DD_ZMB_2008_LFS_v01_M_WB <_value> NADA Labor Force Survey 2008 LFS 2008 ZMB_2008_LFS_v01_M Central Statistical Office (CSO) NADA Labor Force Survey [hh/lfs] Zambia The LFS was a nationwide survey covering household population in all the nine provinces and in both rural and urban areas. The analysis has been done at national, rural/urban and provincial levels. However, the dataset has provisions to generate major indicators at district and constituency levels. The survey excluded institutional populations such as those in Hospitals, Barracks, Prisons or Refugee camps, because the survey was intended to focus only on the usual household members - i.e. members who lived together as a household for at least six months or who intended to live together as a household for at least six months - who constituted a household. Sample survey data [ssd] The main questionnairehas eleven sections namely: 1. Background Characteristics. 2. Demographic Characteristics. 3. Education and School Attendance. 4. Economic Activity. 5. Employment. 6. Unemployment. 7. Health and Safety Issues of persons five years and above. 8. Income. 9. Skills Training. 10. Forced Labour. 11. Child Labour. The sample was designed to allow separate estimates for the nation as a whole, and rural and urban areas. The sample design also allowed for indicators to be estimated for each of the nine provinces, 72 districts and 150 constituencies. A representative probability sample of 30,000 households was selected in two stages. In the first stage 1,500 clusters (Enumeration Areas) were selected from a list or frame of enumeration areas compiled from the 2000 Census of Population and Housing. A cluster is the primary sampling unit, which is equivalent to a Standard Enumeration Area (SEA). In the second stage, 20 households were selected from each of the selected enumeration areas. Sampling Frame and Stratification Zambia is administratively divided into nine provinces. Each province is in turn subdivided into districts. For statistical purposes each district is subdivided into Census Supervisory Areas (CSAs) and these are in turn demarcated into Enumeration Areas (EAs). The Census mapping exercise of 1998-2000 in preparation for the 2000 Census of Population and Housing, demarcated the CSAs within wards, wards within constituencies and constituencies within districts. In 2000, Zambia had 72 districts, 150 constituencies, 1,289 wards, about 4,400 CSAs and close to 17,000 SEAs. Information borne in the list of EAs includes number of households and the population size. The number of households determined the selection of primary sampling units (PSU). Therefore, the sample frame of this survey is the list of EAs developed from the 2000 Population Census. The EAs are stratified as urban and rural strata. Sample Allocation and Selection The total sample of 30,000 households out of 2,382,778 households was first allocated between rural, urban and the provincial domains in proportion to the population of each domain according to the 2000 Census results. The proportional allocation does not however, allow for reliable estimates for smaller domains. Adjustments to the proportional allocation of the sample were made to allow reasonable comparison to be achieved between strata or domains. Therefore, disproportionate allocation was adopted, for the purpose of maximizing the precision of survey estimates. The disproportionate allocation is based on the optimal square root allocation method designed by Leslie Kish. The sample was then selected using a stratified two-stage cluster design. After the households were allocated to the different strata, the number of clusters to be selected was calculated based on an average of 20 completed interviews in each of the selected clusters. Clusters were selected systematically with probability proportional to the number of households. Face-to-face [f2f] A technical team comprising of officers from CSO and MLSS designed the questionnaire. The questionnaire was circulated to various stakeholders such as the ILO and IOM for their comments before it was finalised. The main questionnairehas eleven sections namely: 1. Background Characteristics. 2. Demographic Characteristics. 3. Education and School Attendance. 4. Economic Activity. 5. Employment. 6. Unemployment. 7. Health and Safety Issues of persons five years and above. 8. Income. 9. Skills Training. 10. Forced Labour. 11. Child Labour. Pre-test The pre-test for the Labour Force Survey was conducted in August 2008.The objective of the Pre-test was to test the adequacy of the survey instruments and also served as an opportunity to train trainers for the main survey. The participants in the Pre-test included the survey implementation team members and those who were to train in the main training. The Pre-test exercise consisted of two parts. The first part involved training of team members in a classroom set-up while the second part was meant for fieldwork and review of the survey instruments and special-case experiences. The training included role-plays in which participants demonstrated how the interviews could effectively be conducted in both local and English languages. The participants for the pre-test also met after the exercise for a debriefing and shared experiences, which formed a basis for finalizing all the field instruments. Training of Interviewers and Supervisors About 900 persons were recruited by the Central Statistical Office (CSO) to serve as enumerators and supervisors. In total, there were 750 enumerators and 150 supervisors. In addition, there were 18 master trainers, out of whom 9 were from CSO and 9 were from MLSS. They were all trained during the main training which begun in October 2008 in all the provinces. Master trainers there after trained of enumerators and supervisors for a period of two weeks. Training was guided by the enumerators’ instruction manual that was prepared as part of the survey instruments. The method of training involved having enumerators read through the manual and trainers lecturing on different topics in line with the manual’s prescription. Other training modes included class demonstrations - front of class interviews - and interviews in small groups. After classroom training, master trainers had to go for practicals. Fieldwork At least six enumerators were assigned to one supervisor and they formed a team. Depending on the location of the work area, transport was allocated for some field staff especially for all areas that were hard to reach. Other logistics associated with the fieldwork were provided for all the provinces. Data collection was conducted mainly between November and December 2008. The Central Statistical Office (CSO) coordinated the supervision of fieldwork. Trainers visited the field teams during the implementation of fieldwork. The CSO provincial heads/statisticians who also attended training monitored the quality of data in the field and co-ordinated the provision of logistics in the provinces. There was close contact between the field teams, Provincial staff and Headquarters, which was maintained through out fieldwork. In order to ensure reliability and credibility of data collected some quality control measures designed for the survey included formation of a technical team, which had members from CSO, MLSS and other relevant stake holders. These were involved in the planning and implementation of the survey. The trainers monitored fieldwork throughout and the supervisors remained with their assigned teams until the fieldwork was completed and did some basic edits on a daily basis. The review of generated tables from the survey also involved the stakeholders that took part in the planning. Quality control also involved master trainers going back in the field to verify the data that was collected by enumerators. Due to the non-proportional allocation of the sample to the different strata, sampling weights were required to ensure actual representative ness of the sample at national level. The sampling probabilities at first-stage selection of SEAs and probabilities of selecting the households were used to calculate the weights. The weights of the sample are equal to the inverse of the probability of selection. There were a number of challenges at every stage of the survey. At planning stage, not every concept and definition was included but as far as possible, an attempt was made to conform with international standards. Furthermore, not all the selected work areas were enumerated as planned because a few of them had became inaccessible during data collection mainly due to floods resulting from heavy rains and relocation of local residents due to seasonal economic activities such as fishing in Luapula province. Even though alternative replacements of work areas were made, data collection in these work areas somewhat delayed, resulting in having reference periods shifts. Moreover, not all key indicators of the labour market have been analysed in this report such as labour productivity, labour elasticity etc. However, as far as possible, remedial measures have been taken into account to ensure representativeness and accuracy in the results.