Poverty Income Consumption and Expenditure Survey 2011
Zimbabwe, 2011 - 2012
Identification
ZWE_2011_PICES_v01_M
Poverty Income Consumption and Expenditure Survey 2011
| Name | Country code |
|---|---|
| Zimbabwe | ZWE |
Income/Expenditure/Household Survey [hh/ies]
The Zimbabwe National Statistics Agency (ZIMSTAT) formerly the Central Statistical Office (CSO), conducted the 2011/12 Poverty Income and Expenditure Survey (PICES) from June 2011 to May 2012. Previously, this type of survey was called the Income, Consumption and Expenditure Survey (ICES). These surveys were carried out every 5 years. ZIMSTAT carried ICES surveys in 1990/91, 1995/96 and 2001. The ICES for 2007/08 was conducted during an unstable socio-economic period and could not be completed due to hyperinflation. With the introduction of multiple currencies in January 2006, ZIMSTAT saw the need to conduct another ICES in a more stable environment, in order to determine the weights for the household consumption expenditures for the Consumer Price Index. The 2011/12 PICES is the sixth survey of its kind to be conducted in Zimbabwe by ZIMSTAT.
The Income, Consumption and Expenditure Survey is the main data source for the compilation of national accounts aggregates.
The main objectives of the 2011/2012 PICES were to provide data on:
Poverty;
Income distribution of the population;
Consumption level of the population;
Private consumption;
Consumer Price Index (CPI) weights;
Living conditions of the population;
Production account of agriculture (Communal Lands Small Scale Commercial Farms, Resettlement Areas, A1 and A2 farms and Large Scale Commercial Farms).
Sample survey data [ssd]
Households
Individuals
Scope
Data on socio-demographic characteristics, incomes, receipts from households including agriculture, consumption and other expenditures is collected on a weekly basis and for some items on a monthly basis.
The scope of the PICES includes:
- Population and demographic characteristics
- Household economy and poverty
- Individual incomes and remittances
- Agriculture, production account of agriculture, forestry and fishing
- Informal sector in Zimbabwe
Coverage
National
Producers and sponsors
| Name | Affiliation |
|---|---|
| Zimbabwe National Statistics Agency (ZIMSTAT) | Government of Zimbabwe |
| Name | Role |
|---|---|
| United Nations Development Programme | PICES Technical Committee |
| United Nations Children's Fund | PICES Technical Committee |
| African Capacity Building Foundation | PICES Technical Committee |
| African Development Bank | PICES Technical Committee |
| Department for International Development | PICES Technical Committee |
| USAID-SERA Project | PICES Technical Committee |
| World Bank | PICES Technical Committee |
| Ministry of Labour and Social Services | PICES Technical Committee |
Sampling
The 2002 Zimbabwe Population Census Master Sample frame (ZMS202) provided an area sampling frame for the 2011/12 PICES. The survey was based on a sample of 31,248 households which is representative at province and district levels. The sample design entailed two stages: selection of Enumeration Areas (EAs) as the first stage and selection of households in these EAs as the second stage. In total 2,232 EAs were selected with Probability Proportional to Size (PPS), the measure of size being the number of households enumerated in the 2002 Population Census. Finally the number of each of the EAs which were successfully interviewed in the 12 months of the study was 2,220 giving a covering response rate of 99.5 percent. The sample is representative of all the population in Zimbabwe residing in private households. The population living in institutions such as military barracks, prisons and hospitals was excluded from the sampling frame.
Stratification
In order to increase the efficiency of the sample design for PICES 2010/11, it was important to divide the sample design for PICES 2011/12 it was important to divide the sampling frame of EAs into strata which are as homogeneous as possible. At the first sampling stage the sample EAs are selected independently within each explicit stratum. The nature of the stratification depended on the most important characteristics measured in the surveym as well as the domains of analysis. The strata was made consistent with the geographic disaggregation used in the survey tables.
