““A new Approach to Improving A new Approach to Improving Sample Design for Crop Forecast Sample Design for Crop Forecast and Post – Harvest Estimates in and Post – Harvest Estimates in

Zambia”Zambia”A Contributed Paper Session Presented at the A Contributed Paper Session Presented at the

International Conference on Agricultural International Conference on Agricultural Statistics (ICAS III), Westin Regina Hotel, Statistics (ICAS III), Westin Regina Hotel,

Cancun, Q.R. MexicoCancun, Q.R. Mexico2nd-4th November 20042nd-4th November 2004

ByByJohn KalumbiJohn Kalumbi

Deputy Director – Agriculture & Environment Deputy Director – Agriculture & Environment DivisionDivision

Central Statistical Office (CSO)Central Statistical Office (CSO)Lusaka, ZambiaLusaka, Zambia

I. INTRODUCTION AND I. INTRODUCTION AND INSTITUTIONAL ARRANGEMENTINSTITUTIONAL ARRANGEMENT

Production of Agricultural StatisticsProduction of Agricultural Statistics• Responsibility of the CSO (Agric. & Environment Responsibility of the CSO (Agric. & Environment

Division)Division)• Collection of data is done in conjunction with the Collection of data is done in conjunction with the

Ministry of Agriculture and Cooperatives (MACO) Ministry of Agriculture and Cooperatives (MACO) through the Early Warning and Data Base through the Early Warning and Data Base Management Unit.Management Unit.

Agriculture & Environment DivisionAgriculture & Environment Division

Agriculture BranchAgriculture Branch• Small & Medium Scale Farm SectionSmall & Medium Scale Farm Section• Large Scale Farm SectionLarge Scale Farm Section

Environment BranchEnvironment Branch• Land degradationLand degradation• Air pollutionAir pollution• Water SanitationWater Sanitation• ForestryForestry• WildlifeWildlife• FisheriesFisheries

Evolution of Agricultural Surveys in ZambiaEvolution of Agricultural Surveys in Zambia

The 1970/71 Census of Agriculture marked the The 1970/71 Census of Agriculture marked the beginning of annual agricultural surveys in beginning of annual agricultural surveys in Zambia.Zambia.• Crop Forecast Survey (CFS)Crop Forecast Survey (CFS)• Post Harvest Survey (PHS)Post Harvest Survey (PHS)• Area Measurement and Crop-Cutting Survey (AMCC) – Area Measurement and Crop-Cutting Survey (AMCC) –

has since been discontinuedhas since been discontinued

II. METHODOLOGY USEDII. METHODOLOGY USED

• Survey sampling for both CFS and PHS for Survey sampling for both CFS and PHS for Small & Medium Scale agricultural holdingsSmall & Medium Scale agricultural holdings

• Complete enumeration for Large-Scale Agric. Complete enumeration for Large-Scale Agric. HoldingsHoldings

Sample DesignSample Design• Cartographic census field operation defines Cartographic census field operation defines

Census Supervisory Areas (CSAs), which are Census Supervisory Areas (CSAs), which are further subdivided into Standard Enumeration further subdivided into Standard Enumeration Areas (SEAs)Areas (SEAs)

• An SEA is the unit of data collection point at An SEA is the unit of data collection point at district leveldistrict level

Previous CFS/PHS Sample Previous CFS/PHS Sample DesignDesign

Based on 1990 census dataBased on 1990 census data Only rural districts were coveredOnly rural districts were covered A stratified three stage sample design was usedA stratified three stage sample design was used

• CSAs Primary Sampling Units (PSUs) selected at first CSAs Primary Sampling Units (PSUs) selected at first stage with PPSstage with PPS

• SEAs selected at second stage, with one SEA selected SEAs selected at second stage, with one SEA selected with PPS within each sample CSAwith PPS within each sample CSA

• Measure of size in both cases is number households as Measure of size in both cases is number households as listed in the 1990 census of population and Housing.listed in the 1990 census of population and Housing.

• A total sample of 405 SEAs was drawn out of 9,000 A total sample of 405 SEAs was drawn out of 9,000 rural SEAsrural SEAs

• No provision of replacement of both sample SEAs and No provision of replacement of both sample SEAs and sample households.sample households.

• Sample households were selected at third stage from Sample households were selected at third stage from the listing stratified by farm size category using Linear the listing stratified by farm size category using Linear Systematic Sampling (LSS) procedure with random Systematic Sampling (LSS) procedure with random start.start.

New CFS/PHS Sample DesignNew CFS/PHS Sample Design Based on 2000 census dataBased on 2000 census data A stratified two-stage sample design is now being A stratified two-stage sample design is now being

usedused• SEAs are the Primary Sampling Units (PSUs) SEAs are the Primary Sampling Units (PSUs)

selected at first stage with PPSselected at first stage with PPS• Measure of size is number of agricultural Measure of size is number of agricultural

households as listed in the 2000 census of households as listed in the 2000 census of population and housingpopulation and housing

• A total sample of 410 SEAs were drawn out of A total sample of 410 SEAs were drawn out of the 12,788 agricultural SEAs as defined for the the 12,788 agricultural SEAs as defined for the 2000 Zambia Census2000 Zambia Census

• Sample households are selected at the second Sample households are selected at the second stage from the listing stratified by farm size stage from the listing stratified by farm size category (land area, number of animals, special category (land area, number of animals, special crops criteria)crops criteria)

Other Features of New CFS /PHS Other Features of New CFS /PHS Sample DesignSample Design

Includes a small fraction (5%) of urban SEAs, i.e, Includes a small fraction (5%) of urban SEAs, i.e, urban SEAs in which 70 percent or more households urban SEAs in which 70 percent or more households are agricultural according to the 2000 Zambia are agricultural according to the 2000 Zambia census data (about 586 urban SEAs in total)census data (about 586 urban SEAs in total)

Has provision for the replacement of any sample Has provision for the replacement of any sample SEAs that may not be enumerated for one reason or SEAs that may not be enumerated for one reason or another in order to maintain the effective sample another in order to maintain the effective sample size.size.

