Annual Agricultural Survey 2018-2019
Senegal, 2018 - 2019
Directorate of Agricultural Analysis, Forecasting and Statistics
Created on
November 12, 2020
Last modified
November 12, 2020
Page views
2115
Download
263
Documentation in PDF
Metadata
DDI/XML
JSON
Identification
Survey ID Number
SEN_2018-2019_AAS_v01_EN_M_v01_A_OCS
Title
Annual Agricultural Survey 2018-2019
Country
| Name | Country code |
|---|---|
| Senegal | SEN |
Study type
Agricultural Survey [ag/oth]
Abstract
The agricultural survey in its current version covers all regions of the country and all departments (with the exception of the departments of Dakar, Pikine and Guédiawaye, which are excluded from the scope of the survey because of the weakness or even non-existence of agricultural activity). The agricultural survey is an annual statistical operation whose general objective is to estimate the level of the main agricultural production of family-type farms. It also makes it possible to provide information on the physical characteristics of cultivated plots of land (geo location, surface area) and the major investments made at their level (agricultural inputs, cultivation operations, soil management and restoration). It also addresses, once every 3 years, themes relating to the structure of agricultural households (level of agricultural equipment, agricultural income, agricultural risks and adaptation strategies, etc.).The main indicators relate to yield levels, areas sown, production and means of production.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Agricultural holdings
Scope
Notes
The scope of the study was covered using two questionnaires:
(a) QUESTIONNAIRE 1: CENSUS OF HOUSEHOLD MEMBERS AND HOUSEHOLDS' LIVING QUARTERS:
Section 1: Household identification
Section 2: Census of household members
Section 3: Census of plots operated by the household during the 2018 crop year
Section 4: Plots cultivated in the previous crop year
Section 5: Monitoring of plots operated by the household
Section 6: Survey of the areas of the plots operated by the household
(b) QUESTIONNAIRE 2 - HOUSEHOLD:
Section 1: General farm information
Section 2a: 2018 Rainfed Crop Production
Section 2b: Horticultural Production in 2018
Section 4: Inputs for agricultural production
Section 5: On-farm processing of agricultural products (Marketing of agricultural and processed products)
Section 6a: Participation of household members in the 2018 crop year
Section 6b: Farm Worker Participation in the 2018 Crop Year
Section 7: Livestock
Section 8: Inputs for animal livestock
Section 9: Animal products and their marketing
Section 10: Fisheries and Aquaculture
Section 11: Forest Production
Section 12: Agricultural equipment
Section 13: Shocks and Strategies to deal with production shocks
Section 14: Other Expenditure, investment, finance and insurance
Section 15: Marketing and Storage
Section 16: Other sources of income
(a) QUESTIONNAIRE 1: CENSUS OF HOUSEHOLD MEMBERS AND HOUSEHOLDS' LIVING QUARTERS:
Section 1: Household identification
Section 2: Census of household members
Section 3: Census of plots operated by the household during the 2018 crop year
Section 4: Plots cultivated in the previous crop year
Section 5: Monitoring of plots operated by the household
Section 6: Survey of the areas of the plots operated by the household
(b) QUESTIONNAIRE 2 - HOUSEHOLD:
Section 1: General farm information
Section 2a: 2018 Rainfed Crop Production
Section 2b: Horticultural Production in 2018
Section 4: Inputs for agricultural production
Section 5: On-farm processing of agricultural products (Marketing of agricultural and processed products)
Section 6a: Participation of household members in the 2018 crop year
Section 6b: Farm Worker Participation in the 2018 Crop Year
Section 7: Livestock
Section 8: Inputs for animal livestock
Section 9: Animal products and their marketing
Section 10: Fisheries and Aquaculture
Section 11: Forest Production
Section 12: Agricultural equipment
Section 13: Shocks and Strategies to deal with production shocks
Section 14: Other Expenditure, investment, finance and insurance
Section 15: Marketing and Storage
Section 16: Other sources of income
Topics
| Topic | Vocabulary |
|---|---|
| Agriculture & Rural Development | FAO |
| Land (policy, resource management) | FAO |
| Labor | FAO |
| Livestock | FAO |
| Disaster Risk Management | FAO |
| Access to Finance | FAO |
Coverage
Geographic Coverage
National coverage
Producers and sponsors
Primary investigators
| Name | Affiliation |
|---|---|
| Directorate of Agricultural Analysis, Forecasting and Statistics | Ministry of Agriculture and Rural Development |
Producers
