The Tanzania National Panel Survey (TZNPS) is the first in a series of nationally representative household panel surveys that assembles information on a wide range of topics including agricultural production, non-farm income generating activities, consumption expenditures, and a wealth of other socio-economic characteristics. The first year of the survey was conducted over twelve months from October 2008 to October 2009. It was implemented by the Tanzania National Bureau of Statistics (NBS). The second wave of the TZNPS is planned for fall 2010.
The main objective of the TZNPS is to provide high-quality household-level data to the Tanzanian government and other stakeholders for monitoring poverty dynamics, tracking the progress of the Mkukuta poverty reduction strategy, and evaluating the impact of other major, national-level government policy initiatives. As an integrated survey covering a wide range of socioeconomic factors, it compliments other more narrowly focused survey efforts such as the Demographic and Health Survey on health, the Integrated Labour Force Survey on labour markets, the Household Budget Survey on expenditure, and the National Sample Census of Agriculture. Secondly, as a panel household survey where households are revisited over time, the TZNPS allows for the study of poverty and welfare transitions and the determinants of living standard changes, rather than only cross-sectional statistics.
NBS was advised on technical issues related to survey design and implementation by the TZNPS Technical Committee, which included representatives from line ministries, government agencies and development partners, such as the Ministry of Agriculture, Food Security and Cooperatives, Ministry of Finance, Millennium Challenge Authority Tanzania, World Bank, DFID, UNICEF, UNFPA, and JICA. The first wave of the TZNPS was supported by several donors, including the World Bank, The Bill and Melinda Gates Foundation (BMGF), UNICEF, UNFPA, and the Royal Danish Embassy, as well as the Government of Tanzania through the pooled Mkukuta funding. NBS also received management and technical support from the LSMS Team in the Development Economics Research Group (DECRG) of the World Bank.
The TZNPS is part of the Living Standards Measurement Study-Integrated Surveys on Agriculture (LSMS-ISA), which supports governments in seven Sub-Saharan African countries to generate nationally representative, household panel data with a strong focus on agriculture and rural development.
This document describes all aspects of the TZNPS 2008/09, including the set of survey instruments, sample design, survey implementation, and the resulting data sets.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Household,Agricultural and Community
v1.1: Edited, anonymous dataset for public distribution.
Non-farm income generating activities,
Wealth of other socio-economic characteristics
Tanzania Mainland, Dar es salaam, other urban areas, Rural areas Zanzi
Unit of Analysis
Household,Agricultural and Community
Producers and sponsors
The Bill and Melinda Gates Foundation
United Nation Children's Funds
United Nation Population Funds
Royal Denish Embassy
The sample size of 3,280 households was calculated to be sufficient to produce national estimates of poverty, agricultural production and other key indicators. It will also be possible in the final analysis to produce disaggregated poverty rates for 4 different strata: Dar es Salaam, other urban areas on mainland Tanzania, rural mainland Tanzania, and Zanzibar. Alternatively, estimates of most key indicators can be produced at the zone level, as used for the Demographic and Health Survey (DHS) reports and other surveys. There are 7 of these zones in total on the mainland: North, Central, Eastern, South, Southern Highlands, West and Lake. As with any survey though, the confidence of the estimates declines as statistics are disaggregated into smaller zones. Due to the limits of the sample size it is not possible to produce reliable statistics at the regional or district level.
The sample was constructed based on the National Master Sample frame which is a list of all populated enumeration areas in the country developed from the 2002 Population and Housing Census. The sample includes a partial sub-sample of households interviewed during the 2006/2007 Household Budget Survey. Sample design was done in spring of 2008.
In total, the target sample was 3,280 households in 410 Enumeration Areas (2,064 households in rural areas and 1,216 urban areas).
In order to produce nationally representative statistics with the NPS data, it is necessary to apply weighting or expansion factors. These survey weights adjust for differences in the probability of selection into the NPS sample for observations in various strata.
The NPS sample is a multi-stage clustered sample design. First stage sampling involved the selection of survey clusters with the probability of selection proportional to cluster size within a stratum. In rural areas a cluster is defined as an entire village. In urban areas a cluster is defined as a census enumeration area. As a general rule, the probability of selection was higher for clusters within strata where existing data sources showed that the variance of key variables of interest for the NPS (e.g., household consumption and maize production) were likely to be very high - implying the need for more observations to produce reliable estimates. In practice, this implies that a disproportionate number of clusters were selected in urban areas including Dar es Salaam, and thus a larger weight or expansion factor is placed on rural observations in the NPS in order to reconstruct national averages.
