i
Capacity Needs Assessment for Improving
Agricultural Statistics in Uganda
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
ii
Standard Disclaimer:
This volume is a product of the staff of the International Bank for Reconstruction and Development/The
World Bank with external contributions. The findings, interpretations, and conclusions expressed in this
paper do not necessarily reflect the views of the Executive Directors of The World Bank or the
governments they represent. The World Bank does not guarantee the accuracy of the data included in this
work. The boundaries, colors, denominations, and other information shown on any map in this work do
not imply any judgment on the part of The World Bank concerning the legal status of any territory or the
endorsement or acceptance of such boundaries.
Copyright Statement:
The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work
without permission may be a violation of applicable law. The International Bank for Reconstruction and
Development/The World Bank encourages dissemination of its work and will normally grant permission
to reproduce portions of the work promptly.
For permission to photocopy or reprint any part of this work, please send a request with complete
information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA,
telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/.
All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of
the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-
mail [email protected]
iii
Table of Contents
Preface ......................................................................................................................................................... vi
Acknowledgement .................................................................................................................................... viii
Acronyms .................................................................................................................................................... ix
Executive Summary .................................................................................................................................... 1
Chapter 1: Introduction ............................................................................................................................. 4
Agricultural Statistics and the Minimum Set of Core Statistics ............................................................... 4
Need for Agricultural Statistics in Uganda ............................................................................................... 6
Users of Agricultural Statistics in Uganda ................................................................................................ 7
Agricultural Statistics Support and Best Practices for Developing Countries .......................................... 7
Capacity Assessments in Uganda ........................................................................................................... 10
Purpose of This Report ........................................................................................................................... 11
Chapter 2: The Agricultural Statistics System in Uganda .................................................................... 12
UBOS’s Role in the NASS ..................................................................................................................... 12
MAAIF’s Role in the NASS ................................................................................................................... 14
Other Agency Contributions to the NASS .............................................................................................. 17
Local Government Contributions to the NASS ...................................................................................... 18
Current Sources of Uganda Agricultural Statistics ................................................................................. 18
Censuses .................................................................................................................................................. 18
Crop and Livestock Statistics .................................................................................................................. 19
Forestry Statistics .................................................................................................................................... 19
Fisheries and Aquaculture Statistics ....................................................................................................... 19
Agricultural Markets and Price Information Systems ............................................................................. 19
Water and Environment Statistics ........................................................................................................... 19
Rural Development Statistics .................................................................................................................. 19
Food Security and Nutrition .................................................................................................................... 20
Chapter 3: Methodology ........................................................................................................................... 22
Identification of Key Stakeholders ......................................................................................................... 22
Initial Desk Review and Stakeholder Consultation ................................................................................ 22
Panel Discussions and Key Informant Interviews .................................................................................. 22
Standard Assessment Questionnaire ....................................................................................................... 23
Participatory Local Organizational Assessment Interview ..................................................................... 24
Limitations of the Study .......................................................................................................................... 25
Chapter 4: Findings from the Assessment .............................................................................................. 27
Themes from Stakeholders Interviews .................................................................................................... 27
Capacity Assessment of DAES and MAAIF .......................................................................................... 29
iv
Capacity Assessment at the District Level .............................................................................................. 31
Chapter 5: Challenges in the Uganda NASS .......................................................................................... 36
Institutional ............................................................................................................................................. 36
Methodological ....................................................................................................................................... 37
Scant Statistics at the District Level ....................................................................................................... 37
Personnel ................................................................................................................................................. 38
Technological .......................................................................................................................................... 38
Financial .................................................................................................................................................. 38
Chapter 6: Recommendations for Strengthening Agricultural Statistics in Uganda ......................... 40
Chapter 7: Global Best Practices for Agricultural Data ....................................................................... 51
Country Example of Agricultural Data Collection and Survey Programs .............................................. 51
Rwanda ................................................................................................................................................... 51
South Africa ............................................................................................................................................ 53
Sweden .................................................................................................................................................... 55
Best Practices for Agricultural Data: Probability Samples and Two-stage Multiframes ........................ 57
International Initiatives That Can Be Leveraged to Build Capacity Around Agricultural Statistics ...... 58
Collaborations between the Public and Private Sectors .......................................................................... 60
Technology and Quality Assurance Standards ....................................................................................... 61
Bibliography .............................................................................................................................................. 64
Appendix 1: Documents Reviewed .......................................................................................................... 66
Appendix 2: Cost Assumptions. ............................................................................................................... 67
List of Figures
Figure 2.1: Current Structure of the Uganda NASS ..................................................................... 12 Figure 2.2: DAES Organizational Structure ................................................................................ 14
Figure 2.3: Division of Statistics Organizational Structure ......................................................... 16
Figure 6. 1: Harmonized Uganda NASS....................................................................................... 41
Figure 7.1: Organogram of the Ministry of Agriculture, Forestry, and Fishing, South Africa .... 54 Figure 7.2: Distribution of Budget Shares .................................................................................... 59
List of Tables
Table 1.1: Minimum Set of Core Data ............................................................................................ 4
Table 2. 1: Status of the Minimum Set of Core Statistics Collected within the Uganda NASS ... 20
Table 3.1: Key Stakeholders Interviewed ..................................................................................... 23
Table 3.2: ASCI Classification ..................................................................................................... 24
v
Table 3.3: Core Functions Examined in the PLOCA ................................................................... 25
Table 3.4: Districts Surveyed ........................................................................................................ 25
Table 4.1: Standard Assessment ASCIs for DAES and MAAIF .................................................. 30 Table 4.2: Districts Scores for Different Measures of Statistical Capacity .................................. 32
Table 4. 3: Correlation Between Core Functions Critical to Organizational Performance at the
District Level ................................................................................................................................ 35
Table 6. 1: Recommendations for a Harmonized NASS .............................................................. 42
Table 7.1: Land use strata codes, definition, and areas ................................................................ 52 Table 7. 2: Projected AGRISurvey Budget................................................................................... 59 Table A2.1: Exchange Rate .......................................................................................................... 68
Table A2.2: Workshop, Seminar, and Meeting Costs .................................................................. 68
Table A2.3: Consulting Costs ....................................................................................................... 69
Table A2.4: Staffing Costs ............................................................................................................ 69
Table A2.5: Office Costs .............................................................................................................. 70
Table A2.6: Advertising Costs ...................................................................................................... 70
Table A2.7: Office Equipment Costs ............................................................................................ 70
Table A2.8: Meeting and Workshop Costs ................................................................................... 71
List of Boxes
Box 7.1: Sampling frames for agricultural statistics ..................................................................... 58
Box 7.2: Use of technology in collecting agricultural data .......................................................... 62
Box 7.3: Use of technology in data dissemination: Examples of publishers that are Data
Documentation Initiative compliant and of data visualization tools ............................................ 63
vi
Preface
Agriculture is the main source of livelihood for about two thirds of Africa’s population. It
accounts for 70% of employment, overwhelmingly on small farms; occupies half of all land area,
and provides half of all exports and one-quarter of GDP in Uganda. It is considered a leading
sector for future economic growth and economic inclusion in the current National Development
Plan. Thus, enhancing its performance is central to food security and sustainable poverty
reduction. According to Uganda Vision 2040, agriculture contributed approximately 21 percent
of the gross domestic product (GDP) and employed roughly 65 percent of the labor force in 2010
(National Planning Authority 2013).
The agricultural sector in Africa is however faced with increasing demand for agricultural data,
but the Agricultural Planning Department and National Statistical Agencies have many
challenges in making the required data available. There is lack of capacity to provide reliable
statistical data on food and agriculture and to provide a blueprint for long-term sustainable
agricultural statistical systems. A Partnership in Statistics for Development in the 21st Century
(PARIS21)1 review found that only 10% of International Development Association (IDA)
countries2 had included agriculture in the National Strategies for Development of Statistics
(NSDS) process. Even so, agriculture-related NSDS quality is very low as reflected in
agricultural policy and development in most IDA countries. It is imperative that these challenges
of reliable and accurate statistics are addressed.
A number of development organizations are working with various developing countries to
improve their agriculture statistics systems. For example, the World Bank is actively working
with the Government of Uganda to improve the quality and quantity of agricultural statistics
through the Living Standards Measurement Survey (LSMS) - Uganda National Panel Survey
(UNPS). “UNPS is a national panel household survey that has been collecting multi-sectoral
micro data with a strong focus on agriculture. FAO on the other hand supports the Government
of Uganda to generate reliable and detailed information on the nature of food security and
malnutrition for decision making through implementation of the Integrated Food Security Phase
Classification (IPC) in Uganda. FAO facilitates the analysis of food security using the IPC
1 A Partnership in Statistics for Development in the 21st Century (PARIS21) is a global partnership of national, regional, and international
statisticians, analysts, policy-makers, development professionals, and other users of statistics. The PARIS21 Consortium was established as a
global forum and network to promote, influence, and facilitates statistical capacity development and the better use of statistics. 2 The International Development Association (IDA) is the part of the World Bank that helps the world’s poorest countries. Overseen by 173
shareholder nations, IDA aims to reduce poverty by providing loans (called “credits”) and grants for programs that boost economic growth,
reduce inequalities, and improve people’s living conditions.
vii
Analytical Protocols resulting in the availability of up-to-date and reliable food security
information, which is used for planning and early warning. AfDB has developed “Country
Assessment of Agricultural Statistical Systems in Africa - Measuring the Capacity of African
Countries to Produce Timely, Reliable, and Sustainable Agricultural Statistics”.
It is crucial for developing countries to develop their agriculture statistics systems as it is a
critical resource for public policy analysis and design, policy implementation and monitoring,
and decision making. Further, they provide a key input into other statistics, including the national
accounts. For this reason, agricultural statistics need to be comprehensive, reliable, up-to-date,
consistent, and available in a form that renders them intelligible and usable (FAO. 2011)
viii
Acknowledgement
This report summarizes the findings and recommendations of the Capacity Needs Assessment for
Improving Agricultural Statistics led by Ademola Braimoh at the World Bank and conducted by
Frederick Smith, Michael Jacobsen and Karis McGill of RTI International and Paul Kibwika,
Joseph Mugagga Sengendo, Richard Kibombo, Florence Birungi Kyazze, and Rosemirta Birungi
of Development Research and Social Policy Analysis Center (DRASPAC). The work was
carried out under the overall guidance of Diarietou Gaye, Trichur Balakrishnan, Christina
Malmberg Calvo and Dina Umali-Deininger.
The team gratefully acknowledges the support to this work by Patrick Okello, Flavia Oumo,
Contace Nakiyemba, Emmanuel Menyha, Daphne Arinda, Israel Nsiko and Mulmina Maloru
(UBOS); Richard Ndikuryayo, Medard Nabaasa, Efulansi Mutesi, Jovan Lubega, Steven
Kayongo, Agnes Nagayi, Kyagaba Ssekimwany (MAAIF); Tonny Odokonyero, Mildred
Barungi, Francis Mwesigye and Swaibu Mbowa (EPRC); Juma Ndhokero, Jimmy Semakula
and Losira Nasirumbi Sonya (NARO); Emmanuel Iyamulemye Niyibigira, James Kizito-
Mayanja and Samuel Samson Omwa (UCGA); John Diisi, Julius Ariho, Ssenyonjo Edward and
Joseph Mutyaba (NFA); Caleb Gumisiriza and Mwenda Augustin (UNFFE); Robert Kalyebara
and Paul Dhabunansi (aBi Trust); Yamagami Keisuke and Lubega Paul (JICA); Nangulu Moses
and Nabbosa Maxensia (UNADA); Martin Fowler and Ochieng (USAID); Kamugisha Godwin
(NEMA); Martin Emau, Edward Tanyima and Andrew Ateny (FAO); and Vuzzi Azza Victor
and Joyce Alaro (DANIDA).
The report benefited from invaluable suggestions from peer reviewers. We would like to thank
Johan Mistiaen, Forhad Shilpi; Talip Kilic; Carolina Mejia, John Ilukor; Joanne Gaskell. Special
thanks are due to Gandham Ramana, Holger Kray, Kevin Crockford, Joseph Oryokot and Jane
Nalunga for the support provided to this work.
We are also grateful to all the stakeholders who attended the validation workshop for their active
engagement and for the valuable inputs and assistance from Damalie Nyanja and Janet Christine
Atiang of the World Bank.
ix
Acronyms
aBi Agricultural Business Initiative
AfDB African Development Bank
ASCI Agricultural Statistics Capacity Indicator
ASSP Agricultural Sector Strategic Plan
BOU Bank of Uganda
CAPI Computer-Assisted Personal Interview
CDO Cotton Development Organization
CIO Chief Information Officer
COA Census of Agriculture
DAES Directorate of Agriculture and Environment Statistics
DANIDA Danish International Development Agency
DDA Dairy Development Authority
DHS Demographic and Health Survey
DRASPAC Development Research and Social Policy Analysis Center
EPRC Economic Policy Research Centre
FAO Food and Agriculture Organization of the United Nations
GDP Gross Domestic Product
GIS Geographic Information System
HR Human Resources
ICT Information and Communication Technology
ITC Informational and Computational Technology
JICA Japan International Cooperation Agency
KII Key Informant Interview
LSMS Living Standards Measurement Survey
M&E Monitoring and Evaluation
MAAIF Ministry of Agriculture, Animal Industries, and Fisheries
MAFAP Monitoring African Food and Agricultural Policies
NAADS National Agricultural Advisory Services
NAGRIC National Animal Genetic Resource Centre
NARO National Agricultural Research Organization
NASS National Agricultural Statistics System
NASTC National Agricultural Statistics Technical Committee
NDP2 Second National Development Policy
NEMA National Environment Management Authority
NFA National Forestry Authority
NFASS National Food and Agricultural Statistics System
NGO Nongovernmental Organization
NPHC National Population Household Census
NSDS National Strategy for the Development of Statistics
NSS National Statistical System
NSSF National Social Security Fund
ODA Official Development Assistance
PARIS21 Partnership in Statistics for Development in the 21st Century
PLOCA Participatory Local Organizational Assessment
PNSD Plan for National Statistical Development
RAADRS Routine Agricultural Administrative Data Reporting System
SAQ Standard Assessment Questionnaire
SSPS Sector Strategic Plan for Statistics
x
UBOS Uganda Bureau of Statistics
UCDA Uganda Coffee Development Authority
UCGA Uganda Coffee Growers Association
UN United Nations
UNADA Uganda National Agro-inputs Dealers’ Association
UNFFE Uganda National Farmers' Federation
UNHS Uganda National Household Survey
UNPS Uganda National Panel Survey
USAID U.S. Agency for International Development
UTCC Uganda Trypanosomiasis Control Council
WCA World Program for the Census of Agriculture 2020
1
Executive Summary
Agriculture is a key driver of Uganda’s economy accounting for 70% of employment, providing
half of all exports, and one-quarter of GDP in Uganda. Thus, enhancing its performance is
central to food security and sustainable poverty reduction. Recent policies have called for
analyzing and monitoring the growth of the agricultural sector. To accomplish this, policy
makers have identified the need for a strong agricultural statistics system to collect and
disseminate timely, accurate, and relevant statistics.
Ugandan agricultural statistics are used by numerous entities both within and outside of Uganda.
Policy and decision makers within the national government use statistics to enable effective
governing. Agricultural statistics are also needed for planning, administration, monitoring, and
accounting at subnational level. Good statistics are required for exploring the profitability of
agribusiness opportunities, planning and investment, monitoring, evaluation, and reporting of
business activities. Non-Governmental Organizations use statistics to plan, implement, monitor,
and evaluate their activities. They also use statistics to monitor and inform government policy,
lobby politicians, hold governments accountable, and report to their key stakeholders. Lastly,
development partners use a country’s agricultural statistics to determine the need for and impact
of assistance or the requirements for participation in development initiatives.
The current system struggles to provide this level of data. To improve the system so that it can
provide the appropriate data, a capacity needs assessment was undertaken to determine areas of
improvements to the current system. The purpose of this report is to describe the assessment, its
findings, and the recommendations for updating Uganda’s agricultural statistics system.