The first level of stratification corresponded to the 60 administrative districts of Zimbabwe, which are the geographic domains of analysis defined for the PICES. The rural and urban areas are domains at the national level. Some of the administrative districts are completely rural or urban, while most districts have a combination of rural and urban EAs. Since many districts have relatively few urban sample EAs, it would not be effective to use explicit urban and rural stratification within each district. Instead, the sampling frame of EAs for each district was sorted first by the rural/urban code in order to provide implicit stratification. Given that the sample EAs were selected systematically with Probabilty Proportional to Size (PPS), this provided a proportional allocation of the sample within each district by rural and urban areas. The sampling frame includes codes for land-use sectors, which can also be used for implicit stratification. The following land-use sextors have been identified:
1- Communal land
2- Small scale commercial farming area
3- Large scale commercial farming area
4- Resettlement area
5- Urban council area
6- Administrative centres (districts)
7- Growth Point
8- Other Urban Area, e.g. Service Centres and Mines
9- State Land, e.g. National Parks, Safari Areas
Sections 1.4 - 1.6 of the survey report (provided as external resources) provide more information on Sample size and allocation, Sample selection and Systematic selection of EAs.
Out of a total of 30,838 households interviewed 29,765 questionnaires were fully completed. Partly completed questionnaires were excluded from the analysis as they would distort average incomes and expenditures.
Based on a total of 29,765 households with fully completed questionnaires the response rate calculated using the original sample is 95.3 percent.
Before analysis was done it was essential to know the total number of questionnaires that were returned by the provinces. A total of 30,838 interviews were conducted and these included partially completed questionnaires. After removing the partually completed questionnaires the number of households which were successfully interviewed in the study were 29,756, giving a response rateof 95.3 percent based on the initial sample of 31,248 households. The households with partially completed questionnaires were left out from the analysis as they would distort averages for variables such as income and expenditures. The response rates were highest in Manicaland Province which had 97.9 percent, followed by Masvingo 97.1 percent. Harare province and Bulawayo province had the lowest response rates of 82.8 percent and 85.6 percent respectively. The main reason for these low response rates in Harare and Bulawayo is a large number of households who are not found at home, refusals and relocation of households to other areas within the month of the survey. This was prevalent particularly in dwelling units occupied by lodgers. The number of partly completed questionnaires was also high in urban areas. In terms of enumeration area coverage, a total of 2,220 EAs were enumerated out of a sample total of 2,232 EAs and this represented a coverage response rate of 99.5 percent of the total number of EAs sampled.
The basic weight of each sample household would be equal to the inverse of its probability of selection (calculated by multiplying the probabilities at each sampling stage).Section 1.7 of the survey report (provided as external resources) provides further details on the weighting factors.
Data collection
| Start | End |
|---|---|
| 2011-06 | 2012-05 |
In order to ensure the quality of data the Head Office and Provincial Supervisors put the following quality control measures in place:
Head office Supervisors
- Carry out spot checks and ensure work is undertaken properly
- Ensure all project work is going according to schedule
- Listen to some selected interviews to check if questions were being asked correctly
- Resolve any challenges encountered in the province promptly
- Document challenges faced in the field and the corresponding solutions used
- Ensure appropriate and required quantities of materials are directed to; and reach the desired destination and are properly utilized
- Re-interview selected households as quality control
- Check that the households were completing the daily record books (DRBs) provided by enumerators
- Ensure that all field staff have received their monthly field allowances
- Write field reports which are used as monitoring tools
Provincial Supervisors
- See to it that enumerators are preforming their work according to instructions
- Ensure that enumerators go to all selected households; and verify cases of non-contact
- Re-interview selected households to ensure complete enumeration
- Check each questionnaire for completeness and consistency before submitting the questionnaires to the Head Office
- Ensure that all questionnaires are submitted to Head Office on time
- Carry out the day-to-day supervision of enumerators' work
- Liaise with Head Office to discuss any technical or logistics problems encountered in the field
- Ensure that enumerators are paid their field allowances
The plan of the 2011/12 PICES was to collect data for a year from sampled households from June 2011 to May 2012. During the first 2-3 months technical problems arose from the field such as how to collect information on certain questions. The questions were addressed through communicating with the Provincial Supervisors. The PICES Technical Committee also met regularly to discuss field work reports and to iron out fieldwork problems. Issues discussed in the committee were then relayed to the provinces by a household surveys coordinator to avoid conflicting instructions.The coordinating team communicated effectively with the Provincial Supervisors through emails and this information was cascaded down to the enumerator in the field. The problems in the field were also minimised because the PICES 2011/12 Enumerator Manual was comprehensive and covered all the likely problems that the enumerator was likely to face.