The expansion factors can be found in the Weights data file. There are two variables of interest included: the first, “hh_weight” gives the expansion factor needed to produce nationally representative estimates. The second, “hh_weight_trimmed” is identical to the first, except that the largest and smallest 1% of observations in have been censored at the 1st or 99th percentile of ethe “hh_weight” distribution, respectively. This censoring, or “trimming” of the weights has little or no measurable effect on statistical means calculated with the NPS data, but significantly reduces the estimated standard errors or confidence intervals of sample statistics.
The Weights data file also contains unique identifiers for the first-stage sampling units (“clustered”) and for the sampling strata (“strataid”). First stage sampling of clusters was stratified along two dimensions: (i) eight administrative zones (seven on Mainland Tanzania plus Zanzibar as an eighth zone), and (ii) rural versus urban clusters within each administrative zone. The combination of these two dimensions yields 16 strata.
Dates of Data Collection (YYYY/MM/DD)
Mode of data collection
The field staff was trained in Korogwe in September 2008 over four weeks, and a second training was held over three weeks after six months of field work in April 2009. The second training was necessitated by a shift in implementation strategy - from having dedicated household and agricultural enumerators to having all enumerators trained to administer both surveys. In practice, this had be the de facto implementation strategy for most teams since the start of the survey, and the main goal of the standardizing techniques across all teams.
Type of Research Instrument
The first wave of the TZNPS consists of several survey instruments: a Household Questionnaire, Agriculture Questionnaire, and Community Questionnaire.
The Household Questionnaire is comprised of thematic sections. A detailed description of the contents of the questionnaire can be found in Table 1. This comprehensive questionnaire allows for the construction of a full consumption-based welfare measure, permitting distributional and incidence analysis. This project also recognizes the imperative to look beyond the household as a unit of analysis in order to improve the quality, relevance and sustainability of agricultural data systems. Although data collection is structured around a household panel survey, the data on labor, education and health status were collected at the individual level. Moreover, in some household activities (like non-farm enterprise), the questionnaire records which specific members are engaged in the activity.
The questionnaires were developed in collaboration with line ministries and donor partners, including the Technical Committee, over a period of several months. The NBS solicited feedback from various stakeholders in regards to survey content and design. The questionnaires were piloted in the Morogoro region in June 2008, in conjunction with supervisor training. After piloting, the questionnaires were further revised and finalized by September 2008.Questionnaire manuals were developed with detailed instructions for field staff for use during training and as a reference over the course of the field work.
The Agricultural Questionnaire collects information relative to a household's agricultural activities. Information is collected at both the p lot and crop level on inputs, production and sales. Table 2 provides a detailed description of the contents of the questionnaire. This questionnaire was administered to any household that engaged in any farming, livestock or fishery activities.
The Community Questionnaire collects information on physical and economic infrastructure and events in surveyed communities, as described in Table 3. The respondents for this questionnaire vary by individual location but generally include the Village/Block Chairperson, the Village/Ward Executive officer, and several sub-village chairpeople. Information about these respondents is collected individually in section CH of community questionnaire. This questionnaire is administered by the field team supervisors in all communities included in the sample. The questionnaire section on market prices was revised mid-year to improve the formatting and, therefore, the quality of the price data. So there are two different price data sets in the community data reflecting the change in formatting of the price section.
National Burteau of Statistics
Data collect or
Data entry was done concurrently with data collection by the data entry technician, using a laptop. The data entry program was a CSPro-based system, developed by NBS with support from the World Bank. This facilitated the performance of internal crosschecks prior to departure from the enumeration area, allowing enumerators to return to households and clarify inconsistent information on the questionnaires. Data files from completed EAs were then e-mailed to headquarters using 3G modems. These files were concatenated and periodic checks were done to ensure the fieldwork was proceeding according to the calendar.
Following the completion of fieldwork and data-checks during data entry, an extensive review of data files was conducted, including verification against the paper questionnaire of outliers and missing values. Observations were returned for manual inspection of the physical questionnaires if:
· Values fell beyond five standard-deviations from the sample mean.
· Total consumption was less than the sum of purchased and home-produced consumption.
· The unit of measure was inconsistent across the questions for a single consumption item in Section K (i.e., purchased amounts were recorded in kilograms but total mounts were recorded in grams).
· Non-zero monetary expenditure on a given item was accompanied by missing values for total consumption.
When it was determined that these values were the result of data-entry error, the values were corrected. In addition, cases deemed to reflect obvious enumerator error were also corrected in this cleaning process. The majority of such cases involved the use of incorrect measurement units, e.g. recording grams as kilograms or vice versa.