Abstract

The Integrated Household Survey is one of the primary instruments implemented by the Government of Malawi through the National Statistical Office (NSO) roughly every 5 years to monitor and evaluate the changing conditions of Malawian households. The IHS data have, among other insights, provided benchmark poverty and vulnerability indicators to foster evidence-based policy formulation and monitor the progress of meeting the Millennium Development Goals (MDGs), the goals listed as part of the Malawi Growth and Development Strategy (MGDS) and now the Sustainable Development Goals (SDGs).

Geographic coverage

National

Analysis unit

• Households
• Individuals
• Children under 5 years
• Consumption expenditure commodities/items
• Communities
• Agricultural household/ Holder/ Crop

Universe

Members of the following households are not eligible for inclusion in the survey: • All people who live outside the selected EAs, whether in urban or rural areas. • All residents of dwellings other than private dwellings, such as prisons, hospitals and army barracks. • Members of the Malawian armed forces who reside within a military base. (If such individuals reside in private dwellings off the base, however, they should be included among the households eligible for random selection for the survey.) • Non-Malawian diplomats, diplomatic staff, and members of their households. (However, note that non-Malawian residents who are not diplomats or diplomatic staff and are resident in private dwellings are eligible for inclusion in the survey. The survey is not restricted to Malawian citizens alone.) • Non-Malawian tourists and others on vacation in Malawi.

Kind of data

Sample survey data [ssd]

Sampling procedure

The IHS4 sampling frame is based on the listing information and cartography from the 2008 Malawi Population and Housing Census (PHC); includes the three major regions of Malawi, namely North, Center and South; and is stratified into rural and urban strata. The urban strata include the four major urban areas: Lilongwe City, Blantyre City, Mzuzu City, and the Municipality of Zomba. All other areas are considered as rural areas, and each of the 27 districts were considered as a separate sub-stratum as part of the main rural stratum. This is the first round of the survey to include the island district of Likoma in the sampling frame. The sampling frame further excludes the population living in institutions, such as hospitals, prisons and military barracks. Hence, the IHS4 strata are composed of 32 districts in Malawi.

A stratified two-stage sample design was used for the IHS4.

Note: Detailed sample design information is presented in the "Fourth Integrated Household Survey 2016-2017, Basic Information Document" document.

Mode of data collection

Computer Assisted Personal Interview [capi]

Research instrument

HOUSEHOLD QUESTIONNAIRE The Household Questionnaire is a multi-topic survey instrument and is near-identical to the content and organization of the IHS3. It encompasses economic activities, demographics, welfare and other sectoral information of households. It covers a wide range of topics, dealing with the dynamics of poverty (consumption, cash and non-cash income, savings, assets, food security, health and education, vulnerability and social protection). Although the IHS4 household questionnaire covers a wide variety of topics in detail it intentionally excludes in-depth information on topics covered in other surveys that are part of the NSO’s statistical plan (such as maternal and child health issues covered at length in the Malawi Demographic and Health Survey).

AGRICULTURE QUESTIONNAIRE All IHS4 households that are identified as being involved in agricultural or livestock activities were administered the agriculture questionnaire, which is primarily modelled after the IHS3 counterpart. The modules are expanding on the agricultural content of the IHS3, IHS2, AISS, and other regional agricultural surveys, while remaining consistent with the NACAL topical coverage and methodology. The development of the agriculture questionnaire was done with input from the aforementioned stakeholders who provided input on the household questionnaire as well as outside researchers involved in research and policy discussions pertaining to the Malawian agriculture. The agriculture questionnaire allows, among other things, for extensive agricultural productivity analysis through the diligent estimation of land areas, both owned and cultivated, labor and non-labor input use and expenditures, and production figures for main crops, and livestock. Although one of the major foci of the agriculture data collection effort was to produce smallholder production estimates for major crops, it is also possible to disaggregate the data by gender and main geographical regions. The IHS4 cross-sectional households supply information on the last completed rainy season (2014/2015 or 2015/2016) and the last completed dry season (2015 or 2016) depending on the timing of their interview.

FISHERIES QUESTIONNAIRE The design of the IHS4 fishery questionnaire is identical to the questionnaire designed for IHS3. The IHS3 fisheries questionnaire was informed by the design and piloting of a fishery questionnaire by the World Fish Center (WFC), which was supported by the LSMS-ISA project for the purpose of assembling a fishery questionnaire that could be integrated into multi-topic household-surveys. The WFC piloted the draft instrument in November 2009 in the Lower Shire region, and the NSO team considered the revised draft in designing the IHS4 fishery questionnaire.