Eight (8) crops were identified to receive special Eight (8) crops were identified to receive special treatment at second sampling stage in order to treatment at second sampling stage in order to improve the precision of survey estimates of crop improve the precision of survey estimates of crop area and production (i.e., these are crops with area and production (i.e., these are crops with limited geographical concentration whose CVs limited geographical concentration whose CVs previously were relatively high due to the smaller previously were relatively high due to the smaller number of observations – sorghum, rice, cotton, number of observations – sorghum, rice, cotton, burley tobacco, Virginia tobacco, sunflower, burley tobacco, Virginia tobacco, sunflower, soybeans, and paprika)soybeans, and paprika)

III.III. STRATIFICATION OF STRATIFICATION OF HOUSEHOLDSHOUSEHOLDS

Previous sample designPrevious sample design Included households listed in sample SEAs by two Included households listed in sample SEAs by two

farm size categories:farm size categories:• Category A: 0 – 4.99 hectaresCategory A: 0 – 4.99 hectares• Category B: 5.0 – 19.99 hectaresCategory B: 5.0 – 19.99 hectares

New sample designNew sample design Stratification of households listed in sample SEAs is Stratification of households listed in sample SEAs is

based onbased on• Farm size (land area) categoryFarm size (land area) category• Number of livestock or poultry raised criteriaNumber of livestock or poultry raised criteria• Growing of special crops criteriaGrowing of special crops criteria

Stratification of households by Stratification of households by farm size (land area)farm size (land area)

• Category A: 0 – 1.99 hectaresCategory A: 0 – 1.99 hectares• Category B: 2.00 – 4.99 hectaresCategory B: 2.00 – 4.99 hectares• Category C: 5.00 – 19.99 hectaresCategory C: 5.00 – 19.99 hectares

In both the previous and new sample In both the previous and new sample design, households with farm size design, households with farm size category of 20 or more hectares were/are category of 20 or more hectares were/are included in the special frame for large-included in the special frame for large-scale farmers and are enumerated scale farmers and are enumerated separately.separately.

Stratification of households by Stratification of households by number of livestock or poultry number of livestock or poultry

raised criteriaraised criteria Based on the number of livestock or Based on the number of livestock or

poultry raised, households are added poultry raised, households are added to category C if the following to category C if the following minimum number are being raised:minimum number are being raised:• Cattle – 50Cattle – 50• Pigs – 20Pigs – 20• Goats – 30Goats – 30• Poultry – 50Poultry – 50

Stratification of households by Stratification of households by growing of special crops criteriagrowing of special crops criteria

Households may be added to category B and Households may be added to category B and C based on the special crops treatment C based on the special crops treatment criteriacriteria• If the sample SEA only has 1 or 2 households If the sample SEA only has 1 or 2 households

with any of these special crops, these with any of these special crops, these households will be asigned to category C ( in households will be asigned to category C ( in case they do not qualify based on land area case they do not qualify based on land area and animals)and animals)

• If the sample SEA has only 3 to 5 households If the sample SEA has only 3 to 5 households with any of these special crops, such with any of these special crops, such households are assigned to category B (if they households are assigned to category B (if they were previously assigned to category A based were previously assigned to category A based on land area and livestock)on land area and livestock)

IV. SAMPLE HOUSEHOLDS IV. SAMPLE HOUSEHOLDS SELECTION IN SAMPLE SEAsSELECTION IN SAMPLE SEAs

Total sample size for each SEA – 20 householdsTotal sample size for each SEA – 20 households Distribution of sample households to each category:Distribution of sample households to each category:

• Category C – 10Category C – 10• Category B – 5Category B – 5• Category A – 5Category A – 5

If there are shortfall in category C, the difference from 20 will If there are shortfall in category C, the difference from 20 will be allocated equally to categories B and A or category B will be allocated equally to categories B and A or category B will have one more sample households than category A, if the have one more sample households than category A, if the difference cannot be allocated equally.difference cannot be allocated equally.

Where there is no household in category C and less than 10 in Where there is no household in category C and less than 10 in category B, the difference from twenty will be met from category B, the difference from twenty will be met from category A.category A.

If all listed households fall in category A, then all 20 sample If all listed households fall in category A, then all 20 sample households will come from category A.households will come from category A.

• For each category in the sample SEA, selection of sample For each category in the sample SEA, selection of sample households is done using Linear Systematic Sampling (LSS) households is done using Linear Systematic Sampling (LSS) Procedure with random start.Procedure with random start.

V. ESTIMATIONV. ESTIMATION

Estimating for the total involves the weighing of variables Estimating for the total involves the weighing of variables before additions are made at district, provincial and before additions are made at district, provincial and national levels. The basic sampling weight is obtained as national levels. The basic sampling weight is obtained as inverse of the probabilities of selecting the SEAs and inverse of the probabilities of selecting the SEAs and households by category size.households by category size.

The formula for estimating the survey estimates of a total is The formula for estimating the survey estimates of a total is as follows:as follows:

^̂Y = Y = W Wshi shi Y Yshijshij

h i s jh i s j

Where: Where: • YYshij shij = = value of variable y for the j-th sample household in value of variable y for the j-th sample household in

category s within the i-th sample SEA in district category s within the i-th sample SEA in district hh

• WWshishi = = basic weight for the sample households in category s basic weight for the sample households in category s within the i-th sample SEA in district h.within the i-th sample SEA in district h.