| Name | Affiliation | Role |
|---|---|---|
| Horticulture Department | Ministry of Agriculture and Rural Development | Support in questionnaire development, investigator training and supervision |
| National Agency for Statistics and Demography | Ministry of Economy, Finance and Planning | Support in questionnaire development, investigator training and supervision |
| Studies and Planning Unit of the Ministry of Livestock | Ministry of Livestock and Animal Production | Support in questionnaire development, investigator training and supervision |
Funding Agency/Sponsor
| Name | Abbreviation | Role |
|---|---|---|
| Food and Agriculture Organization of the United Nations | FAO | Technical and Financial Assistance |
| Government of Senegal | GoS | Technical and Financial Assistance |
Sampling
Sampling Procedure
(a) THE SAMPLING FRAME
The results of the latest RGPHAE include 755,532 farming households engaged in agriculture in the broadest sense. On this basis 455,916 agricultural households practice rain-fed agriculture. The agricultural household file will be used as the sampling frame for the first degree DR draws. The second stage sampling frame consists, at the level of each primary unit (PU) or DR drawn, of the exhaustive list of agricultural households living there and representing the secondary units (US).
(b) SAMPLE DESIGN
The sample design includes a global sample of 6340 farm households in 1260 DRs and the 42 fields of study (department). The distribution of the sample considers the overall sampling rates and the agricultural weight of the stratum. The selected sampling design refers to a two-stage random survey with national coverage, which admits rural census districts (DRs) as primary units (PU) and agricultural households as secondary units (US). The method consists of distributing the statistical population (farm households) into the primary units so that each of them is unambiguously linked to a specific CU. The sample is then drawn in two stages:
- In the first stage (1st stage), a sample of primary units (PR) is drawn,
- In the second stage (2nd stage), a sample of secondary units is selected from each primary unit (DR) drawn in the 1st stage.
The advantage of this method is that it is not necessary to have an exhaustive list of secondary units for the whole national territory, but only those residing in the sample primary units. This reduces travel and consequently the costs of the survey. The sample draws will be organised independently from one field of study (department) to another. The primary units or DRs are drawn with unequal probabilities and with discount (PIAR draw). The probability of exit from a DR at each draw is chosen in proportion to its size expressed in terms of the number of farm households. At the second level, secondary units (or farm households) are drawn with equal probabilities and without discount (PESR draw). A constant number of secondary units is selected from each DR in the first-stage sample. This constant number of secondary units is chosen equal to 5.For the economic module, two households were drawn in each DR. Exactly 2362 sampled households and 2321 households were covered.
More information on the weighting calculation is available in the note on methodology in the external resources.
The results of the latest RGPHAE include 755,532 farming households engaged in agriculture in the broadest sense. On this basis 455,916 agricultural households practice rain-fed agriculture. The agricultural household file will be used as the sampling frame for the first degree DR draws. The second stage sampling frame consists, at the level of each primary unit (PU) or DR drawn, of the exhaustive list of agricultural households living there and representing the secondary units (US).
(b) SAMPLE DESIGN
The sample design includes a global sample of 6340 farm households in 1260 DRs and the 42 fields of study (department). The distribution of the sample considers the overall sampling rates and the agricultural weight of the stratum. The selected sampling design refers to a two-stage random survey with national coverage, which admits rural census districts (DRs) as primary units (PU) and agricultural households as secondary units (US). The method consists of distributing the statistical population (farm households) into the primary units so that each of them is unambiguously linked to a specific CU. The sample is then drawn in two stages:
- In the first stage (1st stage), a sample of primary units (PR) is drawn,
- In the second stage (2nd stage), a sample of secondary units is selected from each primary unit (DR) drawn in the 1st stage.
