randomly selected household from the EA. Replacement households had to be requested by the field teams on a case-by-case basis and the replacement household was sent by the CAPI managers from NBS headquarters. Interviewers were required to submit a record for each household that was replaced, and justification given for their replacement. These replaced households are included in the disseminated data. However, replacements were relatively rare with only 2% of sampled households not able to be interviewed and replaced.

Sampling deviation
---------------------------

Although a sample was initially drawn for Borno state, the ongoing insurgency in the state presented severe challenges in conducting the survey there. The situation in the state made it impossible for the field teams to reach large areas of the state without compromising their safety. Given this limitation it was clear that a representative sample for Borno was not possible. However, it was decided to proceed with conducting the survey in areas that the teams could access in order to collect some information on the parts of the state that were accessible.

The limited area that field staff could safely operate in in Borno necessitated an alternative sample selection process from the other states. The EA selection occurred in several stages. Initially, an attempt was made to limit the frame to selected LGAs that were considered accessible. However, after selection of the EAs from the identified LGAs, it was reported by the NBS listing teams that a large share of the selected EAs were not safe for them to visit. Therefore, an alternative approach was adopted that would better ensure the safety of the field team but compromise further the representativeness of the sample. First, the list of 788 EAs in the LGA master sample for Borno were reviewed by NBS staff in Borno and the EAs they deemed accessible were identified. The team identified 359 EAs (46%) that were accessible. These 359 EAs served as the frame for the Borno sample and 60 EAs were randomly selected from this frame. However, throughout the course of the NLSS fieldwork, additional insurgency related events occurred which resulted in 7 of the 60 EAs being inaccessible when they were to be visited. Unlike for the main sample, these EAs were not replaced. Therefore, 53 EAs were ultimately covered from the Borno sample. The listing and household selection process that followed was the same as for the rest of the states.

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

Computer Assisted Personal Interview [capi]

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

Two sets of questionnaires – household and community – were used to collect information in the NLSS2018/19. The Household Questionnaire was administered to all households in the sample. The Community Questionnaire was administered to the community to collect information on the socio-economic indicators of the enumeration areas where the sample households reside.

Household Questionnaire: The Household Questionnaire provides information on demographics; education; health; labour; food and non-food expenditure; household nonfarm income-generating activities; food security and shocks; safety nets; housing conditions; assets; information and communication technology; agriculture and land tenure; and other sources of household income.

Community Questionnaire: The Community Questionnaire solicits information on access to transported and infrastructure; community organizations; resource management; changes in the community; key events; community needs, actions and achievements; and local retail price information.

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

CAPI: The 2018/19 NLSS was conducted using the Survey Solutions Computer Assisted Person Interview (CAPI) platform. The Survey Solutions software was developed and maintained by the Development Economics Data Group (DECDG) at the World Bank. Each interviewer and supervisor was given a tablet which they used to conduct the interviews. Overall, implementation of the survey using Survey Solutions CAPI was highly successful, as it allowed for timely availability of the data from completed interviews and real-time quality checks.

DATA COMMUNICATION SYSTEM: The data communication system used in 2018/19 NLSS was highly automated. Each field team was given a mobile modem to allow for internet connectivity and daily synchronization of their assignments and completed interviews. This ensured that headquarters in Abuja had access to the data in real-time. Once each interview was completed and uploaded to the server, the data was first reviewed by the data editors. The data was also downloaded from the server, and Stata dofiles run on the downloaded data to check for additional errors that were not captured by the Survey Solutions application during data collection and entry. An excel error file is generated following the running of the Stata dofiles on the raw dataset. Information contained in the excel error files are communicated back to respective field interviewers for action by the interviewers. This action was done on a daily basis for the duration of the survey.

DATA CLEANING: The data cleaning process was done in three main stages. The first stage was to ensure proper quality control during the fieldwork. This was achieved in part by incorporating validation and consistency checks into the Survey Solutions application used for the data collection and designed to highlight many of the errors that occurred during the fieldwork.
The second stage cleaning involved the use of data editors and data assistants. As indicated above, once the interview is completed and uploaded to the server, the data editors review completed interview for inconsistencies and extreme values. Depending on the outcome, they can either approve or reject the case. If rejected, the case goes back to the respective interviewer's tablet upon synchronization. Special care was taken to see that the households included in the data matched with the selected sample and where there were differences, these were properly assessed and documented. Additional errors observed were compiled into error reports that were regularly sent to the teams. These errors were then corrected based on re-visits to the household on the instruction of the supervisor. The data that had gone through this first stage of cleaning was then approved by the Data Editor. After the data editor's approval of the interview on Survey Solutions server, the Headquarters also reviews and depending on the outcome, can either reject or approve.