COMMUNITY QUESTIONNAIRE The content of the IHS4 Community Questionnaire follows the content of the IHS3 & IHPS Community Questionnaires. A “community” is defined as the village or urban location surrounding the enumeration area selected for inclusion in the sample and which most residents recognize as being their community. The IHS4 community questionnaire was administered to each community associated with the 780 cross-sectional EAs. Identical to the IHS3 approach, to a group of several knowledgeable residents such as the village headman, the headmaster of the local school, the agricultural field assistant, religious leaders, local merchants, health workers and long-term knowledgeable residents. The instrument gathers information on a range of community characteristics, including religious and ethnic background, physical infrastructure, access to public services, economic activities, communal resource management, organization and governance, investment projects, and local retail price information for essential goods and services.

Cleaning operations

DATA ENTRY PLATFORM To ensure data quality and timely availability of data, the IHS4 was implemented using the World Bank’s Survey Solutions CAPI software. To carry out IHS4, 1 laptop computer and a wireless internet router were assigned to each team supervisor, and each enumerator had an 8–inch GPS-enabled Samsung Galaxy Tab S2 tablet computer. The use of Survey Solutions allowed for the real-time availability of data as the completed data was completed, approved by the Supervisor and synced to the Headquarters server as frequently as possible. While administering the first module of the questionnaire the enumerator(s) also used their tablets to record the GPS coordinates of the dwelling units. In Survey Solutions, Headquarters can then see the location of the dwellings plotted on a map of Malawi to better enable supervision from afar – checking both the number of interviews performed and the fact that the sample households lie within EA boundaries. Geo-referenced household locations from that tablet complemented the GPS measurements taken by the Garmin eTrex 30 handheld devices and these were linked with publically available geospatial databases to enable the inclusion of a number of geospatial variables - extensive measures of distance (i.e. distance to the nearest market), climatology, soil and terrain, and other environmental factors - in the analysis.

DATA MANAGEMENT The IHS4 Survey Solutions CAPI based data entry application was designed to stream-line the data collection process from the field. IHS4 Interviews were collected in “sample” mode (assignments generated from headquarters) as opposed to “census” mode (new interviews created by interviewers from a template) for the NSO to have more control over the sample.

The range and consistency checks built into the application was informed by the LSMS-ISA experience in IHS3 and IHPS. Prior programming of the data entry application allowed for a wide variety of range and consistency checks to be conducted and reported and potential issues investigated and corrected before closing the assigned enumeration area. Headquarters (NSO management) assigned work to supervisors based on their regions of coverage. Supervisors then made assignments to the enumerators linked to their Supervisor account. The work assignments and syncing of completed interviews took place through a Wi-Fi connection to the IHS4 server. Because the data was available in real time it was monitored closely throughout the entire data collection period and upon receipt of the data at headquarters, data was exported to STATA for other consistency checks, data cleaning, and analysis.

DATA CLEANING The data cleaning process was done in several stages over the course of field work and through preliminary analysis. The first stage of data cleaning was conducted in the field by the field based field teams utilizing errors generated with the Survey Solutions application. For questions that

Abstract

The Integrated Household Survey (IHS-I) was a comprehensive socio-economic survey of the living standards of households in all districts of Malawi. The National Statistical Office administered the IHS questionnaire to about 12,900 households over a 12 month period, November 1997 to October 1998. The IHS had five main objectives: - · To provide a complete and integrated data set to arrive at a better understanding of households in poverty. · To serve a much broader set of applications on policy issues regarding: Household behaviour and welfare, Distribution of income and expenditure, Employment and Migration, Health, fertility and nutrition, Education and Access to Social facilities. · To provide fresh information on expenditure patterns of households. This information could be useful in the revision of commodity weights for the consumer price indices. · To provide estimates of final household consumption expenditure to serve as a basis for deriving direct estimates in the National Accounts of final household consumption expenditure. · To rationalise data collection, since household surveys were carried out in an uncoordinated manner in the past. The IHS addresses the interests of various users in one integrated data set with inter-linked modules.

Geographic coverage

National

The survey covered households from both the urban and rural areas of the country. The sample coverage of 12,960 households was designed such that it would give an overall relative standard error of 15 percent. These households were chosen from 29 Survey Districts (Statistical abstract Main survey.pdf page 5 ).