This assessment was conducted between June 2017 and March 2018. Both external and internal
stakeholders were interviewed to determine the challenges and opportunities facing Uganda’s
agricultural statistics system. Officials from 16 districts were surveyed regarding their district’s
ability to collect and produce agricultural statistics. Representatives from the Uganda Bureau of
Statistics (UBOS) and the Ministry of Agriculture, Animal Industries, and Fisheries (MAAIF)
answered the Standard Assessment Questionnaire (SAQ), and a snapshot of the current capacity
of these institutions to produce agricultural statistics was obtained. A draft report was prepared in
November and a stakeholder workshop was held on March 1, 2018, to get feedback and
information from both users and developers of agricultural statistics in Uganda.
Common themes arose from the stakeholder interviews. These themes, listed below, describe a
disharmonized system that fails to produce the necessary statistics.
• Different agencies create different systems to produce agricultural statistics due to lack of
clarity regarding institutional mandates.
• There is little faith in the reliability of the current agricultural statistics system.
• Administrative data is collected and compiled without employing standard statistical
procedures. There is also an issue of untimely and incomplete flow of data from the lower
to the higher reporting levels
2
• Methodologies for collecting commodity-specific statistics are not adequate.
• There is a lack of investment and prioritization of agricultural statistics.
The districts expressed varying levels of capacity for collecting agricultural data from their
farmers. Districts such as Mayugi and Kaabong reported having moderate capacity, while other
districts, such as Masaka and Mbarara, reported only a basic capacity for collecting agricultural
data. The districts tended to report moderate capacity in the following categories: the mandate,
governance structure, management, and personnel. However, they did not feel they had sufficient
financial resources or the capacity for public dissemination and publicity.
The above findings point to three major challenges within Uganda’s agricultural statistics
system:
1. There are multiple agencies that collect and disseminate agricultural statistics and there
are challenges to build coordination and cooperation between them due to lack of clarity
on institutional mandates.
2. Human capacity constraints hinder the collection of credible statistics at local and
national levels
3. The types of statistics that are considered to be official are neither clearly defined nor the
methodology required to collect them standardized
Therefore, it is recommended that the current structure of Uganda’s agricultural statistics system
should become more harmonized and that an office be created to collect and produce subnational
agricultural statistics. The recommendations for accomplishing these goals are listed below:
Institutional:
• Establish the Global Strategy core minimum set of statistics as the set of official
agricultural statistics.
• Delineate the responsibilities between agencies for collecting the core minimum set of
statistics.
• Develop a coordination committee for agencies that produce agricultural statistics.
• Establish working committees that codify methodologies for collecting the core minimum
set of statistics.
• Provide training to personnel on emerging methodologies to estimate statistics and data
collection.
Methodological:
• Develop commodity-specific methodologies for the collection of agricultural statistics.
• Implement methodologies for improving agricultural statistics from administrative data.
District:
3
• Strengthen the capacity for collection and dissemination of district-level agricultural
statistics by building the required human capacity.
• Monitor district mandate, prioritization, and funding for collection of district-level
statistics.
• Promote the utility and benefit of agricultural statistics to farmers.
Personnel:
• Hire qualified statisticians in UBOS DAES to effectively manage the production of
agricultural statistics. Both UBOS and MAIFF require a cadre of agricultural statisticians
that are highly qualified in data production and properly trained in the latest survey
methods for core data needs, analysis and reporting.
• Promote agricultural statistics to statistics students.
Technological:
• Develop the information communication technology strategy for the collection, analysis,
and dissemination of agricultural statistics. It is however important to properly test the
technologies for their suitability and reliability before they are fully rolled out.
• Create a database of agricultural statistics for data users.
• Update the current computer and network systems within agencies.
• Update the software for data collection and analysis.
Financial:
• The Ugandan government must establish and maintain funding for agricultural statistics
and data collection.
There are two windows for World Bank support for improving agricultural statistics in Uganda.
The first window is through agriculture projects under the MAAIF, with the Agricultural Cluster
Development Project being restructured to include a provision for strengthening the Statistics
Unit. The second window is through Statistics Payment for Results (PforR) Program for
generating better and more accessible data to inform policy-makers and contributing to
strengthening statistical capacity. Funding through these windows can be used to support four
key interventions: (i) developing the legislative framework for agricultural statistics; (ii)
developing the legislative framework for data sharing between county governments and MoALF;
(iii) establishing structures where users and producers of agricultural statistics interact; and (iv)
developing a Seasonal Agricultural Survey (SAS).
4
Chapter 1: Introduction
The Government of Uganda has identified agriculture as a key driver of economic growth and
stability for Uganda. The Uganda Vision 2040 indicates that agriculture contributed
approximately 21 percent of the gross domestic product (GDP) and employed roughly 65 percent
of the labor force in 2010 (National Planning Authority 2013). Because agriculture is such an
important part of the GDP, the Government of Uganda is prioritizing goals that will transform its
agriculture from subsistence to a commercial system. However, progress on reaching these goals
must be measured and evaluated using agricultural statistics.
Agricultural Statistics and the Minimum Set of Core Statistics
Agricultural statistics measure the agriculture industry and farm and rural households. Data users
rely on agricultural statistics to answer different questions and inform decisions and actions at
the political, academic, institutional, and individual levels. Because agriculture is affected by
economic, environmental, and social factors, agricultural statistics must measure the impact of
agriculture on issues within and across these factors. Additionally, agriculture includes other
activities such as agroforestry, land usage, and aquaculture. Therefore, the set of statistics
considered agricultural statistics are broad and multifaceted.
Agricultural statistics can encompass a variety of estimates, each created for different purposes,
such as regulation, enforcement, enterprise management, environment, social, and economic
factors. In this report, agricultural statistics are defined as those outlined in the Minimum Set of
Core Data framework created by the Global Strategy for Improving Agriculture and Rural
Statistics of the Food and Agriculture Organization of the United Nations (FAO) (Table 1.1).
This set of statistics covers many estimates that can be used for other purposes.
Table 1.1: Minimum Set of Core Data
Group of Key
Variables Key Variables Core Data Items
Economic
Output Production Core crops (for example, wheat and rice)
Core livestock (for example, cattle, sheep, and pigs)
Core forestry products
Core fishery and aquaculture products
Area harvested and planted Core crops (for example, wheat and rice)
Yield/births/productivity Core crops, core livestock, core forestry, core fishery
Trade Exports in quantity and value Core crops, core livestock, core forestry, core fishery
Imports in quantity and value Core crops, core livestock, core forestry, core fishery
Stocks Quantities in storage at beginning of harvest Core crops
Stock of resources Land cover and use Land area
Economically active population Number of people in working age by sex
Livestock Number of live animals
Machinery Number of tractors, harvesters, seeders, and other
equipment
5
Group of Key
Variables Key Variables Core Data Items
Inputs Water Quantity of water withdrawn for agricultural
irrigation
Fertilizers in quantity and value Core fertilizers by core crops
Pesticides in quantity and value Core pesticides (for example, fungicides, herbicides,
insecticides, and disinfectants) by core crops
Seeds in quantity and value By core crops
Feed in quantity and value By core crops
Agro processing Volume of core crops/livestock/fisheries used
in processing food
By industry
Value of output of processed food By industry
Other uses (for example, biofuels)
Prices Producer prices Core crops, core livestock, core forestry, core fishery
Consumer prices Core crops, core livestock, core forestry, core fishery
Final expenditure Government expenditure on agriculture and
rural development
Public investments, subsidies, and other expenditure
Private investments Investment in machinery, research and development,
and infrastructure
Household consumption Consumption of core crops/livestock/and so on in
quantity and value
Rural infrastructure
(capital stock)
Irrigation/roads/railways/communications Area equipped for irrigation/roads in km/railways in
km/communications
International transfer Official development assistance (ODA) for
agriculture and rural development
Social
Demographics of urban
and rural population
Sex
Age in completed years By sex
Country of birth By sex
Highest level of education completed One-digit International Standard Classification of
Education by sex
Labor status Employed, unemployed, and inactive by sex
Status in employment Self-employment and employee by sex
Economic sector in employment International standard industrial classification by sex
Occupation in employment International standard classification of occupations
by sex
Total income of the household
Household composition By sex
Number of family/hired workers on the
holding
By sex
Housing conditions Type of building, building character, main material,
and other information
Environmental
Land Soil degradation Variables will be based on above core items on land
cover and use, water use, and other inputs to
production Water Pollution due to agriculture
Air Emissions due to agriculture
Geographic location
6
Group of Key
Variables Key Variables Core Data Items
Geographic
information system
(GIS) coordinates
Location of the statistical unit Parcel, province, region, country
Degree of urbanization Urban/rural area
Source: World Bank 2010.
Need for Agricultural Statistics in Uganda
Agricultural statistics are recognized among policy makers, governmental officials, researchers,
farmers organizations, agribusinesses, and private donors in Uganda as critical to agriculture-
driven economic stability and improvement.
The Uganda Second National Development Policy (NDP2) has identified several weaknesses
within the Ugandan agricultural value chain: low production, little technological innovation and
adoption, a weak agricultural extension service, an inability to match producers with their final
markets, and limited market and production information. One of the goals put forth in the NDP2
was to measure and address the gaps along the value chain (Government of Uganda 2015).
In their ‘Review of Food and Agricultural Policies in Uganda 2005–2011’ report, the FAO
identified the lack of reliable statistics as one of the weaknesses of the current system (MAFAP
2013).
The Uganda National Agriculture Policy calls for investment in agricultural statistics. It defines
the need for a ‘functional system’ that includes all ministries collecting agricultural statistics and
the district governments. Furthermore, it directs the Ministry of Agriculture, Animal Industries,
and Fisheries (MAAIF) to build an agricultural statistics and management system for use in
monitoring and evaluation (M&E) (MAAIF 2011). MAAIF has established a Division of
Statistics to develop and harmonize a system for administrative data3 collection, storage,
analysis, and dissemination to stakeholders.
The Uganda Bureau of Statistics (UBOS) has also outlined the creation of the Directorate of
Agriculture and Environment Statistics (DAES) within its 2013/14–2017/18 Sector Strategic
Plan for Statistics (SSPS). The DAES was created in 2011, and it is responsible for agricultural
and environmental data collection, management, and dissemination (UBOS 2014d).
3 Administrative data refers to non-statistical sources of information obtained through, for example, government programs or agricultural extension, and can benefit the final statistical product in terms of reduced costs or improved small area estimates. An area that could facilitate better linkages between UBOS and MAAIF is the integration of administrative data with household and farm survey data – but an impediment to this is the lack of publicly available, unit-record administrative data. These problems are better articulated under the section MAAIF’s role in the NASS
7
Users of Agricultural Statistics in Uganda
As outlined in the Plan for National Statistical Development (PNSD) 2013/14–2017/18, Ugandan
agricultural statistics are used by numerous entities both within and outside of Uganda. Each
group in the list below requires different sets of statistics to fulfill the users’ various needs
(UBOS 2014d).
• National and local government: Policy and decision makers within the national
government use statistics to enable effective governing. Agricultural statistics are also
needed for planning, administration, monitoring, and accounting.
• Agribusinesses and other economic parties: Good statistics are required for exploring
the profitability of future business opportunities, planning and investment, monitoring,
evaluation, and reporting of business activities.
• Nongovernmental organizations (NGOs): NGOs use statistics to plan, implement,
monitor, and evaluate their activities. They also use statistics to monitor and inform
government policy, lobby politicians, hold governments accountable, and report to their
key stakeholders.
• Media: The media present statistics within their reports and articles to inform the public
about agriculture and report on various developments within the field.
• Research institutions: Researchers rely on good statistics to plan experiments, conduct
research, and present findings.
• Regional organizations: Organizations that foster regional integration and development
use statistics for this cause.
• International organizations: International groups use a country’s agricultural statistics
to determine the need for and impact of assistance or the requirements for participation in
development initiatives.
• General public: The general public requires high-quality statistics to educate themselves
and make decisions that will provide a meaningful impact on their lives.
Agricultural Statistics Support and Best Practices for Developing Countries
In recent years, many international aid organizations have highlighted the growing need for
agricultural statistics in the developing country context, which has led to the institution of global
plans for improving agricultural statistics, such as the Partnership in Statistics for Development
in the 21st Century (PARIS21), the Global Strategy for Improving Rural and Agricultural
Statistics, and the FAO World Program for the Census of Agriculture 2020 (WCA). The goal of
all these programs is to assist developing countries in building a National Agricultural Statistics
System (NASS) that provides useful, accurate, and timely agricultural statistics to national and
international data users.
The mandate of PARIS21 Initiative is “to reduce poverty and improve governance in developing
countries by promoting the integration of statistics and reliable data in the decision-making
process” (PARIS21 2016a). This is accomplished through promoting coordination between data
8
providers, data producers, and data users; improving the use of timely and useful statistics;
assisting with the creation of a National Strategy for the Development of Statistics (NSDS)
document for each participating country; and providing documentation and data archiving
services. All these services are available for each subsector within the National Statistical System
(NSS), including agricultural statistics.
The United Nations (UN) Statistical Commission, FAO, World Bank, and various governmental
agencies that produce and utilize agricultural statistics created the Global Strategy for Improving
Rural and Agricultural Statistics within the FAO to assist developing nations in creating and
disseminating agricultural statistics. The purpose of the Global Strategy is to provide “a
framework and methodology that will lead to an improvement in terms of the quantity and
quality of national and international food and agricultural statistics to guide policy analysis and
decision-making in the 21st strategy” (World Bank 2010). The Global Strategy spells out three
main tasks for countries:
1. Produce a minimum set of core data;
2. Better integrate agriculture into the NSS; and
3. Improve governance and statistical capacity building
The Global Strategy is closely aligned with the creation of the NSDS in that it provides an
assessment of the country’s NASS and where improvements can be made. The findings from the
assessment are used to build the country’s Strategic Plan for Agriculture and Rural Statistics, a
critical piece of the NSDS.
The FAO has been providing assistance through the WCA for countries to develop and conduct a
census of agriculture (COA) since the 1930s. The goal of the WCA is to assist countries in
developing and conducting a COA using standardized methodologies that are internationally
accepted. Every 10 years, the world program is updated to include the latest methodologies and
concepts that all countries can implement. The FAO has recognized the COA as one of the key
components for the Global Strategy.4
The World Bank Living Standards Measurement Study – Integrated Surveys on Agriculture (LSMS-ISA)
initiative has been providing financial and technical support to UBOS since 2009 towards the design,
implementation, analysis and dissemination of the Uganda National Panel Survey (UNPS). UNPS is a
national longitudinal survey that has been collecting multi-sectoral microdata that has a strong focus on
agriculture, and that has been at the heart of rigorous research that has reported state of the agricultural
sector and its linkages to a broad range of development outcomes. Thus far, the UNPS 2009/10, 2010/11,
2011/12, 2013/14, 2015/16 and 2017/18 waves have been implemented (with the anonymized unit-record
microdata being made available within 12 months of completion of fieldwork); the UNPS 2018/19 is
underway at the time of the writing of this report; and the UNPS 2019/20 is in the pipeline.
4 http://www.fao.org/world-census-agriculture/wcarounds/wca2020/en
9
Similar to the approach pursued in other African countries that have been supported by the LSMS-ISA
initiative, the World Bank’s investments into UBOS capability to produce and analyze high-quality
microdata have been leveraged to initiate a parallel program of methodological experiments in the areas
of land area measurement, soil fertility assessment, crop production and yield measurement, crop variety
identification, and remote sensing, and ownership and control of physical and financial assets, with the
idea of developing improved survey methods with downstream linkages to the future UNPS rounds.
These experiments yield peer-reviewed methodological research outputs that are distilled into guidelines
for implementing best practices in data collection, which are in turn used as reference documents by
survey practitioners, including UBOS and the national statistical offices supported by the LSMS-ISA
(Kilic 2017).
Best practices and guidelines have been developed by each of the above-mentioned groups to
assist developing countries in collecting agricultural data and disseminating agricultural
statistics. Guidebooks and technical papers on various statistical and methodological aspects are
published on each group’s website for free use. Each group has given advice on different aspects
of agricultural statistics. The Global Strategy has produced best practices for
• Developing a master sample frame, with examples from different developing and
developed countries;
• Designing fishery survey modules in household surveys;
• Enumerating nomadic and seminomadic livestock counts;
• Costing production surveys and grain stocks surveys;
• Improving crop production forecast surveys
• Remote sensing;
• Creating food balance sheets;
• Calculating gender-based estimates;
• Making international classifications; and
• Providing data users access to agriculture microdata.