The advantage of this method is that it is not necessary to have an exhaustive list of secondary units for the whole national territory, but only those residing in the sample primary units. This reduces travel and consequently the costs of the survey. The sample draws will be organised independently from one field of study (department) to another. The primary units or DRs are drawn with unequal probabilities and with discount (PIAR draw). The probability of exit from a DR at each draw is chosen in proportion to its size expressed in terms of the number of farm households. At the second level, secondary units (or farm households) are drawn with equal probabilities and without discount (PESR draw). A constant number of secondary units is selected from each DR in the first-stage sample. This constant number of secondary units is chosen equal to 5.For the economic module, two households were drawn in each DR. Exactly 2362 sampled households and 2321 households were covered.
More information on the weighting calculation is available in the note on methodology in the external resources.
Response Rate
The response rates are 94% at the cluster level and 89% at the household level.
Weighting
“poids_men”: this is the adjusted household weighting coefficient which was used in the study.
Note: The weighting calculations can be found in the file "Note sur le calcul des coefficients de pondération".
Note: The weighting calculations can be found in the file "Note sur le calcul des coefficients de pondération".
Data Collection
Dates of Data Collection
| Start | End | Cycle |
|---|---|---|
| 2018-08-20 | 2018-11-01 | 1st round |
| 2019-01-01 | 2019-02-13 | 2nd round |
Data Collection Mode
Computer Assisted Telephone Interview [cati]
Data Processing
Data Editing
(a) DATA ENTRY PLATFORM
To ensure data quality and real-time availability, the AAS 2018-2019 was implemented using the World Bank's Survey Solutions CAPI software with version 20.07.03. In order to carry out all the interviews, a local server was installed to manage the interviews and assignments, as well as the creation of user accounts for the supervisors (42) and interviewers (156). At the level of the collection system, each interviewer was trained in the use of the survey solution interviewer collection application. Following this training, the interviewers received a tablet (Samsung, techno), the Garmin 64 tool for measuring plot surface areas and an internet connection for receiving assignments and synchronising interviews on the server. Also the control system at department level, consisting of the heads of department, was trained on the use and management of the Headquaters. To this end, each head of department was given an internet connection to approve or reject interviews.
(b) DATA MANAGEMENT
The AAS 2018-2019 data entry application has been designed to streamline the data collection process in the field. The study interviews were collected in "sample" mode (assignments generated from headquarters). Logical and consistency checks built into the data entry application (interviewer) helped to minimise possible errors in the information collected from the respondent. The headquarters assigned the work to the interviewers according to the coverage of the households to be surveyed. Once the assignments were made, the interviewers synchronized to receive their assignments and proceeded with the administration of the interviews. Each interview was completed and sent to the supervisor's area in his or her department for verification. The verification has two states: approve the interview or reject the interview. Once the interviews are approved by the supervisors, a database is built and can be exported in various formats (stata, spss, tab). For the AAS 2018-2019 database, the data was exported to STATA for further consistency checks, data cleaning and analysis.
(c) DATA CLEANING
The data cleansing process took place in different stages:
- Preliminary processing consists of programming a set of consistency checks in the SUSO application to prevent interviewers from entering certain outliers or incorrect values.
- A second level of control is carried out by supervisors based at the level of the departmental rural development services (sddr) and at the level of DAPSA. It consists of opening each interview to check the completeness and consistency of responses. Supervisors should synchronize the enumerators' shelves as often as possible to avoid having many questionnaires on the shelf and to allow daily checks of the questionnaires. Some supervisors preferred to review completed interviews on the shelves to revise them before synchronization, while recording the notes in the supervisor's account and rejecting the questionnaires accordingly.
- The final stage of data cleaning is carried out by the central DAPSA team using statistical processing software (STATA and SPSS). It consists of exporting all the data from the platform in order to have an overview of the data. During this stage, a distinction is made between two main types of checks: checks on the form of the observations and logical or linkage checks between observations.
* To access thematic tables of the study, refer to the data platform link (DAPSA) provided in external materials under 'other materials' of the documentation section.
To ensure data quality and real-time availability, the AAS 2018-2019 was implemented using the World Bank's Survey Solutions CAPI software with version 20.07.03. In order to carry out all the interviews, a local server was installed to manage the interviews and assignments, as well as the creation of user accounts for the supervisors (42) and interviewers (156). At the level of the collection system, each interviewer was trained in the use of the survey solution interviewer collection application. Following this training, the interviewers received a tablet (Samsung, techno), the Garmin 64 tool for measuring plot surface areas and an internet connection for receiving assignments and synchronising interviews on the server. Also the control system at department level, consisting of the heads of department, was trained on the use and management of the Headquaters. To this end, each head of department was given an internet connection to approve or reject interviews.