Data were collected in monthly rounds of 60 enumeration areas/clusters over a period of 12 months to account for seasonal effects during the year. There were 720 Enumeration Areas/Clusters (EAs) with 20 households being selected from each rural EA and 10 households from each urban EA. The sampling was designed to ensure that at least 240 households were interviewed in each survey district to provide an acceptable level of accuracy for each variable.

Analysis unit

The IHS have the following units of analysis: individuals and households.

Kind of data

Sample survey data [ssd]

Sampling procedure

The survey households were chosen following a multi-stage clustered random sampling approach. The 29 survey districts were grouped into 4 urban and 25 rural districts. All the districts were included in the survey. Separate procedures were followed in urban and rural areas.

This survey was carried out between November 1997 and October 1998 by our counterpart agency, the National Statistical Office, based in Zomba. The desire was to produce information that was representative at the district level. Consequently, all of the 25 administrative districts of the country formed clusters for the first stage of the sampling process. (There are now 26 districts, as one of the 25 was since subdivided.) Additionally, each of the four urban centers of Malawi - Blantyre, Zomba, Lilongwe, and Mzuzu - were treated as a separate statistical district. In total, twenty-nine statistical districts for which representative data was sought were delimited. Slightly different sample selection procedures were used in the rural and urban districts.

Rural district sample selection The next level below the district in the administrative hierarchy of the country is the traditional authority (TA) in rural areas and the ward in urban areas. In each of the 25 rural statistical districts, TAs were randomly selected from comprehensive lists of all TAs in the district. The number of TAs selected was done roughly proportional to population size in the statistical district. In the districts with small populations, only one TA was selected. The median number of TAs selected in a district was two, whereas in Lilongwe district, the most populous district, five TAs were selected. Comprehensive lists of all enumeration areas (EA) - a sub-unit of the TA - in the selected TAs were drawn up. Twelve EAs were randomly selected in each TA. The interview schedule involved interviewing all sample households in one of these twelve EAs each month of the year from November 1997 to October 1998. Comprehensive lists of all households in these selected EAs were then drawn up. Twenty households were randomly selected from these lists in each EA, for a total of 240 households for any one TA.

Urban district sample selection In the urban areas, the wards were ignored. Rather a comprehensive list of the enumeration areas - a sub-unit of the wards - was used for random selection of EAs. The number selected was roughly in proportion to population with the number of EAs being multiples of 12 to reflect the 12 months of the survey year (Blantyre - 60 EAs, Zomba - 24 EAs, Lilongwe - 36 EAs, Mzuzu - 24 EAs). Ten, rather than twenty, households in each of the EAs were then randomly selected from comprehensive lists of households in the selected EAs. Each EA was assigned to be interviewed in one of the twelve months of the survey year, e.g. all of the selected households in five EAs were interviewed each month in Blantyre, three EAs in Lilongwe, and all selected households in two EAs in Zomba and Mzuzu.

Overview The total sample size was 12,960 households, with 11,520 households in rural areas and 1,440 urban households. Each sample household was interviewed at the beginning of one of the twelve months of the survey year. The sample households were provided with a diary of expenditures to complete over the following 28 days. All household expenditures over that period were recorded in the diary. The households were visited every third day during the month to monitor the diary entries.

Mode of data collection

Face-to-face [f2f]

Research instrument

Data were collected using two main forms (Refer appendices). These forms are described in more detail below: - (a) FORM IHS-2: Household Characteristics, Income and Expenditure: This form collected data on almost all the modules of the Integrated Household Survey (37 modules). The data collected included the household roster, vital statistics, fertility and mortality, nutrition and anthropometrics, education, health, household expenditure, crop production and sales, livestock and poultry ownership, non-farm activities and income, assets, employment and migration and access to facilities. (b) Form IHS- 3: Diary of Expenditure: Data on daily expenditure was captured using this type of form. Households were asked to maintain the diary for a period of 28 days.

Cleaning operations

Data entry programs were written in IMPS (Integrated Microcomputer Processing System) and data from the questionnaires was captured onto computers initially at NSO's regional centres, and later at NSO headquarters in Zomba. The output from IMPS was in ASCII format, which had to be translated to SPSS. The data was then cleaned in and tabulated in SPSS and STATA, a process which was largely completed by end May 2000.