• The World Bank LSMS produces guidelines on household and farm survey data
collection on a range of agricultural and non-agricultural topics, anchored primarily
in randomized survey experiments that inform peer-reviewed academic research,
which in turn feed into these guidelines. Relevant to agricultural statistics, the
guidelines are currently available for survey data collection on:
• Land areas
• Soil fertility
• Livestock
10
• Forestry
• Fisheries
• Asset ownership
At the time of the writing of the report, the LSMS, in collaboration with its national and
international partners, was also working on guidelines for survey data collection on
agricultural labor, annual crop production, extended-harvest crop production, crop variety
identification, and remote sensing for measuring crop yields. The LSMS is also part of the
World Bank Development Data Group – Survey Unit, which develops the free, computer-
assisted personal interviewing (CAPI) software known as Survey Solutions, which is now the
UBOS software of choice for surveys collecting data based on CAPI, beyond the UNPS.
The FAO WCA has produced manuals providing best practices on agricultural statistics. Such
manuals are available on several topics including:
• Linking population and housing censuses with agricultural censuses;
• Employment data collection in agricultural censuses; and
• Preparing internationally comparable agricultural statistics.
Some country examples on implementation of best practices are provided in Chapter 7.
Capacity Assessments in Uganda
The first step in the operationalization of these global plans for agricultural statistics is to
determine the status of the NASS in target countries by undertaking capacity assessments.
Capacity assessments provide a snapshot of the NASS across legal, institutional, financial,
methodological, personnel, operational, and technological frameworks. They also provide the
basis for recommendations for improvement.
Uganda has been very involved in capacity assessments for building its NASS. In 2014, the
African Development Bank (AfDB) conducted a capacity assessment of the NASSs of 52
African countries, including Uganda. The assessment reported on the overall status of each
participating county’s NASS, scoring the countries by their institutional infrastructure, resources,
statistical methods and practices, and the availability of statistical information (AfDB 2014).
MAAIF conducted a stakeholder analysis to determine their capacity for building an NASS
(MAAIF 2014). In 2015, the World Bank commissioned a capacity assessment of the Uganda
NASS within UBOS and MAAIF. The current report builds on these assessments to identify the
pertinent areas for improving the production quality and dissemination of agricultural statistics at
national and subnational levels. Subnational statistics are required to support public policy,
manage food security and disaster-risks, and for day-to-day planning, monitoring and decision
making.
11
In Uganda, agriculture is a key driver of the national economy, and the Government of Uganda is
emphasizing the necessity of a functional and comprehensive NASS, including in the 2011
National Agriculture Policy. To assist in that process, UBOS and MAAIF requested that the
World Bank conduct a capacity assessment of the Uganda NASS. The World Bank solicited
applications for the assignment, and RTI International was awarded the contract, with activities
conducted in May–December 2017.
Purpose of This Report
This report presents the results of the capacity assessment with a focus on the ability of the
Uganda NASS to collect and produce agricultural statistics to inform policy and decision-making
for agricultural transformation.
This capacity assessment differs from those previously done in three ways. First, this capacity
assessment includes a comprehensive stakeholder analysis that includes input from data users,
data providers, and donor agencies, in addition to the agencies involved in producing statistics.
MAAIF utilized a stakeholder analysis conducted in 2011 to build the Agricultural Sector
Strategic Plan (ASSP), but it did not conduct a new stakeholder analysis with Ugandan
stakeholders. Second, this report includes a capacity analysis of selected districts’ capacity for
collecting agricultural statistics. None of the previously conducted assessments included a
district-level analysis that can capture the needs of decision-makers at the grassroots. Third, the
study evaluates the role of the two main agencies responsible for producing agricultural statistics
and identifies areas for improving their capacities for producing credible statistics. The AfDB
capacity assessment on the other hand was an overall capacity assessment of the Uganda NASS.
The 2015 assessment only assessed the capacity within MAAIF.5
The remainder of the report is organized as follows. Chapter 2 gives an overview of the
agricultural statistics systems in Uganda and the institutions responsible for collecting and
disseminating statistics on agriculture. Chapter 3 further describes the current sources of Uganda
agricultural statistics and highlights the status of the core statistics collected within the Uganda
NASS. Chapter 4 provides the methodologies employed in conducting the capacity assessments
at national and local government levels, and the limitations of each approach. Chapter 5
discusses the findings from capacity assessments, while Chapter 6 provides recommendations for
improving statistical capacity for agriculture Uganda. Chapter 7 concludes with examples of
global best practices for agricultural data.
5 The term ‘data’ means values provided by the selected farmers or observations that are used to calculate statistics.
The term ‘statistics’ means estimates calculated from the data with an associated measure of uncertainty calculated
from the data.
12
Chapter 2: The Agricultural Statistics System in Uganda
Uganda’s current NASS was established through the Ugandan Bureau of Statistics Act of 1998
providing the mandate for multiple institutions to collect agricultural data. The NASS is a
decentralized system with multiple agencies charged with collecting and disseminating statistics
on agriculture, as shown in Figure 2.1. UBOS and the Division of Statistics in MAAIF are the
two main groups that capture agricultural statistics. Other agencies within the national
government and local governments also collect agricultural data for their own uses. Finally,
NGOs collect agricultural statistics for M&E of various programs.
Figure 2.1: Current Structure of the Uganda NASS
UBOS DAES
MAAIF Statistics Division
NGO Funders
Other Governmental Agencies
Official agricultural statistics
Non-officialagricultural statistics
Statistics for agricultural projects
UBOS’s Role in the NASS
The primary agency for statistics dissemination is UBOS, as stated in the Uganda Bureau of
Statistics Act of 1998 (Government of Uganda 1998). This act gives power to UBOS as the
prime agency in the Government of Uganda for the national statistics system. This act also gives
UBOS the authority to determine what statistics are collected and how. Furthermore, UBOS can
work with and assign duties to other agencies to collect and disseminate statistics.
The primary unit within UBOS for collecting agricultural statistics is the DAES. It was founded
in 2011, and its structure is shown in
13
Figure 2.2: .2. The mandate of the DAES is to be the official source of agricultural statistics for
Uganda. Currently, it produces statistics on crops, livestock, and the environment from the
surveys and censuses. The DAES plans to expand its responsibilities to include aquaculture and
fisheries statistics and will add a senior statistician of Fisheries to the organizational structure.
The goal of the DAES is to collect data and disseminate the official set of core agricultural
statistics. As a department within UBOS, and under the UBOS Act of 1998, the DAES also has
the following responsibilities within the agricultural statistics field according to the Government
of Uganda (1998):
• “Provide high quality central statistics information services.
• Promote standardization in the collection, analysis and publication of statistics to ensure
uniformity in quality, adequacy of coverage and reliability of statistics information.
• Provide guidance, training and other assistance as may be required to other users and
providers of statistics.
• Promote cooperation, coordination and rationalization among users and providers of
statistics at national and local levels to avoid duplication of effort and ensure optimal
utilization of scarce resources.
• Promote and being the focal point of cooperation with statistics users and providers at
regional and international levels.”
The DAES primarily collects data using censuses and surveys. It either designs the data
collection tools or works with outside groups seeking statistics on various projects to create the
surveys. It trains the enumerators on statistically sound methods of collecting data. The DAES
tabulates and disseminates the final statistics to the government, the public, or the organizations
partnering with the DAES to collect statistics on their projects. The DAES can also use
secondary statistics from other data producers to calculate and disseminate statistics.
14
The DAES possesses computers and statistical software to process and produce estimates
(Nalunga 2015). However, it does not possess GIS equipment.
Figure 2.2: DAES Organizational Structure
Director
Principal Statistician
Senior Statistician -
Crops
Senior Statistician -
Livestock
Senior Statistician - Environment
Statistician Statistician Statistician
MAAIF’s Role in the NASS
MAAIF and its various agencies collect vast amounts of agricultural data in the form of
administrative records. These data are secondary products of the normal operations or part of
M&E of the different policies enforced by MAAIF. Administrative data are primarily collected
by agricultural extension agents as a part of their activities.
15
In 2014, MAAIF created the Division of Statistics within the ministry to build the capacity of
MAAIF to collect and disseminate agricultural statistics as a part of its constitutional mandate
“to promote and support sustainable and market oriented agricultural production, food security
and household incomes.” The Division of Statistics is part of the Agriculture Planning
Department.
This goal of this division is to support the NASS for Uganda, collecting and providing
agricultural statistics at the national and district levels using censuses, surveys, and
administrative data. The current organizational structure is shown in Figure 2.3: . As of the
writing of this report, the Division of Statistics has not yet produced statistical reports because it
is new and still being formalized. However, it is undertaking a pilot study with USAID to
examine the use of sentinel farms for the routine collection of agricultural data.
Administrative data refers to non-statistical sources of information obtained through, for
example, government programs or agricultural extension, and can benefit the final statistical
product in ways ranging from reduced costs to improved small area estimates.6 Examples include
data collected through soil information, farm assistance programs (e.g. subsidies and insurance),
land registration and cadastral records, grain associations, and monitoring programs (e.g.
livestock tracing systems). Administrative data has a variety of uses such as improving statistical
sampling frame construction and sample design; filling data gaps from surveys and censuses;
forecasting; planning; and provision of small area estimates and administrative uses, thereby
leading to improved policy and decision-making.
Despite its importance, much of the administrative data is collected and compiled without
employing standard statistical procedures or researchers trained in statistical methods. Research
has shown that a large proportion of administrative data consists of guess-estimates and is
believed to be of questionable quality. There is also an issue of untimely and incomplete flow of
data from the lower to the higher reporting levels. This may lead to delays in the ability of
governments to make policy decisions or a general lack of understanding and hence proper
utilization of own country’s data (UBOS, 2007).7
Uganda face enormous challenges in the compilation of agricultural statistics from
administrative records. First, farmers do not keep records on area planted, animals kept and
production levels. Second, the quality and timeliness of the data is generally poor. Third, Local-
level financial and human resources to support administrative data generation are limited. For
instance, the number of local governments compiling administrative data has been on a decline,
6 Global Strategy to improve Agricultural and Rural Statistics (GSARS). 2017. Improving the methodology for
using administrative data in an agricultural statistics system. Final Report. Technical Report n.24. Global Strategy
Technical Report: Rome 7 Uganda Bureau of Statistics (UBOS). 2007. The Development of the Agricultural Sector Strategic Plan for
Statistics: A Data Collection Plan for Agricultural Statistics in Uganda. Final Report to the Uganda Bureau of
Statistics by the National Consultant: February 2007. UBOS Publication: Kampala.
16
although the MAAIF has been engaging in efforts to develop the capacity of the local
government staff involved in generating agricultural statistics8
One of the key challenges facing the National Statistical System is the generation and utilization
of administrative data. A large volume of administrative data is produced; however, it is of
inadequate quality due to the following reasons (GSARS, 2017):
• Poor data flow, due to unclear reporting mechanisms;
• Submission of incomplete returns or reports;
• Failure of some units to submit returns;
• Data may be collected but not used for planning purposes;
• Poor documentation of the data production processes;
• The reporting mechanisms of different sectors or institutions vary considerably, which
delays the data collection process;
• The skills of the staff involved in data management are limited; and
• High turnover of the professional staff
MAAIF has been strengthening its capacity to produce, store, and analyze statistics and
administrative data through the National Food and Agricultural Statistics System (NFASS)
Project within the Agriculture Planning Department/Division of Statistics. The NFASS Project
has three integrated components:
• Development of a data center for all agricultural statistics;
• An institutional data module; and
• Routine Agricultural Administrative Data Reporting System (RAADRS)
Figure 2.3: Division of Statistics Organizational Structure
8 Global Strategy to improve Agricultural and Rural Statistics (GSARS). 2017. Improving the methodology for
using administrative data in an agricultural statistics system. Final Report. Technical Report n.24. Global Strategy
Technical Report: Rome
17
Source: Nalunga 2015.
Note: ICT = information and communication technology.
Other Agency Contributions to the NASS
In addition to UBOS and MAAIF, other agencies within the Government of Uganda collect data
and disseminate agricultural statistics. There are seven semiautonomous agencies within MAAIF
that produce agricultural statistics according to their mandates, needs, and routine activities.
These statistics are used by UBOS, MAAIF, and organizations such as the World Bank, UN,
International Monetary Fund, Bank of Uganda (BOU), and Ministry of Finance Planning and
Economic Development. The seven agencies are as follows:
• National Agricultural Research Organization (NARO): This is the agricultural research
organization within MAAIF. It collects data and produces statistics related to the
experiments conducted within the organization and its partners.
• Uganda Coffee Development Authority (UCDA): This is the regulatory agency for coffee
production within Uganda. UCDA produces statistics on coffee production and coffee
farm numbers.
• Cotton Development Organization (CDO): It is the regulatory agency for cotton
production within Uganda. The CDO produces statistics on all aspects of the cotton
industry within Uganda.
Division
Agricultural
Statistics
Assistant Commissioner
Principal
Statistician
Fisheries Livestock Crops
3 Statisticians
3 Statisticians
3 Statisticians ICT
4 Staff
Senior Statistician
Senior Statistician
Senior Statistician
18
• Dairy Development Authority (DDA): It is the regulatory agency for dairy production
within Uganda. The DDA has produced statistics on milk production and milk prices.
• National Animal Genetic Resource Centre (NAGRIC) Data Bank: The NAGRIC
oversees the national animal breeding program in Uganda. The NAGRIC Data Bank
contains genetic data on both commercial and indigenous livestock breeds.
• National Agricultural Advisory Services (NAADS): It exists to improve Ugandan
agriculture as an advisory service to farmers and agribusinesses. The NAADS collects
data from farmers though participatory M&E activities in programs that improve farm
household welfare through modernized farm practices. The NAADS also produces
statistics on the quantity sold and value obtained of various agricultural products.
• Uganda Trypanosomiasis Control Council (UTCC): It seeks to eradicate trypanosomiasis
and tsetse in Uganda. UTCC produces statistics on tsetse and trypanosomiasis control and
eradication projects.
In addition, the following directorates within UBOS produce statistics that pertain to agriculture:
The NFA; Directorate of Statistical Capacity Services; Directorate of District Statistics and
Capacity Development; Directorate of Business and Industry Statistics; Directorate of Population
and Household Statistics; and Directorate of Socioeconomic Statistics.
To support coordination, technical issues, and dissemination of agricultural statistics, there are
three additional committees and activities:
• The National Agricultural Statistics Technical Committee (NASTC) is formed by the
primary stakeholders in agricultural statistics and provides a forum for discussion on
concepts, methods, and technical issues. The committee is chaired by MAAIF, cochaired
by the School of Statistics and Planning Department at Makerere University while UBOS
serves as the secretariat. The NASTC meets quarterly.
• The PNSD is developed through sector-specific plans as its building blocks. The
framework serves as the coordinating mechanism for agencies that produce agricultural
statistics.
• The Country STAT Technical Working Group is made up of major producers and users
of agricultural statistics and reviews and discusses statistics before dissemination through
the UBOS statistics abstract each year.
Local Government Contributions to the NASS
Finally, local and district governments collect their own sets of agricultural statistics. The
District Planning Unit within each district collects various agricultural data for purposes of
19
planning and monitoring.9 Data are primarily collected through district officers as a part of their
normal activities. Statistics are calculated from these administrative data for the purposes of
monitoring and policy development, enactment, and enforcement.
Current Sources of Uganda Agricultural Statistics
Censuses
UBOS conducts the National Population Household Census (NPHC) roughly every 10 years.
The NPHC captures demographic information on the population of Uganda. The goal of the
NPHC is “… to ensure availability of bench-mark demographic and socio-economic data for use
in planning, policy formulation and program evaluation” (UBOS 2014b). The most current
census was performed in 2014. This census contains an agriculture module that asks the
respondent the type of animal or crop farming the household engaged in. The module also asks if
land was owned by the head of the household and whether the household used irrigation.
UBOS has conducted the COA three times: once in 1967, once in 1990/91, and, most recently, in
2007/08 (UBOS 2016). The COA provides a comprehensive snapshot of Ugandan agriculture
with statistics on crops, livestock, economics, socio-demographics, agro-forestry, and irrigation.
To conduct the COA, a 1 percent sample of farms is drawn from the respondents to the
agriculture module in the most recent NPHC, and enumerators are sent back to those households
to collect the additional agricultural data. UBOS plans to conduct a new Census of Agriculture
and Aquaculture in 2019/20 in cooperation with the WCA.