Data processing
PICES 2011/2012 data was captured by the ZIMSTAT data entry unit and CSPro was used to develop data entry programmes. About 80 people were involved in data processing each month from December 2011 to the end of July 2012. These members of staff worked overtime on average for 20 days in a month. Data was captured twice by different people for purposes of verification. Statistical Analysis System (SAS) was used for data processing programmes. Data cleaning was done at all stages i.e. data entry and data processing to check for the consistency of the data.
Quality Control Measures Used During Data Processing
Data processing involved coding and editing of the questionnaires and data entry. The main reason why data processing was started early was to ensure that data processing is started whilst data collection was in progress. This enabled field staff to be informed of the quality of data collection whilst they were still in the field. It was also found necessary that any queries on the data could be resolved whilst the field staff remembered what transpired. This was also deemed necessary because the number of questionnaires reveived could be checked promptly and discrepancies on the questionnaires received and those expected would be investigated immediately and resolved.
During data processing one member of staff was given 4 batches to be completed in six days. About 80 ZIMSTAT staff members were requested to work outside normal business hours on workdays and on Saturdays. The first two days were for initial entry while the other two days were for verification entry. Two persons exchanged questionnaires during the verification stage. The third stage was to check for differences between the two entries and any errors in initial entry were corrected at that stage. A clean file was then set aside to be copied by programmers at the end of each data processing exercise.
Control sheets were used for monitoring the movement of questionnaires from one person to another during the editing and data processing stage. Any errors made during the data entry were corrected and all data capture operators were informed of these errors to avoid the same errors being repeated. Furthermore, as part of quality control, the data entry programme had inbuilt quality control programmes such as the skip patterns of the questionnaire and the automatic refusal if an unknown identification code (Geocode) or inconsistent code was entered. Data Entry Supervisors also made spot checks to see work being entered while a Statistical Officer was placed in each of the data entry pools to correct any errors or inconsistencies in a process known as "online editing".
In order to check the quality of data processing ZIMSTAT staff began to generate tables to do validity checks using Population Census data for 2002 and other surveys such as Zimbabwe Demographic and Health Survey (ZDHS 2010-11). The Finscope Zimbabwe 2011 Survey Results were also used in validating the data. The validation exercise was done for both the 6 months data and the 12 months data and any deviations from the norm were investigated. An audit of the questionnaires received and processed was also done and any discrepancies were investigated and resolved. ZIMSTAT also compared the geocodes sampled and the geocodes with processed data and any differences were also corrected. As a quality control measure, a Sampling Consultant was engaged to work with ZIMSTAT PICES team to check and review the PICES weights for the 6 months data and 12 months data respectively.
Data Access
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.
Contacts
| Name | Affiliation | URL | |
|---|---|---|---|
| Zimbabwe National Statistics Agency (ZIMSTAT) | info@zimstat.co.zw | www.zimstat.co.zw | |
| World Bank Microdata Library | World Bank |
Metadata production
DDI_ZWE_2011_PICES_v01_M_WB
| Name | Affiliation | Role |
|---|---|---|
| Development Economics Data Group | The World Bank | Documentation of the DDI |
2014-04-08
Metadata version
Version 01 (April 2014)