The third stage of cleaning involved a comprehensive review of the final raw data following the first and second stage cleaning. Every variable was examined individually for (1) consistency with other sections and variables, (2) out of range responses, and (3) outliers. However, special care was taken to avoid making strong assumptions when resolving potential errors. Some minor errors remain in the data where the diagnosis and/or solution were unclear to the data cleaning team.

摘要
---------------------------
2018/19年国家生活水平调查（NLSS）的主要目标包括：
i) 为生产广泛的社会经济和人口统计数据提供关键信息，包括用于基准和监测可持续发展目标（SDGs）；
ii) 监测人口福祉的进展；
iii) 提供统计证据并衡量当前及预期政府政策对家庭的影响。此外，2018/19年NLSS可用于改进其他非调查统计信息，例如，确定并校准家庭最终消费支出对国内生产总值（GDP）的贡献；更新国家消费者价格指数（CPI）的权重和确定篮子；改进尼日利亚微观经济和福利统计的方法和传播。
2018/19年NLSS收集了关于家庭日常基本需求和条件下的全面且多样的社会经济和人口数据。2018/19年NLSS问卷包括广泛的模块，涵盖人口指标、教育、健康、劳动、食品和非食品商品支出、非农企业、家庭资产和耐用消费品、安全网可及性、住房条件、经济冲击、犯罪暴露和农业生产指标。
地理覆盖范围
---------------------------
全国覆盖
分析单位
---------------------------
家庭
总体
---------------------------
调查涵盖所有法定家庭，不包括监狱、医院、军事营房和学校宿舍。
数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
抽样程序
2018/19年NLSS样本旨在为尼日利亚36个州和联邦首都特区（FCT）阿布贾提供代表性估计。通过扩展，该样本在国家及区域层面也具有代表性。尽管样本未明确按城市和农村地区分层，但可以从NLSS数据中获取全国层面的城市和农村估计。在所有阶段，城市和农村地区抽样单位的相对比例均得到维持。在为2018/19年NLSS设计样本之前，分析了2009/10年HNLSS的结果，以提取抽样属性（方差、设计效应等）并估计达到2018/19年NLSS贫困估计所需精度的所需样本量。
EA选择：2018/19年NLSS的抽样框架基于由国家统计局（NBS）开发的国家级主样本，称为NISH2（尼日利亚家庭综合调查2）。该主样本基于国家人口委员会（NPopC）进行的2006年尼日利亚人口普查住房和人口中定义的普查区域（EA）。NBS开发了NISH2，用作具有州级域的调查的框架。NISH2 EA来自国家统计局（NBS）为具有地方政府级域的调查开发的另一个主样本（称为“地方政府主样本”）。NISH2包含每个州200个EA，由每个州10个样本EA的20个复制品组成，这些样本EA是从完整的地方政府主样本中系统性地选出的。由于2018/19年NLSS需要州级域，因此NISH2作为调查的抽样框架。由于NISH2由州级复制品组成，每个州有10个样本EA的20个复制品，因此为每个州选择了6个复制品，以提供每个州60个EA的总样本。为2018/19年NLSS在每个州选择的6个复制品使用随机系统抽样方法进行选择。这种抽样程序提供了与直接从EA普查框架中选择60个EA的系统样本相似的每个州内样本EA的分布。
在为2018/19年NLSS选择的EA中进行了新的家庭清单。在整个清单过程中，有139个（约6%）选定的EA无法由现场团队进行清单编制。现场团队无法在这些EA中进行清单编制的主要原因是国家存在的安全问题。2018/19年NLSS的现场工作期间，发生了与该国东北部叛乱、农民与牧民之间的冲突以及游荡的匪徒团伙相关的事件。这些事件使得访谈者无法访问受冲突事件影响的村庄和地区。除了安全问题外，一些EA自2006年人口普查以来已被拆除或废弃。为了不损害样本量及其统计估计的统计能力，决定用139个EA替换这些EA。从剩余的NISH2 EA中随机选择了相同州和部门的EA，以替换每个无法由现场团队进行清单编制的EA。这种必要的排除受冲突影响的地区意味着样本代表了2018/19年NLSS现场工作期间可访问的尼日利亚地区。该样本将不会反映当时处于冲突状态地区的状况。这种妥协是为了确保访谈者的安全。