In 2008, MAAIF and UBOS conducted a census of livestock to provide data on livestock
agriculture for the National Livestock Productivity Improvement Project (UBOS 2009). This
census provided estimates on livestock production in household farms and institutional farms
within a sample of enumeration areas in 80 districts. Estimates were produced on livestock and
poultry, heads of households, economic inputs and costs, labor use and costs, and some livestock
prices.
Crop and Livestock Statistics
The Uganda National Panel Survey (UNPS) includes a strong focus on agriculture since 2009 towards
the design, implementation, analysis and dissemination. It is funded by World Bank’s LSMS-ISA and
conducted by the Government of Uganda.
9 The data includes production of primary food crops (crop production, area harvested and yields), use of land, farm
machinery, fertilizers and pesticides, fisheries, food availability for consumption, population, and labor force at the
district level
20
Forestry Statistics
The GIS and Mapping Unit in the NFA maps the availability of forest wood within the central
forest reserves. This information is included in UBOS’s statistical releases.
Fisheries and Aquaculture Statistics
There are no current statistics on fisheries and aquaculture. MAAIF is currently working on
reestablishing the data collection tools that use the Beachhead Management Unit as the sampled
observation (MAAIF 2014).
Agricultural Markets and Price Information Systems
The BOU publishes monthly, quarterly, and annual price data on agricultural exports. Price data
are available for coffee, cotton, tea, fish, maize, simsim (sesame), tobacco, beans, sugar, and
other agricultural products. The BOU obtains price data from UBOS, MAAIF, and other
agencies that track specific commodities. The price data are disseminated on the BOU’s website.
Water and Environment Statistics
The NPHC counts the number of households that used irrigation in agricultural production.
The National Service Delivery Survey is a multistage survey that reviews the trends in service
delivery. The most recent iteration was conducted in 2015 and provided statistics on agricultural
inputs and costs, extension activities, and the environment (UBOS 2015).
Rural Development Statistics
The 2012/13 UNHS published statistics on rural poverty (UBOS 2014a). As a part of monitoring
NAADS program implementation among farmers, the Agriculture Technology and Agribusiness
Advisory Services Project conducted a survey of 15,010 farmers over 111 districts that
monitored the use of improved crop and livestock technologies during 2010 and 2011 (NAADS
2013).
Food Security and Nutrition
The 2007/08 COA published statistics on food security. The 2012/13 UNHS published statistics
on food poverty (UBOS 2014a). The 2006/07 wave was the reference wave and looked at rates
of growth. This survey includes crop and livestock modules (UBOS 2014b).
Table 2.1 shows the list of statistics currently collected in the Uganda NASS and their source and
year of availability. The source and year of release were reported by either MAAIF or UBOS in
the SAQ. The releases were then confirmed through Internet searches.
Table 2.1: Status of the Minimum Set of Core Statistics Collected within the Uganda NASS
Statistic Agency with Most Recent Data
Year of
Most Recent
Release
Crops
Crop production: quantity UBOS 2016
Crop production: value None -
Crop yield per area UBOS 2015
Area planted UBOS 2013/14
Area harvested UBOS 2013/14
21
Statistic Agency with Most Recent Data
Year of
Most Recent
Release
Livestock
Livestock production: quantity UBOS 2013/14
Livestock production: value UBOS 2013/14
Fisheries and aquaculture
Fishery and aquaculture production: quantity MAAIF 2017
Forestry and wood products
Forest production of wood16: quantity None -
Forest production of wood: value UBOS 2016
Forest production of non wood17: quantity None -
Forest production of non-wood: value UBOS 2016
External trade
Export: quantity None -
Export: value BOU 2017
Import: quantity None -
Import: value BOU 2017
Stock of capital and resources
Livestock inventories MAAIF 2016
Agricultural machinery MAAIF 2016
Stocks of main crops: quantity None 0
Land and use None 0
Water related
Irrigated areas UBOS 2013/14
Types of irrigation None 0
Irrigated crops None 0
• Quantity of water used None 0
• Water quality None 0
Inputs
Fertilizer quantity UBOS 2013/14
Fertilizer value UBOS 2013/14
Pesticide quantity UBOS 2013/14
Pesticide value UBOS 2013/14
Seeds quantity UBOS 2013/14
Seeds value UBOS 2013/14
Animal feed quantity MAAIF 2016
Animal feed value MAAIF 2016
Forage quantity None 0
Forage value None 0
Animal vaccines and drugs quantity MAAIF 2016
Animal vaccines and drugs value MAAIF 2016
Aquatic seed quantity None 0
Aquatic seed value None 0
Agro-processing
Main crops None 0
Post-harvest losses None 0
Main livestock None 0
Fish: quantity MAAIF 2016
Fish: value MAAIF 2016
Prices
Producer prices MAAIF 2017
Wholesale prices None 0
Consumer prices MAAIF 2017
Agricultural input prices MAAIF 2017
Agricultural export prices MAAIF 2017
Agricultural import prices MAAIF 2017
Investment subsidies or taxes
Public investment in agriculture None 0
22
Statistic Agency with Most Recent Data
Year of
Most Recent
Release
Agricultural subsidies None 0
Fishery access fees None 0
Public expenditure for fishery management None 0
Fishery subsidies None 0
Water pricing None 0
Rural infrastructure and services
Area equipped for irrigation None 0
Crop markets None 0
Livestock markets None 0
Rural roads (km) Uganda National Roads Authority 2016
Railways (km) None 0
Communication Uganda Communication Commission 2017
Banking and insurance BOU 2017
Social
Population dependent on agriculture UBOS 2015
Agricultural workforce (by gender) UBOS 2008/9
Fishery workforce (by gender) MAAIF 2016
Aquaculture workforce (by gender) MAAIF 2016
Household income UBOS 2015
Environmental
Soil degradation None 0
Water pollution due to agriculture None 0
Emissions due to agriculture None 0
Water pollution due to aquaculture None 0
Emissions due to aquaculture None 0
Geographic location
Geo-coordination of the statistical unit (parcel, province, region,
country) UBOS 2013/14
23
Chapter 3: Methodology
This study employed mixed methodologies to conduct the capacity assessment. This chapter
describes those methodologies and the goals and limitations of each approach.
RTI conducted the study with the assistance of local subcontractor Development Research and
Social Policy Analysis Center (DRASPAC) between May 2017 and October 2017. Data
collection began in June 27, 2017, with kickoff meetings and key interview meetings with
UBOS, MAAIF, the World Bank, and other stakeholders. Data collection occurred from June 27,
2017, to August 19, 2017. Tabulation and data analysis were performed between August 20,
2017, and September 21, 2017. The draft report was written between September 21, 2017, and
September 29, 2017.
Identification of Key Stakeholders
In a series of meetings and discussions between the World Bank and RTI, a list of all key
stakeholders in the Ugandan agricultural statistics system was identified. Key stakeholders
included the following:
• The directors of the agencies and ministries responsible for producing agricultural
statistics
• Key agricultural data users from the agribusiness, government, and academic fields
• Key members of farmers’ groups and farmers in strategic agricultural fields that provide
the data
• Key members of groups that fund agricultural statistics production
These members formed the group that RTI and DRASPAC met and conducted interviews with
during the study. The stakeholders included public and private organizations and development
partners (Table 3.1).
Initial Desk Review and Stakeholder Consultation
RTI conducted a desk review of relevant documents, including national agricultural and
statistical policies to collect secondary data to supplement stakeholder interviews. The full list of
secondary documents consulted can be found in Appendix 1. The goal of this desk review was to
assist in examining the current structure of the Uganda NASS and preparing for the stakeholder
discussions.
Panel Discussions and Key Informant Interviews
An in-person interview stage followed the desk review. Panel discussions were conducted at the
national level, bringing together multiple stakeholders to identify comprehensive ideas. Panel
discussion attendees included relevant officials at UBOS and the Ministry of Agriculture
Planning Department and representatives from the private sector, civil society, farmer and
agribusiness associations, NGOs, and donors.
24
These panel discussions were carried out by DRASPAC based on panel questions and facilitation
guidance from RTI. In addition to panel discussions, RTI conducted a series of key informant
interviews (KIIs) to ascertain the views of individuals involved in each stage of data collection,
analysis, dissemination, and use. Table 3.1 shows the stakeholders that were interviewed, the
dates they were interviewed, and their role within the Uganda NASS.
Table 3.1: Key Stakeholders Interviewed
Date Organization Role in NASS
June 27, 2017 UBOS Producer
MAAIF Producer
World Bank User/funder
June 28, 2017 Economic Policy Research Centre (EPRC) User
Makerere University User
June 29, 2017 National Agricultural Research Organization (NARO) Producer/user
UBOS Producer
June 30, 2017 Uganda Coffee Development Authority (UCDA) Producer
July 5, 2017 National Forestry Authority (NFA) Producer
July 14, 2017 Uganda National Farmers’ Federation (UNFFE) Provider
July 17, 2017 Agricultural Business Initiative (aBi) Trust User
August 1, 2017 Japan International Cooperation Agency (JICA) Funder
August 2, 2017 Danish International Development Agency (DANIDA) Funder
August 4, 2017 U.S. Agency for International Development (USAID) Funder
August 8, 2017 National Environment Management Authority (NEMA) User
August 9, 2017 FAO Funder/user
Standard Assessment Questionnaire
The Standard Assessment Questionnaire (SAQ) was designed by the AfDB to assess national
capacity for collecting and producing agricultural statistics. RTI employed the SAQ as one of the
tools for this assessment to examine the current capacity of the Uganda NASS for agricultural
statistics. The SAQ was administered to officials within the DAES in UBOS and the Division of
Statistics in MAAIF to assess the capacity of these organizations to collect and generate
agricultural statistics.
25
The SAQ was administered in parts to the appropriate personnel who could provide the most
accurate responses.10 The SAQ was sent to UBOS and MAAIF to assess the capacity within
these agencies.11 The SAQ was not administered to the seven semiautonomous agencies that also
produce agricultural statistics because of time and budget constraints. Each part was mailed to
the DAES director and statisticians and to the assistant commissioner and statisticians in the
Division of Statistics in MAAIF. Responses were returned by mail after follow-up contacts.
Missing data were left as missing because of budgetary constraints. These missing values were
not imputed using other data sources. The data were then converted into agricultural statistics
capacity indicators (ASCIs) using the methodology created by the AfDB and grouped into the
same dimensions of institutional infrastructure, resources, statistical methods and practices, and
availability of statistical information and their elements (AfDB 2014). Each ASCI was
categorized based on the level of capacity as shown in 3.2.
Table 3.2: ASCI Classification
ASCI Capacity Classification
0 ≤ ASCI < 20 Very weak
20 ≤ ASCI < 40 Weak
40 ≤ ASCI < 60 Moderate
60 ≤ ASCI < 80 Strong
ASCI ≥ 80 Very strong
Participatory Local Organizational Assessment Interview
The Participatory Local Organizational Assessment (PLOCA) tool, developed by RTI and
adapted for this survey, is a comprehensive capacity assessment tool that seeks to capture the
capacity of local organizations and institutions in management and practices, policies, personnel,
and materials. This provides a holistic view of the capacity of organizations, including in this
instance, the capacity to collect agricultural statistics. The PLOCA examines 10 core functions
(Table 3.3) considered critical to organizational performance (RTI 2014).
10 The SAQ tool can be found at
https://www.afdb.org/fileadmin/uploads/afdb/Documents/Publications/AfricaCountryAssessment_ASCI_Report_Fi
nal_Web_11_2014.pdf.
11 The findings of the self-assessment were validated in a workshop in March 2018 with key development partners
and other stakeholders in attendance. The identified constraints were further discussed at the validation workshop
together with a synthesis of recommendations to address the constraints. These are discussed further in Chapter 5.
26
Table 3.3: Core Functions Examined in the PLOCA
Core Organizational Functions
1. Mission, vision, values
2. Governance
Management and Implementation
3. Strategy
4. Leadership and internal collaboration
5. Learning and innovation
6. Project implementation and service delivery
7. Human resources (HR)
8. Financial and administrative management
9. Collaboration and networking
10. Fundraising and sustainability
To examine the capacity of the government to obtain agricultural statistics at the district level,
district officials were given the PLOCA questionnaire during the district meetings conducted by
DRASPAC on July 12, 2017, in Masaka District and July 18, 2017, in Mbale District. The goal
of the questionnaire was to determine the ability and capacity of officials in the districts to
collect agricultural statistics.12 The districts were selected across agroecological zones to capture
diversity in Uganda’s agricultural system. The selection was based on inputs from both the
World Bank and UoG teams. While the study of these sixteen districts can only provide case study
insights, the common issues which have emerged imply that the analysis and recommendations are useful
beyond the districts visited. The districts surveyed and the dates they were surveyed are listed in
Table 3.4.
Table 3.4: Districts Surveyed
Date Region District
July 12, 2017 Northern Adjumani, Apac, Gulu, Kaabong
Eastern Kumi, Mayuge, Mbale, Tororo
July 18, 2017 Western Bushenyi, Hoima, Kisoro, Mbarara
Central Kayunga, Kiboga, Luwero, Masaka
Limitations of the Study
The current study has the following limitations because of time and budgetary constraints that
prevented analyses beyond those presented in this report:
12 The meetings were titled ‘Consultation on the Development of Sentinel Farmers Sampling Methodology and Data
Collection Tools Under the NFASS’.
27
• The capacity was analyzed at the national and district levels. No attempts were made to
analyze capacity at geographic levels below the district as time and resources did not
permit this level of detail.
• The completed SAQs from both UBOS and MAAIF contained uncompleted sections.
o UBOS did not respond to questions on informational and computational technology
(ITC), financial, personnel, and district capacity. However, the AfDB capacity
analysis from 2014 looked at the capacity of UBOS in these areas and has been used
for this analysis.
o MAAIF did not respond to questions on the overall structure of its Division of
Statistics and ITC, financial, personnel, and district capacity. However, it outlined the
needs in these areas in their 2012 ASSP.
While this study cannot provide a complete capacity analysis of UBOS and MAAIF where data
are missing, the capacity in these areas has previously been assessed. Thus, by pairing existing
and new information, the study team was still able to obtain a comprehensive view of the
Uganda NASS and structure recommendations accordingly.
28
Chapter 4: Findings from the Assessment
Themes from Stakeholders Interviews
Several common themes arose from the KIIs and panel interviews. These themes are outlined in
this chapter, along with statements from the interviews.
National-level Capacity for Agricultural Statistics Exists
Despite questions about the reliability of the data being reported, stakeholders did identify some
capacity for agricultural statistics embedded within the current NASS.
• UBOS: It is an established organization with the mandate for and the structure to get data.
• MAAIF: It has hired 20 statisticians and obtained GIS equipment.
• aBi Trust: The macro-level statistics such as the Demographic and Health Survey (DHS)
and census are good and available with UBOS.
• USAID: It also generates some good statistics for the agricultural sector.
District-level Statistics Are Greatly Desired, But the Capacity for Them Is Lacking
Although there is capacity at the national level, as stated above, at the district level, Ugandan
institutions lack the human capital and economic resources needed for thorough data collection.
• NFA: More detailed statistics need to be gathered from the grassroots.
• UBOS: District officers cannot collect survey data.
• MAAIF: It feels that statisticians are needed at the district level for this to happen.
• EPRC: One of the biggest problems with data availability is the lack of access to data
from the ground.
Different Agencies Create Different Systems
As described in Chapters 3 and 4, the NASS currently comprises many organizations and many
independent surveys and censuses. This reflects the persistent lack of clarity in institutional mandates
concerning collection and dissemination of agricultural statistics. It is known that UBOS has the overall
mandate of production and dissemination of official statistics, production of statistics is a combined effort
of various stakeholders including Ministries, Departments and Agencies (MDAs). Specifically, the
Directorate of Agriculture and Environmental Statistics under UBOS holds the primary responsibility for
production and management of agricultural statistics. However, the actual collection, analysis and
dissemination involves more stakeholders than those directly under that directorate. For instance, the
Division of Agricultural Statistics under MAAIF is also directly mandated by the constitution to take lead
and establish a system and institutional framework for agricultural data collection, analysis, storage and
dissemination to stakeholders, including UBOS.
29
Thus, multiple agencies reported that their mandate included the collection and dissemination of
data as outline below.
• MAAIF: It stated they are creating the NASS.
• UBOS: It stated they are the clearinghouse of official agricultural statistics.