(b) DATA MANAGEMENT
The AAS 2018-2019 data entry application has been designed to streamline the data collection process in the field. The study interviews were collected in "sample" mode (assignments generated from headquarters). Logical and consistency checks built into the data entry application (interviewer) helped to minimise possible errors in the information collected from the respondent. The headquarters assigned the work to the interviewers according to the coverage of the households to be surveyed. Once the assignments were made, the interviewers synchronized to receive their assignments and proceeded with the administration of the interviews. Each interview was completed and sent to the supervisor's area in his or her department for verification. The verification has two states: approve the interview or reject the interview. Once the interviews are approved by the supervisors, a database is built and can be exported in various formats (stata, spss, tab). For the AAS 2018-2019 database, the data was exported to STATA for further consistency checks, data cleaning and analysis.
(c) DATA CLEANING
The data cleansing process took place in different stages:
- Preliminary processing consists of programming a set of consistency checks in the SUSO application to prevent interviewers from entering certain outliers or incorrect values.
- A second level of control is carried out by supervisors based at the level of the departmental rural development services (sddr) and at the level of DAPSA. It consists of opening each interview to check the completeness and consistency of responses. Supervisors should synchronize the enumerators' shelves as often as possible to avoid having many questionnaires on the shelf and to allow daily checks of the questionnaires. Some supervisors preferred to review completed interviews on the shelves to revise them before synchronization, while recording the notes in the supervisor's account and rejecting the questionnaires accordingly.
- The final stage of data cleaning is carried out by the central DAPSA team using statistical processing software (STATA and SPSS). It consists of exporting all the data from the platform in order to have an overview of the data. During this stage, a distinction is made between two main types of checks: checks on the form of the observations and logical or linkage checks between observations.
* To access thematic tables of the study, refer to the data platform link (DAPSA) provided in external materials under 'other materials' of the documentation section.
Other Processing
The AES 2018-2019 microdata contain information on households and individuals and therefore need to be anonymised before they can be used for the analysis. Attached to the external materials is the full procedure used in the anonymization process.
Data Appraisal
Data Appraisal
1. Error correction
The controls have made it possible on the one hand to check that no formal errors remain in the database and on the other hand to reduce the errors of funds even if they are difficult to correct. Two types of corrections are used: automatic correction and manual correction. It should be noted that these corrections are made on outliers and partial non-responses. The automatic correction in the form of a programme to be executed is preferred because it allows to keep a traceability of all the corrections made. Corrections by standard profile, by average, nearest neighbour or by ratio are often used to correct or impute certain values.
2. Treatment of total non-responses
Total non-response occurs when no information is provided on a household. These cases do not exist in the database because they have been managed from the beginning. If the respondent does not consent to answer the questions, he or she is replaced by another household.
The controls have made it possible on the one hand to check that no formal errors remain in the database and on the other hand to reduce the errors of funds even if they are difficult to correct. Two types of corrections are used: automatic correction and manual correction. It should be noted that these corrections are made on outliers and partial non-responses. The automatic correction in the form of a programme to be executed is preferred because it allows to keep a traceability of all the corrections made. Corrections by standard profile, by average, nearest neighbour or by ratio are often used to correct or impute certain values.
2. Treatment of total non-responses
Total non-response occurs when no information is provided on a household. These cases do not exist in the database because they have been managed from the beginning. If the respondent does not consent to answer the questions, he or she is replaced by another household.
Access policy
Contacts
| Name | Affiliation | URL | |
|---|---|---|---|
| Agricultural Documentation and Information Statistics Division | Directorate of Agricultural Analysis, Forecasting and Statistics | [email protected] | http://www.dapsa.gouv.sn |
| Mme Fall Sylvie Dasylva | Directorate of Agricultural Analysis, Forecasting and Statistics | [email protected] | http://www.dapsa.gouv.sn |
Confidentiality
The microdata of the Annual Agricultural Survey are confidential. They are anonymized to preserve the confidentiality of the identity of respondents.