家庭选择：在清单编制之后，从清单中的家庭中选择10个家庭进行访谈。这些家庭在按家庭清单编制顺序排序后进行系统选择。这种系统抽样有助于确保所选家庭在EA中得到良好分散，从而限制所选家庭在EA内聚集的可能性。偶尔，访谈者会遇到无法进行访谈的选定家庭（例如，由于迁移、拒绝等）。为了保持样本量和统计能力，用从EA中随机选择的额外家庭替换了无法进行访谈的家庭。这些替换家庭需要由现场团队逐案提出请求，并由国家统计局总部从CAPI经理处发送替换家庭。访谈者必须提交每个被替换家庭的记录，并给出其替换的理由。这些替换家庭包含在发布的数据中。然而，替换相对罕见，只有2%的抽样家庭无法进行访谈并被替换。
抽样偏差
---------------------------
尽管最初为博尔诺州抽取了样本，但该州持续的叛乱对该州进行调查造成了严重挑战。该州的情况使得现场团队无法在不危及他们安全的情况下进入该州的大部分地区。鉴于这一限制，很明显，无法为博尔诺州抽取代表性样本。然而，决定在团队可以进入的地区进行调查，以收集有关该州可访问部分的某些信息。
博尔诺州现场工作人员可以安全操作的区域有限，这需要从其他州选择替代样本选择过程。EA选择发生在几个阶段。最初，尝试将框架限制为被认为可访问的选定地方政府。然而，在从确定的LGA中选择EA后，国家统计局的清单团队报告称，所选EA中有很大一部分对他们来说不安全。因此，采用了另一种方法，这更有利于确保现场团队的安全，但进一步妥协了样本的代表性。首先，由博尔诺州国家统计局工作人员审查了788个EA的LGA主样本清单，并确定了他们认为可访问的EA。该团队确定了46%（359个）可访问的EA。这些EA作为博尔诺样本的框架，并从中随机选择了60个EA。然而，在整个NLSS现场工作期间，发生了更多与叛乱相关的事件，导致在计划访问时，7个EA无法访问。与主样本不同，这些EA没有进行替换。因此，最终覆盖了博尔诺样本的53个EA。随后进行的清单编制和家庭选择过程与其他州相同。
数据收集方式
---------------------------
计算机辅助个人访谈 [capi]
研究工具
---------------------------
NLSS2018/19使用了两组问卷——家庭问卷和社区问卷——来收集信息。家庭问卷对所有样本家庭进行行政。社区问卷用于社区，以收集样本家庭居住的普查区域的 socioeconomic 指标。
家庭问卷：家庭问卷提供有关人口统计、教育、健康、劳动、食品和非食品支出、家庭非农收入生成活动、粮食安全和冲击、安全网、住房条件、资产、信息和通信技术、农业和土地产权以及家庭其他收入来源的信息。
社区问卷：社区问卷征求有关交通和基础设施的可及性、社区组织、资源管理、社区变化、关键事件、社区需求、行动和成就以及当地零售价格信息。
数据清理操作
---------------------------
CAPI：2018/19年NLSS使用Survey Solutions计算机辅助个人访谈（CAPI）平台进行。Survey Solutions软件由世界银行的发展经济学数据组（DECDG）开发和维护。每位访谈者和监督员都获得了一个平板电脑，他们使用该平板电脑进行访谈。总的来说，使用Survey Solutions CAPI进行调查的实施非常成功，因为它允许及时获得已完成访谈的数据，并进行实时质量检查。
数据通信系统：2018/19年NLSS使用的数据通信系统高度自动化。每个现场团队都获得了一个移动调制解调器，以允许互联网连接和每日同步其任务和已完成访谈。这确保了阿布贾总部可以实时访问数据。一旦每个访谈完成并上传到服务器，数据首先由数据编辑进行审查。数据还从服务器下载，并在下载的数据上运行Stata dofiles，以检查在数据收集和输入期间未由Survey Solutions应用程序捕获的额外错误。在运行Stata dofiles后的原始数据集上生成一个excel错误文件。excel错误文件中的信息被传达给相应的现场访谈者，以便访谈者采取行动。这项行动在整个调查期间每天进行。
数据清理：数据清理过程分为三个主要阶段。第一阶段是在现场工作期间确保适当的质量控制。这通过将验证和一致性检查纳入用于数据收集的Survey Solutions应用程序来实现，该应用程序旨在突出许多在实地工作期间发生的错误。第二阶段清理涉及使用数据编辑和数据助理。如上所述，一旦访谈完成并上传到服务器，数据编辑将审查完成的访谈，以查找不一致性和极端值。根据结果，他们可以批准或拒绝案例。如果拒绝，则案例在同步时返回到相应访谈者的平板电脑。特别关注确保包含在数据中的家庭与所选样本相匹配，并在存在差异的情况下，对这些差异进行适当的评估和记录。根据监督员的指示，对观察到的其他错误进行了编制，并定期将错误报告发送给团队。这些错误根据对家庭的重新访问进行了纠正。经过第一阶段清理的数据由数据编辑批准。在数据编辑在Survey Solutions服务器上批准访谈后，总部也进行审查，并根据结果，可以拒绝或批准。
第三阶段清理涉及对第一阶段和第二阶段清理后的最终原始数据进行全面审查。每个变量都单独检查，以（1）与其他部分和变量的一致性，（2）超出范围的反应，以及（3）异常值。然而，在解决潜在错误时，特别小心避免做出强烈的假设。一些小错误仍然存在于数据中，其中对数据清理团队来说，诊断和/或解决方案是不清楚。