• JICA: It supported a pilot to generate statistics on rice from 44 districts at the regional
level in collaboration with the NAADS and NARO.
• UCDA: It has a very good export database but not a good database on farmers.
• DANIDA: Each project/organization tries to make its own baseline survey.
There Is Little Faith in the Reliability of the Current Agricultural Statistics System
Although numerous organizations are involved in the collection and dissemination of agricultural
statistics, the data are not necessarily dependable. Actors throughout the sector highlighted
weaknesses in the current data being collected and reported on.
• EPRC: Data quality depends on the size of the project; larger projects tend to have fewer
issues.
• aBi Trust: UBOS data are generated after long intervals. MAAIF data are quite unreliable
and are scanty.
• USAID: The statistics generated by MAAIF are based on estimates and not real hard
data.
• NEMA: The geomapping done by the NFA cannot be trusted because it is using out-of-
date methodologies and equipment.
There Is Little Attention Paid to Agricultural Statistics
Although it faces challenges and capacity/resource constraints, the current NASS in Uganda does
produce a wide array of statistics. However, stakeholders found that those statistics are not being
effectively used to inform the government and private sector planning.
• JICA: Utilization of agricultural statistics in planning and decision making is low at all
levels (district and MAAIF levels).
• aBi Trust: The extent and rigor in use of statistics in the country is generally low.
• Makerere University: Graduates from the Department of Statistics at Makerere University
do not see a future in agricultural statistics.
Methodologies for Collecting Commodity-specific Statistics Are Not Adequate
Commodity-specific statistics, where they are being collected, are not as comprehensive as they
should be and are generally specific to price data.
30
• JICA: The UBOS statistics are not comprehensive on individual commodities such as
rice.
• Uganda Coffee Growers Association (UCGA): Coffee production statistics are
problematic. Production should be calculated as the number of trees times yield.
• UBOS: Data on animal permits are only captured when animals are moved.
• UNFFE: UBOS uses approaches and methodologies that farmers do not understand and
so it never gets the correct information from the farmers.
New Technologies Can Be Utilized for Digital Data Collection
Stakeholder agreed that as UBOS, MAAIF, and other organizations aim to improve agricultural
statistics, the use of new technologies should be incorporated.
• NFA: Using digital devices, for example, mobile phones in the generation and
transmission of data. These can easily be integrated into geospatial mapping to get more
detailed and accurate data.
• EPRC: It would be ideal if we have the farmers report their data on mobile phones.
• Makerere University: UBOS has improved its data collection capabilities. It is using
CAPI for data collection and it has a pilot study going using Open Data Kit.
Capacity Assessment of DAES and MAAIF
The SAQ results collected from the DAES and MAAIF are shown in Table 4.1: 4.1. Overall,
UBOS and MAIFF reported an average capacity for agricultural statistics, but each agency had
strengths in different dimensions.
The DAES indicated that it had higher capacity within the institutional infrastructure dimension
with an average ASCI of 69.4 versus 18.2 for MAAIF. Consistent with the national mandate of
UBOS, its parent organization, the DAES felt that it had strong coordination within the NSS and
a strategic vision and planning for agricultural statistics. However, MAAIF only reported weak
capacity in the integration of agriculture within the national statistics system and no capacity in
any other element.
No agency reported any capacity for the resources dimension. This same pattern of nonresponse
was discovered in the AfDB capacity assessment (AfDB 2014). However, it used other sources
to fill in the missing data to calculate its ASCIs. As for the resources dimension, no agency
reported any capacity for statistical software, data collection technology, or information
technology infrastructure. This was also reported in the AfDB report.
MAAIF tended to have stronger capacity in the statistical methods and practices dimension,
excluding technology. It felt it had strong capacity in the adoption of international standards and
producing agricultural markets and price information. The DAES only reported moderate
capacity in adopting international standards and weak to very weak capacity in the remaining
nontechnological elements.
31
Each agency reported moderate capacity in the availability of statistical information dimension.
MAAIF reported moderate capacity in the availability of the minimum set of core statistics. Both
agencies felt they had moderate capacity in overall data quality perception and data accessibility.
Despite not reporting information on personnel and technological capacity, UBOS and MAAIF
provided some information during KIIs. As shown in
Figure 2.2: .2, the DAES employs a director, an assistant director, three senior statisticians, and
three supporting statisticians. In addition, the DAES can rely on other directorates within UBOS
to assist with producing agricultural statistics, such as the Directorate of District Statistics.
Additionally, the DAES has agreements with outside agencies such as the NFA to produce
statistics. Finally, the DAES utilizes the administrative, information technology, and HR
directorates within UBOS to handle tasks outside of its mandate.
The Division of Statistics within MAAIF is a part of the Agricultural Planning Department. The
division can rely on the Administration Planning Department for administrative and personnel
tasks. However, ICT is handled separately within each agency. In the kickoff meeting, the
Division of Statistics reported having a GIS unit that contains a map printer and several
computers obtained using USAID funding. It has also set up a data processing unit that contains
four servers for the transmission of data. In terms of personnel, more than 20 statisticians work
within the division.
These semiautonomous agencies may have varying levels of personnel and ITC capacity for
producing agricultural statistics. Although all agencies were not interviewed for this assessment,
the interview with the UCGA provided some insight into its personnel and ITC capacities.
UCGA has two statisticians who produce agricultural statistics. They do not see this as enough
statisticians, and they would ideally like to have more. For ITC capacity, they reported that they
would need software for statistical analysis.
32
Table 4.1: Standard Assessment ASCIs for DAES and MAAIF
Dimensions Elements DAES MAAIF
Institutional
infrastructure
Legal framework 60.0 —
Coordination in the NSS 100.0 —
Strategic vision and planning for agricultural statistics 100.0 —
Integration of agriculture in the NSS 45.5 18.2
Relevance of data 41.7 —
Average ASCI 69.4 18.2
Resources Financial resources — —
HR: staffing — —
HR: training — —
Physical infrastructure — —
Average ASCI 0.0 0.0
Statistical methods and
practices
Statistical software capability — —
Data collection technology — —
Information technology infrastructure — —
Adoption of international standards 46.9 84.4
General statistical activities 28.6 14.3
Agricultural markets and price information 10.0 100.0
Agricultural surveys 21.1 15.8
Analysis and use of data 11.1 11.1
Quality consciousness 25.0 25.0
Average ASCI 23.8 41.8
Availability of
statistical information
Core data availability 14.9 62.2
Timeliness 66.7 66.7
Overall data quality perception 60.0 40.0
Data accessibility — —
Average ASCI 47.2 56.3
Overall ASCI 45.1 43.8
Capacity Assessment at the District Level
The results of the PLOCA administered to the 16 districts are shown in 4.2. The overall score for
each function was the average of all reported scores (excluding missing and ’not applicable’
answers) from the districts. The total score for each district was a weighted average of all
reported scores for each district with an equal weight applied to each section. The weighted
average was used to prevent the responses from any one section from skewing the capacity score.
33
Table 4.2: Districts Scores for Different Measures of Statistical Capacity
Assessment Indicators
Apac Mbale Kumi Tororo Mayugi Adjumani Gulu Kaabong Kiboga Kisoro Kayunga Luwero Hoima Masaka Mbarara Bushenyi Overall
Function I: Mission,
Vision, Values
2.6 4.00 2.4 3.2 3.2 3.2 2.4 3.2 2 2.4 2 2.6 2.8 3.2 2 3.8 2.81
Function II: Governance 3.55 3.27 3.18 3.55 3.64 2.36 2.64 3.73 2.27 3.55 3.09 3.2 3.27 2.73 2.73 3.27 3.13
Function III: Management
and Implementation 3.14 3.04 2.73 2.82 3.29 2.74 2.56 3.13 2.81 2.09 2.96 2.77 2.76 2.08 1.86 3.29 2.75
Function IV: HR 3.00 3.1 3.45 3.06 3.53 3.32 2.65 3.15 2.9 2.9 3.35 3.15 3.2 2.05 3.35 3.4 3.10
Function V: Financial and
Administrative
Management
3.23 2.73 3.65 2.96 3.29 3.7 2.72 3.04 3.42 3.31 3.65 3.2 3.25 2 3.62 2.8 3.17
Function VI:
Collaboration and
Networking
3.3 3.1 2.8 2.5 3.1 2.2 2.7 2.9 2.1 2.9 3.4 2.4 2.7 2.3 2 2.1 2.66
Function VII: Fundraising
and Sustainability 2.12 2.31 2.44 1.72 2.64 1.81 1.83 2.71 1.83 1.78 2.39 2.17 1.94 1.72 2.5 2.27 2.14
District Average 2.99 3.08 2.95 2.83 3.24 2.76 2.50 3.12 2.48 2.70 2.98 2.78 2.86 2.30 2.58 2.99
Average Score 1 2 3 4
Capacity Nascent Basic Moderate High
Legend
34
Function I: Mission, Vision, Values
Two districts, Mbale and Bushenyi or 13 percent of the districts felt they had a very clear
understanding of the mission, vision, and values for collecting agricultural statistics, 11 districts
(69 percent) reported moderate understanding, while the remaining 3 districts – Kiboga,
Kayunga, and Mbarara districts (18 percent) recorded basic understanding for mission, vision
and values for agricultural statistics. The average capacity score for Function 1 across the
districts was moderate.
Function II: Governance
Apart from Apac, Tororo, Mayugi, Kaabong, and Kisoro districts (31 percent) that felt that the
governance structure and capacity was well in place, all the rest (69 percent) felt they had nearly
moderate capacity for collecting agricultural statistics. They tended to report that the governance
structure was democratically elected and had a clear set of bylaws and a constitution defining the
roles and actions of the governing board. However, they felt that the governing board had only
basic capacity in mobilizing resources and representing the district externally in matters of
advocacy and lobbying.
Function III: Management and Implementation
Although the overall capacity in management and implementation was rated nearly moderate,
individual districts reported varying degrees of capacity in this area. 13 districts (81 percent)
reported moderate capacity while Mbarara, Kisoro districts (13 percent) reported basic capacity
and Masaka district reported below basic (nascent) capacity in management and implementation.
Districts felt that they had implementation plans that were linked to strategic plans. Districts also
believed that M&E plans and that leadership succession plans in the districts were weak.
Function IV: HR
In terms of HR capacity Mayugi district had nearly high capacity, 14 districts (88 percent) felt
that they had moderate capacity while Masaka district that felt they had basic capacity in this
area. Districts felt that the job descriptions within each district were clearly defined and that
salaries are clearly structured. However, they felt that staff are not properly motivated to do their
jobs and that HR policies are not regularly reviewed, updated, and distributed to staff.
Function V: Financial and Administrative Management
Kumi, Adjumani, Kayunga, and Mabara districts (25 percent) felt that they had nearly high
capacity while the remaining 11 districts (75 percent) had basic capacity. Majority of the districts
felt that there was clear organization of financial duties, institutional bank accounts are in place,
and clear staff travel per diem policies and that external audits are conducted annually by a
registered firm. However, they felt that there was inadequate insurance in place to protect the
district and safeguard their assets.
Function VI: Collaboration and Networking
Ten districts (62 percent) reported moderate or nearly moderate capacity in collaboration and
networking while Adjumani, Kiboga, Luwero, Masaka, Mbarara and Bushenyi districts (38%)
35
felt that they had only basic capacity. They had basic capacity on external marketing plan or any
external communication tools that are regularly updated.
Function VII: Fundraising and Sustainability
Mayugi, Kaabong, and Mbarara districts (19 percent) reported moderate capacity while Apac,
Mbale, Kumi,Kayunga, Luwero and Bushenyi districts (37 percent) reported basic capacity.
Tororo, Adjumani, Gulu, Kiboga, Kisoro, Hoima and Masaka districts (44 percent) reported less
than basic capacity in this function. Although they felt they had moderate capacity for
establishing policies that meet donor requirements, accepting input from relevant stakeholders,
and institutionalizing programs for beneficiaries to take ownership, the districts did not believe
that their revenue stream was stable or that they had the financial means to increase programs.
In general, the 16 districts reported basic to moderate capacity across all functions. Mayugi,
Mbale and Kaabong districts reported the highest average capacity across all functions while
Masaka and Kiboga districts recorded the lowest.
The lowest score for the functions was recorded for Fund raising and Sustainability, while the
highest was recorded for Financial and Administrative management. Table 4.3 reveals significant
but negative correlations between functions I and V, indicating that a clear understanding of the
mission, vision, and values for collecting agricultural statistics is not backed up with financial
administrative capacity in the districts. Conversely, there is significant positive correlation
between functions II and VI, suggesting that good governance structure supports collaboration
and networking. There is a significant positive correlation between HR and financial
administration management, and between HR and Fundraising and Sustainability. This suggests
that well-articulated HR policies and proper staff motivation can be instrumental in enhancing
financial management and sustainability for improving agricultural systems in the districts.
36
Table 4. 3: Correlation between Core Functions Critical to Organizational Performance at the
District Level Function
I:
Mission,
Vision,
Values
Function II:
Governance
Function III:
Management
and
Implementation
Function
IV: HR
Function V:
Financial and
Administrative
Management
Function VI:
Collaboration
and
Networking
Function VII:
Fundraising
and
Sustainability
Function I:
Mission, Vision,
Values
1.00
Function II:
Governance
0.33 1.00
Function III:
Management and
Implementation
0.45 0.41 1.00
Function IV: HR 0.01 0.29 0.47 1.00
Function V:
Financial and
Administrative
Management
-0.55* -0.08 0.06 0.74* 1.00
Function VI:
Collaboration
and Networking
-0.00 0.61* 0.41 0.15 0.09 1.00
Function VII:
Fundraising and
Sustainability
0.06 0.42 0.39 0.64* 0.30 0.34 1.00
Correlation coefficients with * are significant at 5 percent probability level.
37
Chapter 5: Challenges in the Uganda NASS
Based on the interviews and capacity assessments the following challenges were identified
within the Uganda NASS. These issues were also confirmed during the validation workshop.
Institutional
One of the most serious challenges facing the NASS is the lack of coordination and
communication between the different agencies collecting agricultural data. Because the NASS is
a decentralized system, agencies can and have produced differing values for the same estimate.
MAAIF is taking steps to set up the NFASS to collect administrative data from districts in
coordination with UBOS’ system for collection of official statistics. There are coordination and
reporting systems in place, but the data collection, storage, and analysis systems are still
emerging. The structure proposed in the ASSP and the Plan for National Statistical Development
will allow for the efficient collection of agricultural data. UBOS is also building its capacity to
generate the official set of agricultural statistics. The plans are in place and work toward an
efficient system is ongoing, but the institutional structures are not fully operational yet. It is a
critical challenge in preparation for the 2020 COA that these systems should rapidly be brought
to full operational status. Oversight of overall development of the NASS is needed to ensure that
these institutional challenges are resolved.
Statistical methodologies present a second institutional challenge. Developing capacity to collect
accurate data using standardized methods is important to avoid inefficiencies and inaccuracies in
the system. A lack of transparency in presenting statistical methodologies or the use of outdated
methods can lead to inefficiencies and a general sense of mistrust of the data among data users.
There is a clear challenge to the NASS to set appropriate standards for measurement, making
data available for analysis and consistent operational mandates for each agency collecting data.
Data releases are inconsistent across all agencies, coming in different years and at different
times. Data users have identified this issue as an institutional challenge. Finally, while there is a
clear institutional mandate for agricultural statistics, some of our respondents felt the lack of
available resources or prioritization among many other needs has constrained the improvement
of the NASS. In summary,
• There is coordination and cooperation between UBOS and MAAIF but resource
challenges slow improvements to the NASS;
• There is confusion between the agencies within the NASS as to who produces and
collects each type of agricultural statistic;
• There is no agreement or standardization of statistical methodologies between agencies;
• Data releases are inconsistent; and
• There is a lack of prioritization in funding decisions that limits improvement in the
collection and analysis of agricultural statistics.
38
Methodological
The methodologies for agricultural statistics production are either not transparent or have not
been updated to reflect changes within Ugandan agriculture. Data users have criticized UBOS for
not providing enough details on how statistics are produced and MAAIF for not producing any
methodologies on statistics. Additionally, changes to programs or practices within Ugandan
agriculture have not been implemented within Ugandan agricultural statistics. MAAIF possess
administrative data that can be used to augment UBOS censuses and surveys (for example,
livestock data). Additionally, agencies within the NASS are not effectively collaborating and
using their institutional strengths to fulfill their mandates. For example, the NFA can produce
geospatial maps for UBOS, but the spatial statistics are not being released.