Access conditions
Any use of the data produced or managed by DAPSA must be subject to an authorization request addressed to its director. It will clearly indicate the use that will be made of the data and a copy in French or in English of the study project must be attached to the request. The requesting service or organization must undertake to comply with the following conditions:
1. The user must comply with the provisions of articles 6 and 7 chapter 2 section 2 of the Senegalese law No 2004.21 relating to the organization of statistical activities which obligation for any person handling individual data, collected by the services and bodies under the national statistical system, to guarantee the anonymity of the natural or legal persons concerned by the survey, and to use this data only for the purpose of disseminating or publish aggregated statistical results.
2. The user is fully responsible for his conclusions or studies drawn from these data and in this,
3. The data must not be copied or transmitted to other persons or organizations, directly or indirectly, without the prior written consent of DAPSA.
4. Any publication written using data from the DAPSA must include the following statement "Source: Survey X *, Year N *, Directorate of Analysis, Forecasting and Agricultural Statistics (DAPSA) of the Republic from Senegal, www.dapsa.gouv.sn "
5. Three copies of any report produced on the basis of the data must be sent for information and comments to DAPSA.
6. The user is informed that DAPSA reserves the right to use any means it deems useful in the event of non-compliance with any of these commitments.
7. The user can, if necessary, enlist the assistance of DAPSA in carrying out the study.
*: X being the full name of the survey followed by its acronym in parentheses, example "Annual agricultural survey (EAA)".
N being the year of the survey, example "2002".
1. The user must comply with the provisions of articles 6 and 7 chapter 2 section 2 of the Senegalese law No 2004.21 relating to the organization of statistical activities which obligation for any person handling individual data, collected by the services and bodies under the national statistical system, to guarantee the anonymity of the natural or legal persons concerned by the survey, and to use this data only for the purpose of disseminating or publish aggregated statistical results.
2. The user is fully responsible for his conclusions or studies drawn from these data and in this,
3. The data must not be copied or transmitted to other persons or organizations, directly or indirectly, without the prior written consent of DAPSA.
4. Any publication written using data from the DAPSA must include the following statement "Source: Survey X *, Year N *, Directorate of Analysis, Forecasting and Agricultural Statistics (DAPSA) of the Republic from Senegal, www.dapsa.gouv.sn "
5. Three copies of any report produced on the basis of the data must be sent for information and comments to DAPSA.
6. The user is informed that DAPSA reserves the right to use any means it deems useful in the event of non-compliance with any of these commitments.
7. The user can, if necessary, enlist the assistance of DAPSA in carrying out the study.
*: X being the full name of the survey followed by its acronym in parentheses, example "Annual agricultural survey (EAA)".
N being the year of the survey, example "2002".
Citation requirements
(a) English version:
Source: "Annual agricultural survey (AEE), 2018/2019, Directorate of Analysis, Forecasting and Agricultural Statistics (DAPSA) of the Republic of Senegal, www.dapsa.gouv.sn"
(b) French version:
Source: "Enquête agricole annuelle (AEE), 2018/2019, Direction de l'Analyse, de la Prévision et des Statistiques agricoles (DAPSA) de la République du Sénégal, www.dapsa.gouv.sn"
Source: "Annual agricultural survey (AEE), 2018/2019, Directorate of Analysis, Forecasting and Agricultural Statistics (DAPSA) of the Republic of Senegal, www.dapsa.gouv.sn"
(b) French version:
Source: "Enquête agricole annuelle (AEE), 2018/2019, Direction de l'Analyse, de la Prévision et des Statistiques agricoles (DAPSA) de la République du Sénégal, www.dapsa.gouv.sn"
Disclaimer and copyrights
Disclaimer
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
Metadata production
DDI Document ID
DDI_SEN_2018-2019_AAS_v01_EN_M_v01_A_OCS_FAO
Producers
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| Office of Chief Statistician | OCS | Food and Agriculture Organization | Adoption of metadata for FAM |
| Directorate of Agricultural Analysis, Forecasting and Statistics | DAPSA | Ministry of Agriculture and Rural Development | Study documentation |
DDI Document version
SEN_2018-2019_AAS_v01_EN_M_v01_A_OCS_v01