Additionally, several types of statistics are not actively being collected within the NASS. These
statistics are identified in the Global Strategy as a basis for the work of national and international
data users. Therefore, the lack of these statistics prevents analysis and decision making relating
to them. In summary,
• Methodologies for producing agricultural statistics are not adequate;
• Statistical methodologies for the minimum core set are not made available;
• Administrative statistics are not properly collected, stored, and analyzed effectively,
limiting their use within the agricultural statistics system; and
• The competencies of individual agencies within the NASS are not effectively utilized by
other agencies.
Scant Statistics at the District Level
One of the major challenges of the current system is the lack of statistics at the district level.
Every data user group stated that their analyses were hampered by the lack of statistics at the
district level. The Directorate of District Statistics within UBOS oversees the production of
official statistics at the district level; however, district-level agricultural statistics are scant. The
Division of Statistics within MAAIF is working on producing statistics from administrative data
collected by agricultural extension workers. MAAIF currently produces statistical abstracts, but
data quality is hampered by poor farm record keeping, inadequate estimation procedures, and
logistical limitations affecting travel by extension workers. The new systems coordinating and
upgrading data collection by the NFASS have not yet become fully operational. Other agencies
within the NASS produced agricultural statistics on an as-needed basis or for the specific
commodity within their mandate. In summary,
• District-level statistics are often unavailable or of questionable accuracy;
• Resource constraints limit capacity for collecting accurate, timely data at the district
level; and
• Districts have varying levels of capacity or statistical personnel for collecting data and
producing agricultural statistics.
39
Personnel
Personnel needs vary among the agencies within the NASS. The DAES has two statisticians (one
senior and one junior) on secondment from MAAIF who handle different types of commodities.
The Division of Statistics in MAAIF has hired 20 statisticians to produce agricultural statistics.
However, other agencies outside of UBOS and MAAIF have reported not having the number of
people needed to produce agricultural statistics. Additionally, statistics students do not see a
future in agricultural statistics. Consequently, training new statisticians in agricultural statistical
methods has been challenging. In summary,
• Not enough statisticians are present in agencies outside of UBOS and MAAIF’s Statistics
Department that produce agricultural statistics and
• There are not enough new statisticians being trained in agricultural statistical methods
and apparently no specialization of agricultural statistics.
Technological
ICT equipment needs vary within agencies. The DAES reported having enough computers to
perform their duties. The Division of Statistics within MAAIF has received ICT equipment from
external stakeholders. However, other agencies expressed the need for new or updated equipment
to perform their duties.
However, two deeper ICT capacity needs among all agencies were uncovered by our
conversations with key stakeholders. First, all parties said that emerging data collection
technologies have not been implemented. As an example, they noted the increasing use of mobile
phones and the number of data collection software packages that could be used for data
collection. Second, all agencies stated that no centralized internal databases are used to store
agricultural data and statistics. They see these databases as becoming necessary as the volume of
data grows. Furthermore, these databases could reduce the lack-of-access problem among data
users by providing an electronic portal to the needed statistics. In summary,
• Emerging data collection technologies are not adequately used;
• As data volume increases, the lack of a centralized database and the opportunity for
improved analytics is a challenge; and
• Mobile phones and data collection software are not adequately used.
Financial
Above all, the main threat outlined within this report and in previous capacity assessments is the
lack of a dedicated and renewable source of funding for agricultural statistics. Currently, the
Government of Uganda does not create set budgets for all agencies participating in the national
agriculture statistics system. Some programs that are of interest to the government received
funds, which were used to establish a budget to collect statistics. However, not all agencies have
this type of dedicated funding. This hampers the ability of the Uganda NASS to produce
agricultural statistics in many ways: hiring data collectors and statisticians is difficult, ICT
upgrades and replacements are not occurring, data collection efforts are reduced or canceled,
40
statistics releases are delayed or eliminated, research into new statistical methodologies is
lacking, and the administration of the statistical agencies is constrained.
In summary, there is no established funding source to produce, improve, and maintain the system
for agricultural statistics. As the world COA approaches, there is an urgent need for funding the
systems and personnel required for Uganda to participate effectively in this international effort.
41
Chapter 6: Recommendations for Strengthening Agricultural
Statistics in Uganda
Based on the abovementioned findings, the current structure has many useful aspects that need to
be harmonized to strengthen the system.
Figure 6. 1: 6.1 shows a structure that could potentially help improve coordination of Uganda
NASS. UBOS shall remain the agency in charge of coordinating and collecting the official
agricultural statistics that continue to be informed by the administrative data collected by
MAAIF. UBOS would act as the agency in charge of coordinating and collecting the official set
of agricultural statistics. It would work with both MAAIF and the other agencies that collect
administrative data and other agricultural statistics to coordinate the work and assess statistical
methodologies. Furthermore, national- and district-level statistics would be created based on the
work performed by MAAIF district-level personnel for administrative data and district
statisticians at district offices. Finally, NGOs would work with UBOS, MAAIF, or the other
agencies collecting data to provide the necessary statistics for their projects.
The NASS will be successfully improved if data collected at the district level can be modernized
with methodologies consistent with international guidelines, more efficient means of collection,
trained staff at the district level, and the resources to properly collect the data. The NFASS
Project is developing a data center that can collect, assess data quality and analyze administrative
data coming from the field. UBOS will require district level statisticians who can administer
official data collection processes. There are multiple plans in place for many of these
improvements to be made. Oversight of improvements to the overall system will be required to
ensure that each agency is effectively meeting their commitments to the system’s improvement.
The assessment team has also reviewed a system for regional statisticians that could provide a
more uniform approach among districts. While this may be a technically valid intervention, it is
outside the norms of Ugandan administrative structures and was reconsidered. However,
oversight of the development of district-level statistical capacity will remain a challenge for
donors, UBOS and MAAIF.
42
Figure 6. 1: Harmonized Uganda NASS
UBOS • Other agencies (BOU, URA etc)
• NGOs
• Minimum Core Set of Statistics
• District-Level Statistics
MAAIF
- Districts- MAAIF Agencies (e.g NARO, UCDA,NAADS, DDA etc)
This new structure is similar to the Health Management Information System for collecting health
statistics and the Education Information System for collecting education statistics. In these
systems, the data providers (health centers and schools, respectively) submit their data to a
central database from which the relevant statistics are derived. However, agricultural data
collection differs in that farms are not required to provide their data to UBOS. Furthermore,
farmers do not have the same level of ITC capacity to transmit their data to a central data
repository. Instead, administrative data is collected via extension agents and MAAIF offices and
transmitted to the MAAIF data center and made available to stakeholders. This harmonized
system builds upon development of UBOS systems for collecting official statistics and the
43
NFASS Project’s establishment of a data center, analytical capacity, and administrative data
collection processes.
UBOS will act as the clearinghouse for methodologies for generating the official set of statistics
and support the creation of methodologies for other agencies. UBOS will work with Makerere
University to test and implement new methodologies for data collection and estimation. Both
agencies will make their methodologies transparent to data users. This is important because the
current statistical methodologies need to be made relevant and transparent before data users will
accept them.
This system relies on the strength of its personnel. People with training and experience in data
collection and statistical estimation are needed at the national and district levels to produce
accurate and relevant statistics. Enumerators will be hired from and collect data within their
home districts, and training will be provided by their associated regional offices. This strategy
will have the dual benefits of using people with knowledge of the district and improving
employment within the district.
The evolving system will need increased ITC capacity. Data collection on mobile devices will
allow for the district enumerators to quickly collect data. Additional ITC infrastructure will be
needed to support the movement of data between the national and district offices and UBOS for
official data and MAAIF for administrative data. It will also allow for the most efficient
implementation of best practices for agricultural statistics. UBOS will be responsible for
collaboration with stakeholders to choose the most appropriate guidelines to use since it is
responsible for disseminating the core set of statistics. These guidelines can be distributed to its
partners and established within the surveys that collect the data for the official core set of
statistics.
Financing of the system will come from both internal and external institutions. The Government
of Uganda must recognize and support the NASS and provide it with its own stable line of
funding. External stakeholders seeking data for their projects, such as NGOs, will work with and
provide the funding to UBOS, which will collect and supply the necessary statistics.
Based on the abovementioned results, the following recommendations are proposed to harmonize
the NASS, improve coordination between the existing structures, and develop a system for
obtaining subnational-level statistics (Table 6.1). The costs are based on current knowledge of
Ugandan finances and were calculated using the assumptions outlined in Appendix 2. In many
instances, the cost estimates include processes e.g. workshops and consultations that leads to the
production of the required outputs.
Table 6. 1: Recommendations for a Harmonized NASS
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
Institutional
Establish the Global
Strategy core
minimum set of
Update the UBOS
Act of 1996 to
include the set of
National UBOS Short term 5,000 One off
44
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
statistics as the set
of official
agricultural
statistics
statistics as part of
the mandate for
UBOS
Communicate the
set to agencies
who collect
agricultural
statistics
UBOS Short term 10,000 One off
Meet with external
stakeholders to
determine their
data needs
UBOS Short term 2,000 One off
Clearly delineate
the responsibilities
between agencies
for collecting the
core minimum set
of statistics
Establish an
agricultural
statistics sector
committee
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 3,000 One off
Draft a charter and
rules
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 10,000 One off.
Cost
includes
processes
e.g.
workshops
and
consultations
that leads to
the output
Set the UBOS
DAES director as
chair and MAAIF
statistics director
as cochair
UBOS and
MAAIF
Short term 6,000
One off
Identify the
supporting
members
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 1,000 One off
Produce clear and
defined roles and
responsibilities for
each type of
agricultural
statistic among the
members
UBOS,
MAAIF and
agencies
producing
agricultural
statistics at
national and
district level.
Short term 13,000 One off
Engage a
coordination
committee for
agencies that
produce agricultural
Establish a formal
agricultural
statistics
coordination
committee
National UBOS,
MAAIF and
agencies
producing
agricultural
Short term 3,000 Yearly
45
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
statistics statistics
Draft a charter and
rules
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 10,000 One off
Establish the
governing body
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 1,000 One off
Identify the
supporting
members
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 1,000 One off
Schedule regular
biannual meetings
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short to
medium
term
15,000 Yearly
Serve as the bridge
between agencies
in coordinating
agricultural data
collection and
addressing cross-
agency issues
Coordination
committee
Medium
term
20,000 Yearly
Establish working
committees that
codify
methodologies for
collecting the core
minimum set of
statistics
Establish a formal
agricultural
statistics technical
committee
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 3,000 Yearly
Draft a charter and
rules
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 10,000 One off
Establish the
governing body
between UBOS,
MAAIF, and
UBOS,
MAAIF and
Makerere
University
Short term 3,000 One off
46
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
Makerere
University
Identify the
supporting
members
UBOS,
MAAIF and
Makerere
University
Short term 3,000 Yearly
Schedule annual
meetings
UBOS,
MAAIF and
Makerere
University
Short to
medium
term
13,000 Yearly
Produce
statistically sound
data collection and
estimation
methodologies for
the core minimum
set of statistics that
are achievable for
each agency
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Medium to
long term
749,000 One off cost
Work with
supporting
agencies to
establish those
methodologies
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Medium
term
24,000 One off
Develop a calendar
of statistical
releases via
statistical abstracts
and other methods
for dissemination
Map out the
production cycles
of the production
commodities
National
MAAIF
Medium
term
76,000 Yearly
Work with data
users to determine
the times each
statistical release
will have the
greatest impact
and relevance
MAAIF and
UBOS
Medium
term
33,000 Yearly
Work through the
coordination
committee to
organize the
schedules of each
agency's activities
UBOS and
MAAIF
Medium
term
58,000 One off
Create a yearly
calendar of
statistical releases
for the core set of
statistics
UBOS Medium
term
65,000 Yearly
Update the
calendar within the
coordination
committee
UBOS and
MAAIF
Short to
medium
term
46,000 Yearly
47
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
Promote the benefit
and utility of
statistics outside of
the NASS
Prepare an
advocacy plan for
promoting
agricultural
statistics to the
public
National UBOS and
MAAIF
Short to
medium
term
20,000 Yearly
Meet regularly
with governmental
officials
concerning the
needs and
achievements
surrounding
agricultural
statistics
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short to
medium
term
41,000 Yearly
Promote statistical
releases using
traditional and
only media
platforms
UBOS Short to
medium
term
225,000 Yearly
Organize a yearly
agriculture
statistics forum
with governmental
officials and
external
stakeholders
Coordination
Committee
Medium
term
750,000 Yearly
Methodological
Develop
commodity-specific
methodologies for
the collection of
agricultural
statistics
Work with subject
matter experts to
identify the
subject-specific
needs for data
collection
National UBOS and
MAAIF and
Academic
and research
Institutions
Short to
medium
term
9,000 Yearly
Collaborate with
academic
institutions to
research
statistically sound
and current
methodologies
UBOS and
MAAIF and
Academic
and
Research
Institutions
Short to
medium
term
13,000 One off
Conduct pilot
studies to test the
new
methodologies
UBOS Medium
term
62,000 One off
Implement the new
methodologies
within the existing
agricultural
statistics program
UBOS and
MAAIF
Medium
term
42,000 One off
Develop
methodologies for
Catalog current
administrative data
National UBOS and
MAAIF
Short term 9,000 One off
48
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
creating agricultural
statistics from
administrative data
sources
Assess the
adequacy of use of
each source within
the NASS
UBOS Short term 7,000 One off
Research
methodologies to
incorporate
appropriate
administrative data
sources in data
collection and
estimation
UBOS Short term 9,000 One off
Prepare a plan
within each agency
stating how
administrative data
will be used for
agricultural
statistics
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 7,000 One off
Coordinate
between agencies
for the desired
administrative data
UBOS Medium
term
7,000 Yearly
District
Establish statistical
personnel in district
offices whose sole
purpose is to collect
agricultural data
and disseminate
district-level
agricultural
statistics. They will
be responsible for
coordination with
MAAIF and other
agencies collecting
statistics in their
districts
Identify the
physical,
technological, and
personnel needs
for each office
District UBOS Medium
term
5,000 One off
Promote statistics
functions within
the district offices
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Medium
term
70,000 Yearly
Establish a
sustainable line of
funding for the
district statistics
personnel,
activities, and
logistics
UBOS and
MAAIF
Medium
term
Per regional
office needs
Yearly
Promote the utility
and benefit of
agricultural
statistics to farmers
Meet with farmers’
groups on a regular
basis to discuss
agricultural
statistics needs and
activities
District UBOS and
MAAIF
Short to
medium
term
1,310,000 Yearly
Promote data
collection efforts
UBOS and
MAAIF
Short to
medium
85,000 Yearly
49
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
through traditional
and online media
platforms
term
Offer a platform
for farmers to
provide input on
agricultural
statistics
UBOS and
MAAIF
Medium
term
73,000 Yearly
Personnel
Hire qualified
statisticians/develop
skills in agencies
that produce
agricultural
statistics
Conduct a
personnel needs
assessment in each
agency
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Medium
term
8,000 One off
Promote
agricultural
statistics to
statistics students
National UBOS Medium
term
47,000 Yearly
Strengthening
computing skills in
agencies that
produce agricultural
statistics
Conduct a
personnel needs
assessment in each
agency
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Medium
term
8,000 One off
Technological
Utilize innovative
data collection
software for mobile
devices
Research available
software for data
collection on
mobile devices
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short to
medium
term
14,000 One off
Perform pilot
studies on the
effectiveness of
collecting and
transmitting data
UBOS Medium
term
59,000 One off
Create the
infrastructure for
transmitting and
storing the
collected data
UBOS Medium
term
1,000,000 One off
Train enumerators
on the use of the
data collection
software
UBOS and
MAAIF
Short to
medium
term
1,285,000 One off
Create a database of
agricultural
statistics for data
users
Draft a plan for the
aggregation and
storage of the core
minimum set of
statistics and other
National UBOS and
MAAIF
Short to
medium
term
13,000 One off
50
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
agricultural
statistics
Obtain the ICT
equipment for the
database
UBOS and
MAAIF
Medium
term
Per needs
assessment
One off
Reinstate the use
of Country Stat to
disseminate
agricultural
statistics to an
international office
UBOS Medium
term
27,000 One off
Prepare and update
a metadata
dictionary
UBOS Medium
term
50,000 One off
Hire database
specialists to
maintain the
database
UBOS Medium
term
706,000 One off
Provide a portal
for data users to
access the data
UBOS Medium
term
74,000 One off
Update the current
computer and
network systems
within agencies
Conduct a
technology needs
assessment for
agricultural
statistics within
each agency
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short to
medium
term
35,000 One off
Purchase computer
equipment
specifically for
agricultural
statistics
UBOS Medium
term
Per needs
assessment
Update software for
data collection and
analysis within
agencies
Determine the
statistical
estimation needs
for the type of
statistics produced
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short to
medium
term
13,000 One off
Review the current
software used for
statistical
estimation
UBOS Short term 8,000 One off
Purchase the
necessary software
UBOS Short term Per needs
assessment
One off
Develop the ICT
strategy for
agricultural
statistics collection,
analysis, and
dissemination
Identify the
technological
needs and capacity
for the core
minimum set of
statistics within the
National UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 72,000 One off
51
Area of
Recommendation Activity Level
Responsible
entity
Timeframe Average
Yearly Costs
(US$)
Remarks
agencies
Draft an overall
ICT plan for
agricultural
statistics at the
national and
district level
UBOS and
MAAIF
Short term 3,000 One off
Work with the
chief information
officer (CIO)
within each agency
to create an agency
ICT plan for
agricultural
statistics
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short term 22,000 One off
Financial
The Ugandan
government must
establish and
maintain funding
for agricultural
statistics and data
collection
Identify
‘champions’ of
agricultural
statistics within the
government
National UBOS and
MAAIF
Short term 12,000 One off
Lobby for
continued funding
to be put into the
national budget
UBOS and
MAAIF
Short to
medium
term
32,000 One off
Training
Train district- and
national-level staff
in emerging
methodologies and
data collection
processes
As methods and
data collection
evolves, conduct a
training needs
analysis to develop
training plans
National
and
District
UBOS,
MAAIF and
agencies
producing
agricultural
statistics
Short to
medium
term
To be decided Yearly
Note: Short term:1-3 years; medium term: 3-5 years; long term: 5-10 years
Please see Appendix 2 for more details, including cost assumptions
World Bank Support for Improving Agricultural Statistics
There are two windows for World Bank support for improving agricultural statistics in Uganda.
The first window is through agriculture projects under the MAAIF, with the Agricultural Cluster
Development Project being restructured to include a provision for strengthening the Statistics
Unit. The second window is through Statistics Payment for Results (PforR) Program for
generating better and more accessible data to inform policy-makers and contributing to
strengthening statistical capacity. Funding through these windows can be used to support four
key interventions: (i) developing the legislative framework for agricultural statistics; (ii)
developing the legislative framework for data sharing between county governments and MoALF;
52
(iii) establishing structures where users and producers of agricultural statistics interact; and (iv)
developing a Seasonal Agricultural Survey (SAS).
53
Chapter 7: Global Best Practices for Agricultural Data
Country Example of Agricultural Data Collection and Survey Programs
The World Bank highlighted the role of South-South Learning in building capacity around
agricultural statistics in Africa. Two countries: Rwanda, which is part of the East African
community, and South Africa, can provide opportunities for learning and country case studies.
Rwanda has a very good agricultural survey program while the South Africa administrative data
collection experience provides some pointers for improving data collection. In addition, as part
of the action plan, the team recommends undertaking country study tours and/or desk-based
research to gathering learnings relevant to Uganda in terms of agricultural survey programs but
also a devolved structure where statutory powers are delegated from the central government to
the subnational level.
Rwanda
The National Institute of Statistics of Rwanda conducts two survey programs around agricultural
statistics.
National Agricultural Survey
The National Agricultural Survey (NAS), last conducted between September 2007 and August
2008, collected information on the two agricultural seasons and covered a sample of 10,080
agricultural households over 30 districts.
The survey collects data on
• Demographic and social characteristics of agricultural farmers;
• Farms characteristics;
• Agricultural practices and crop production;
• Livestock practices and production;
• Fishery, aquaculture, and beekeeping practices;
• Forestry practices and income; and
• Food stocks and nutrition.
SAS
The SAS aims to cover all three agricultural seasons in Rwanda: Season A, which starts in
September and ends in February of the following year; Season B, which commences in March
and ends with June of the same year; and Season C, which starts in July and ends in September
of the same year. The National Institute of Statistics of Rwanda (NISR) conducted the first SAS
in 2013 and the last survey was conducted between September 2016 and February 2017. The
54
respondents of the survey are categorized into two groups, namely, agricultural operators (small-
scale farmers) and large-scale farmers (LSFs). The NISR classifies LSFs according to specified
criteria, namely, farmers growing crops on 10 ha or more of land or any farmer raising 70 or
more cattle, 350 goats and sheep, 140 pigs, or 1,500 chicken or managing 50 bee hives.
The survey collects information on the characteristics of the agricultural operators, the farm
characteristics including the area yield and production, agricultural practices, inputs, equipment,
and use of crop production (NISR 2016). The survey uses multiple-frame sampling techniques
based on probability sampling and estimation techniques combining an area and list frame.
Imagery with a very high resolution of 25 cm is used to divide the county into strata (12 strata in
total). The survey interviewed a sample of 195 LSFs (out of 774) and 5,089 of a total of 25,346
agricultural operators. Data collection is undertaken through paper-based questionnaires but data
entry was completed through the CSPro data entry software, while summary tables were created
through SPSS and Excel.
A total number of 540 segments were spread throughout the country as coverage of the survey,
with 25,346 and 23,286 agricultural operators in Season A and Season B, respectively. From
these numbers of agricultural operators, subsamples were selected during the second phases of
Seasons A and B. Furthermore, the total number of enumerated LSFs was 774 in Season A and
622 in Season B. Season C considered 152 segments counting 8,987 agricultural operators from
which 963 agricultural operators were selected for survey interviews.
Table 7.1 shows the five strata that were selected for sampling based on cultivated land and other
land use characteristics.
Table 7.1: Land use strata codes, definition, and areas
Stratum Description Total (ha) Percent
1.1 Intensive agricultural land (Season A and B) 1,479,081 81.9
1.2 Intensive agricultural land (Season A and B with potential for C) 48,388 2.7
2.1 Other marshlands 95,821 5.3
2.2 Marshlands potential for rice 20,201 1.1
3.0 Rangeland 133,849 7.4
10.0 Tea plantations 28,763 1.6
Total agricultural
land
1,806,103
Source: SAS, NISR 2016.
The results of the SAS are presented based on the five strata defined. Other sources of
agricultural data in Rwanda include:
• Comprehensive Food Security and Vulnerability Analysis and Nutrition Survey
(CFSVA) (2012);
• Census of Population and Housing (most recent in 2012); and
• Integrated Household Living Conditions Survey (most recent in 2015).
55
South Africa
Department of Agriculture, Forestry, and Fishing13 (Administrative Data)
The following institutions exist under the ambit of the department:
• Meat Inspection Scheme. Setting out of the legislative mandate, authority for inspection
services, procedures, and standards. Inspection services also distinguish between low-
frequency slaughter houses and high-frequency slaughter houses and collect data in these.
• Crop Estimating Committee.14 Comprises officials from the following institutions:
Department of Agriculture, Forestry and Fisheries; Provincial Departments of
Agriculture; various Agricultural Research Council (ARC) -Institutes (Soil, Climate and
Water; Small Grains Institute; and Grain Crops Institute); Bureau for Food and
Agricultural Policy (BFAP) and Statistics South Africa (SA).
• Abstract of Agricultural Statistics. South African Grain Information Services (SAGIS)
is the main source of information on crop production, boards such as Sugar Cane Board,
Customs and Excise Data (tax authority and South African Revenue Service (SARS)),
Red Meat Abattoir Association, Cape Wool SA, and Milk SA.
Figure 7.1 below indicates the Organogram of the Ministry of Agriculture, Forestry, and Fishing,
South Africa
13 http://www.daff.gov.za/. 14 South African Grain Information Services: http://www.sagis.org.za/.
56
Figure 7.1: Organogram of the Ministry of Agriculture, Forestry, and Fishing, South Africa
Source: MoALF South Africa Strategic Plan 2015/2016–2019/2020.
Statistics South Africa (Survey and Census)
Statistics South Africa (Stats SA) based on the Population Census of 2011 published an
‘agricultural households’ report. This report covers three types of agriculture, namely,
subsistence, smallholder, and commercial. The census provided some information on subsistence
and smallholder agriculture but excluded important data on land farmed and yields.
The Census 2011 questionnaire included questions on the following agricultural activities:
1. What kind of agricultural activity is the household involved in?
2. How many of the following (livestock) does the household own?
3. Where does this household operate its agricultural activities?
In addition, a regular survey program also collects information related to agriculture through two
surveys:
1. The Quarterly Labour Force Survey (QLFS) collects detailed information on employment
in the agricultural sector on a quarterly basis. It is a panel survey in that 25 percent of the
sample is rotated out every quarter. Employment in the sector can be disaggregated by
sex, age, and province as well as remuneration levels. The sample is representative at a
provincial level and within provinces at the metro/non-metro level.
57
2. The Annual General Household Survey (GHS) collects information on food security and
agricultural activity based on a sample of 21,228 households. Characteristics of
households involved in agriculture, main reason for involvement in agricultural activity,
and type of agricultural production activity are collected (livestock, poultry, grain and
food crops, industrial crops, fruit and vegetables crops, fodder grazing, pasture grass for
grazing). The sample is representative at a provincial level and within provinces at the
metro/non-metro level.
Sweden
The System of Official Statistics in Sweden
Statistics Sweden is a central government authority for official statistics and other
government statistics. In 1994 a statistical reform was implemented of Sweden’s official
statistics, implying a decentralised system for official statistics and 25 government
authorities were given responsibility for official statistics in defined sectoral areas instead
of a centralised system and one governmental authority responsible. One of the main
purposes of the 1994 statistical reform was to give the users more influence over the
statistics, for flexibility and that the efficiency of statistics production would improve.
The System for Official Statistics includes the statistics, statistical products, metadata, the
production systems, final observation registers, publications, separate tables and
databases. Databases can be interactive or include fixed tables that the user cannot
change. The system also includes laws, ordinances, regulations, general
recommendations, guidelines, tools (that are developed for the system such as methods,
classifications, etc.), the statistical authorities, the Council for the Official Statistics, and
Statistics Sweden as the coordinating authority.
According to the decision by Parliament, the Government determines the subject areas
and statistical areas for which official statistics are to be produced, and which authorities
are to be given the responsibility. For the moment there are 22 different subject areas.
The statistical authorities decide on the content and scope of statistics within the statistics
area(s) for which unless otherwise specified by the government. The statistical authorities
also decide, in consultation with important users of the statistics and taking into account
the demands made by the European Union, which objects and variables are to be studied,
which statistical measurements and study domains are to be used, the periodicity of the
surveys etc. Except for Statistics Sweden there is normally no special appropriation for
statistics; funding for statistics is included in the authorities’ appropriation framework for
their main task. The System for Official Statistics includes the statistics, statistical
products, metadata, the production systems, final observation registers, publications,
separate tables and databases. Databases can be interactive or include fixed tables that the
user cannot change. The system also includes laws, ordinances, regulations, general
recommendations, guidelines, tools (that are developed for the system such as methods,
classifications, etc.), the statistical authorities, the Council for the Official Statistics, and
Statistics Sweden as the coordinating authority.
58
A Council for Official Statistics was established in 2002 with the purpose to improve
coordination and overall view of the system for official statistics. The Council, which is
an advisory body, deals with matters of principle concerning the availability, quality and
usefulness of the official statistics, as well as issues on facilitating the response process
for data providers. The Council works to improve cooperation between the statistical
authorities, and to develop and manage a statistics network. It consists of one chair and
six other representatives who are managers at the statistical authorities. The Council is
supported by a secretariat and different workgroups. All authorities responsible for
official statistics are invited to participate in the different workgroups. Due to the users of
official statistics the system and the cooperation is judged to function rather well. The
duties of the Council are set out in Statistics Sweden's Directives. The authorities to be
represented in the Council are appointed by Statistics Sweden after consultations with all
the statistical authorities. Members serve on the Council for a period of not more than
three years. Statistics Sweden’s Director General is Chair of the Council, and the Council
appoints its own Deputy-Chair.
To provide a picture of this, the statistical authorities annually complete questionnaires
on the provision of data and on costs and staff who work with the official statistics. The
authorities also submit a list of their active products. As a complement to this
information, special measurements have been made on punctuality and production time,
documentation, the use of the Official Statistics of Sweden (SOS) logotype and reporting
by sex in the statistics.
The cooperation within and improvement of the system Statistics Sweden, in its role as
coordinator, has the mandate to issue regulations to statistical authorities regarding
documentation, quality declarations and publication. The main coordination tool since the
Council was established has been coordination by cooperation (soft coordination) and the
development of a well-functioning infrastructure. Participation in the workgroups has
been on a voluntarily basis and great interest in participating has been observed.
Common guidelines for deciding what Official Statistics are and a definition of what a
statistical product is, for sufficient quality, for preliminary statistics, for the websites at
different authorities have been developed. There are specified routines for deciding on
which statistics are to be official. There is a database of all Official Statistics and all
changes in the statistical system are continuously registered in the database. It is therefore
possible to follow a statistical product from cradle to grave. The users have now one main
single point of contact with the Official Statistics via Statistics Sweden’s website, though
there is a decentralised system. There are slightly more than 300 statistical products
within the Official Statistics and they are described in a consistent manner on the website.
There is a common publishing plan that is continuously updated and there are links to the
different authorities' websites where Official Statistics are published.
To date, the cooperation has led to a common view of Official Statistics, an increase in
competence, more systematic assessments related to user needs of what should be
included in the Official Statistics as well as a much better overview of the content of the
Official Statistics. The authorities responsible for official statistics have generally
organised contact nets with their users. The availability of statistics for users who have an
59
interest in statistics covering different areas has improved. The work is still in an initial
phase. Today we deal with aspects of statistics such as quality, documentation, response
burden, use of administrative data and security of information. Other aspects will emerge
in the future. The value of systematic cooperation has the potential to increase as there
are mutual benefits which can be derived from the joint development of statistics and
common statistical systems rather than the development of separate solutions for each
authority
Best Practices for Agricultural Data: Probability Samples and Two-stage Multiframes
Evidence-based decision making relies on information that is based on timely, consistent, and
statistically sound information, from either probability sample surveys, censuses, or
administrative data. The widely used unscientific practice of ‘eye observations’ by agricultural
officers, farmer groups, village elders, and other local officials who provide an opinion on the
total areas planted and harvested is no longer an acceptable practice, especially in the context of
climate change and the importance of monitoring impact on food security.
In the absence of highly developed administrative data systems, the use of probability sampling
surveys is regarded as the most appropriate approach for obtaining robust estimates with
acceptable periodicity of data collection. A sample is the collection of data from a sample of
units, unlike a census that would contact all units in the population. With good fieldwork
planning and management, a well-designed sample survey can be completed relatively quickly
and is representative of the population with known probabilities and measures of sampling
variability. In addition, a well-designed sample for producing national estimates also require a
surprisingly small number of agricultural holdings.
Two-stage multiple-frame surveys use two or more sampling frames. One frame is an area frame
used to collect data from small farms and the other is a list frame to collect data from large
farms. List frames normally provide good coverage of the large commercial farms.
The use of multiple frames brings a great degree of flexibility to the statistician because the
sampling methods can be unique to each frame. The only requirement is the need to identify any
overlap between the two frames to avoid the possibility of any double counting. In addition, the
classification of farms as small, medium, and commercial is required.
Two-stage sampling is a means of surveying large populations using relatively small samples
and ensuring that all statistical units have an equal chance (probability) of being included in the
sample to be interviewed. The course of action is to divide the area to be surveyed into small
geographical units called ‘census enumeration areas’).
60
Box 7.1: Sampling frames for agricultural statistics
A Master Sampling Frame (MSF) forms the basis for the selection of probability-based samples of farms and
households. The first step in the development of the MSF is to identify the data items to be measured, for example,
the total production of maize, the number of beef cattle, or the changes in land cover. The MSF should link the farm
or agricultural holding, the household, and the land. The possible sampling frames are the listing of maize fields,
animals, people by gender, or land parcels. The MSF comprises a listing of the sampling units that would provide a
complete coverage of the population of interest. The listing of the sampling units can comprise the names of farm
operators (from an Agricultural Census), the names of households (from a Population Census), a list of commercial
agricultural enterprises not linked to households, or a list of area units defined geographically. The MSF is the joint
use of two or more of these listings of sampling units.
Source: GSARS 2015a.
International Initiatives That Can Be Leveraged to Build Capacity Around Agricultural
Statistics
Internationally there are a number of initiatives underway in Uganda to strengthen the NASS,
including the introduction of FAO’s Agricultural Integrated Surveys15 (AGRISurveys) program.
USAID are working with the FAO team and with UBOS and MAAIF on plans to take forward
the AGRISurvey assessment that was undertaken in Uganda in January 2018. This includes an
identification of the statistical indicators that the LSMS-ISA16 (from the Uganda National Panel
Survey) caters for and the gaps that the AGRISurvey would fill from the lists of core SDG
indicators and CAADP monitoring of agricultural statistics. AGRISurveys collects economic
data on farms and agriculture sector every year, while alternating modules. It is based on
standard methodology and tailored to country needs. The current funding includes USAID grant
to collect in 4 countries and BMGF grant to support initial TA in up to 15 countries. Table 7.2
summarizes the anticipated budget costs for AGRISurvey to be carried out on an annual basis
and Figure 7.2 summarizes the funding gap.
15 A farm-based modular survey that builds on an agricultural census and operates over a 10-year cycle, providing tthhee ccrriittiiccaall ddaattaa aa ccoouunnttrryy nneeeeddss ttoo uunnddeerrssttaanndd iittss aaggrriiccuullttuurraall sseeccttoorr.. 16 A household survey project that conducts multiple rounds of a nationally representative panel survey with a multi-
topic approach. In eight countries to date:
61
Table 7. 2: Projected AGRISurvey Budget Funding source 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028
15% 20% 30% 40% 50% 60% 70% 80% 90% 100% 100%
$67,500 $170,000 $255,000 $320,000 $400,000 $480,000 $560,000 $640,000 $720,000 $800,000 $800,000
70% 60% 50% 40% 30% 20% 10% 0% 0% 0% 0%
$315,000 $510,000 $425,000 $320,000 $240,000 $160,000 $80,000 $0 $0 $0 $0
5% 5% 5% 0% 0% 0% 0% 0% 0% 0% 0%
$22,500 $42,500 $42,500 $0 $0 $0 $0 $0 $0 $0 $0
10% 15% 15% 20% 20% 20% 20% 20% 10% 0% 0%
$45,000 $127,500 $127,500 $160,000 $160,000 $160,000 $160,000 $160,000 $80,000 $0 $0
Estimated
Survey Total $450,000 $850,000 $850,000 $800,000 $800,000 $800,000 $800,000 $800,000 $800,000 $800,000 $800,000
Funding Gap
Government of
Uganda
USAID Funding
BMGF Funding
(Open Data)
Source: Emily Hogue, FAO Statistics Division
Figure 7.2: Distribution of Budget Shares
Source: Emily Hogue, FAO Statistics Division. AGRISurveys requires partners to: i) support and advocate, particularly through data demand; ii) support the consolidation of agriculture data collection through the Government of Uganda; iii) support the institutional framework; and give financial support for the funding gap.
Other are the GODAN initiative, the Advanced Data Planning Tool (ADAPT) developed by
Partnership in Statistics for Development in the 21st Century (PARIS21), the Global Strategy for
Improving Rural and Agricultural Statistics, the FAO World Program for the Census of
Agriculture 2020 (WCA) as well as various data quality (Eurostat 2007) assurance frameworks.
PARIS21 ADAPT Tool
The tool has been designed to bring together stakeholders to develop the indicators framework
related to monitoring development outcomes. The frameworks can be to measure national
development plans or the Sustainable Development Goals (SDGs). The tool can also be used to
62
identify reporting, financial, data, or geographic gaps related to the data for measuring indictors
(World Bank, 2004)
One of the important elements of the ADAPT tool is its flexibility to map national priorities to
global requirements. The Costing Module supports stakeholders in estimating the cost related to
data collection for long-term planning and program-specific budgeting, once unit cost
information for specific data collections has been entered into the tool. Another important
element of the tool is to produce a gap analysis, for data (absolute data gaps, frequency, or
disaggregation gaps), methodology, capacity, and funding gaps. The gap identification, before
starting the process, requires stakeholders to undertake the costing of activities including
identification of activities where there is insufficient funding, while also identifying which SDG
indicators are not collected or where the data collection does not align with what is demanded.
The resulting plans can then be integrated into the country NSDS.
GODAN
The GODAN initiative “seeks to support global efforts to make agricultural and nutritionally
relevant data available, accessible, and usable for unrestricted use worldwide. The initiative
focuses on building high-level policy and public and private institutional support for open data.”
It is a voluntary association launched in October 2013, currently comprising over 600 partners
from national governments, non-governmental, international and private sector organizations.
The aims of GODAN are to
• Advocate for new and existing open data initiatives to set a core focus on agriculture and
nutrition data;
• Encourage the agreement on and release of a common set of agricultural and nutrition
data;
• Increasing widespread awareness of ongoing activities, innovations, and good practices;
• Advocate for collaborative efforts on future agriculture and nutrition open data
endeavors; and
• Advocate programs, good practices, and lessons learned that enable the use of open data
particularly by and for the rural and urban poor.
This initiative can be used to support the initiatives to improve agricultural data collection
activities. It promotes collaboration to harness the growing volume of data generated by new
technologies to solve long-standing problems and to benefit farmers and the health of consumers.
Collaborations between the Public and Private Sectors
Collaborations between the public and private sectors around data collection and funding can
present opportunities for improving the quality of agricultural data through sharing of
information and freeing up of financial resources. There are a number of models for this
interaction.
63
PPP is one avenue for this collaboration, where the private sector can invest in technology
creation, adaption, and transfer through the investment in research and skills development and
the dissemination of knowledge, data, and scientific knowledge. FAO (2013a) identifies that the
contributions of the private sector can be financial and nonfinancial and engagements are based
on the principles of mutual collaboration and sponsorships. The six areas identified for
collaboration are
1. Knowledge management and dissemination;
2. Norms and standards setting;
3. Mobilization of resources;
4. Development and technical programs;
5. Policy dialogue; and
6. Advocacy and communication.
Data collaborative is a new form of partnership through which a number of stakeholders from the
public and private sectors and research institutions can share and use data to help solve public
problems. For this type of collaborations to be applied, there is a need to train data producers and
users, matching the public demand for data and the private supply of data in a secure and
confidential way, documenting activities and finally using experimentation and focusing on
scaling initiatives with potential.
In the sharing of data between the public and private sectors, it is important to set the
frameworks through which data sharing will occur, including establishing a code of practice,
fairness and transparency, security, governance, individual rights to access information and data,
and freedom of information (ICO 2011).
Technology and Quality Assurance Standards
Technology presents various opportunities to improve data quality and timeliness with which
data can be disseminated. However, technology is only one aspect of a successful survey design
and can only build on the existing good practices for data collection and the skills set of data
collectors. To ensure that quality data are collected, a Survey Quality Assessment Framework
(SQAF)17 checklist can be utilized. This framework asks questions around the survey process
17 A generic format for surveys is provided by the following resource prepared in collaboration with PARIS21:
Statistical Services Centre of University of Reading. 2009. “International Household Survey Network Survey
Quality Assessment Framework (SQAF).” http://www.ihsn.org/projects/survey-quality-assessment-framework-
SQAF.
64
and emphasizes checking, documentation, and the implementation of the systems to minimize
errors and ensure the completeness of information.
Box 7.2: Use of technology in collecting agricultural data
GPS
An important element of agricultural data is reliable information related to land, either cultivated land, grazing or
fertilized land, or wood land. However, farmers often are not able to provide their land size in a standard format.
In addition, the traditional measure using a rope in compass leads to sampling errors and is a very time-
consuming activity. The advances in geo-positioning and GPS provide the cropped area directly without the need
for distance and angle measurements.
Remote sensing
Remote sensing can be used to identify and monitor crops; this type of information combined with GIS can serve
as a useful tool regarding crops and assist in decision making around agricultural strategies. Remote sensing can
be used to identify crop status including stressed plants, crop yield estimation, and identification
Crop identification
By observing the various kinds of crops, it is possible to map the boundaries of the fields. Mapping of the
boundaries of land parcels provides information for the creation of cadastral maps. Cadastral maps are usually in
a vector format and in this form can be used in a GIS, along with other types of data (ownership, crop types
cultivated, and so on).
CAPI
CAPI is increasingly being used in the collection of data. It involves an interviewer collecting information from a
respondent via a questionnaire residing on a laptop, smartphone, or tablet.
CATI
Computer-assisted telephone interviewing (CATI) and self-administered web completion of questionnaires are
additional ways in which the high cost of personal interviewing can be reduced.
Software (examples)
Survey solutions is a tool for creating surveys using the World Bank CAPI platform and is provided free of cost.
The goal of the tool is to assist developing countries’ National Statistical Offices and other data producers with a
sustainable method for conducting complex and large-scale surveys. The tool provides functionality for data
capturing, survey, and data management.
CSPro refers to the Census and Survey Processing System and was developed by the U.S. Census Bureau. The
bureau maintains the system and makes it available at no cost. The system can be used for entering, editing,
tabulating, mapping, and disseminating census and survey data and is in use in a number of developing countries.
Technology should also be used in the dissemination of data. The OECD defines data
dissemination as “consisting of distributing or transmitting statistical data to users.” There are
various release media that can be used for dissemination purposes including the Internet; CD-
ROM; paper publications; files available to authorized users or for public use; fax response to a
special request; public speeches; and press releases. Dissemination formats according to the
Special Data Dissemination Standards (SDDS) include hardcopy and electronic formats that
detail the reference documents through which users can access the data described in the metadata
or any additional data not routinely provided.
65
Box 7.3: Use of technology in data dissemination: Examples of publishers that are Data Documentation
Initiative compliant and of data visualization tools
Nesstar Publisher
This is an editor for the preparation of metadata and data for publishing in an online catalogue. It is provided free of
charge and allows for the editing, creation, and exporting of data and is aligned to the Data Documentation Initiative
(DDI). The publisher includes tools to validate metadata and variables, compute/recode/label new or existing
variables to be added to a dataset before publishing and is multilingual covering a number of languages including
English, French, and Arabic (http://www.ihsn.org/software/ddi-metadata-editor).
Microdata Cataloguing Tool National Data Archive (NADA)
NADA is a web-based cataloguing system that serves as a portal for researchers to browse, search, compare, apply
for access, and download relevant census or survey information. It was originally developed to support the
establishment of national survey data archives but is increasingly being used across a number of organization across
the world.
Microsoft Power BI
It is a cloud-based service that allows for the creation of visualizations, reports, and dashboard by the users. It is
based on Excel and related PowerPivots.
66
Bibliography
AfDB (Agricultural Business Initiative). 2014. Country Assessment of Agricultural Statistical
Systems in Africa. Abidjan, Côte d'Ivoire: AfDB.
African Development Bank, Organisation for Economic Co-operation and Development, United
Nations Development Programme and Economic Commission for Africa (2013). African
Economic Outlook 2013, Structural Transformation and Natural Resources . African
Development Bank.
AfDB. 2014. Country Assessment of Agricultural Statistical Systems in Africa
https://www.afdb.org/en/knowledge/publications/country-assessment-of-agricultural-
statistical-systems-in-africa/
Global Strategy to Improve Agricultural and Rural Statistics: Report of the Friends of the Chair
on Agricultural Statistics. United Nations Economic and Social Council Statistical
Commission Forty ‐ first session. 2010 https://unstats.un.org/unsd/statcom/doc10/BG-
AgriStats.pdf
FAO. 2011. Improving Statistics for Food Security, Sustainable Agriculture, and Rural
Development: An Action Plan for Africa 2011-2015
http://www.fao.org/fileadmin/templates/ess/documents/meetings_and_workshops/Resour
ce_Partners_Oct2011/AfDB_Agriculture_long_web_PrintVersion_Final_04July2011.pdf
FAO. 2018. Uganda at a glance. http://www.fao.org/uganda/fao-in-uganda/uganda-at-a-
glance/en/Government of Uganda. 1998. Uganda Bureau of Statistics Act. Entebbe,
Uganda: Government of Uganda.
Government of Uganda. 2015. Second National Development Plan (NDPII) 2015/16–2019/20.
Kampala, Uganda: Government of Uganda.
Gourlay S., Kilic T., and Lobell D. 2017. “Could the Debate Be Over? Errors in Farmer-
Reported Production and Their Implications for the Inverse Scale-Productivity
Relationship in Uganda” Policy Research Working Paper 8192.
MAFAP (Monitoring African Food and Agricultural Policies). 2013. “Review of Food and
Agricultural Policies in Uganda.” MAFAP Country Report Series. Rome, Italy: FAO.
MAAIF (Ministry of Agriculture, Animal Industries, and Fisheries). 2011. National Agriculture
Policy. Entebbe, Uganda: Government of Uganda.
———. 2014. Agriculture Sector Strategic Plan for Statistics 2013/14–2017/18. Entebbe,
Uganda: MAAIF.
Nalunga, J. 2015. Agricultural Statistics Capacity Needs Assessment Report. Kampala, Uganda.
NAADS (National Agricultural Advisory Services). 2013. Cumulative Progress Report: FY
2011/2012–2012/2013.
67
http://www.naads.or.ug/files/downloads/August%2020%20CUMULATIVE%20NAADS
%20PROGRESS%20REPORT%20FY%202011_13_%20(3).pdf.
National Planning Authority. 2013. Uganda Vision 2040. Kampala, Uganda.
PARIS21 (Partnership in Statistics for Development in the 21st Century). 2016a. Our Mandate.
http://www.paris21.org/.
———. 2016b. Coordination and Monitoring. http://www.paris21.org/news-
center/news/strategy.
———. 2016c. Promoting Statistics. http://www.paris21.org/advocacy.
———. (2016d). National strategies for the development of statistics.
http://www.paris21.org/NSDS.
RTI. 2014. Participatory Local Organizational Assessment. Research Triangle Park, NC: RTI.
UBOS (Uganda Bureau of Statistics). 2009. 2008 National Livestock Census.
http://catalog.ihsn.org/index.php/catalog/3788/download/52466.
———. (2014a). 1012/13 Uganda National Household Survey.
http://www.ubos.org/onlinefiles/uploads/ubos/UNHS_12_13/2012_13%20UNHS%20Fin
al%20Report.pdf.
———. (2014b). 2014 Census. http://www.ubos.org/2014-census.
———. (2014c). Plan for National Statistical Development 2014/14–2017/18. Kampala,
Uganda: Government of Uganda.
———. (2014d). Strategic Plan 2013/14–2017/18. Kampala, Uganda: UBOS.
———. (2015). National Service Delivery Survey: 2015 Report.
http://www.ubos.org/onlinefiles/uploads/ubos/pdf%20documents/2015%20NSDS%20rep
ort.pdf.
———. (2016). Agricultural statistics. http://www.ubos.org/publications/agriculture/.
World Bank. 2010. Global Strategy to Improve Agricultureal and Rural Statistics. Washington,
DC: World Bank.
68
Appendix 1: Documents Reviewed
• Uganda Vision 2040
• Uganda National Agricultural Policy 2011
• Uganda Census of Agriculture 2008/2009
• Uganda Bureau of Statistics Strategic Sector Plan for Statistics 2013/14–2017/18
• UBOS Act of 1998
• UBOS Statistics Abstract 2016
• FAO CountryStat Panorama Report: Uganda 2008
• Ministry of Agriculture, Animal Industries and Fisheries Sector Strategic Plan for Statistics
2007–2011
• MAAIF National Coffee Policy
• The Republic of Uganda Plan for National Statistical Development 2007–2011
69
Appendix 2: Cost Assumptions.
The cost estimates are hinged upon recommendations made by the researchers for the different
interventions that can help improve the collection, reporting, and dissemination of the core set of
agricultural statistics. This report describes the Capacity Needs Assessment for Improving
Statistics for Sustainable Agriculture in Uganda.
The interventions are grouped into six categories: institutional, methodological, district,
personnel, technological, and financial. To execute each of the proposed activities, assumptions
were made for every activity in these categories. Although some general assumptions cut across
all the categories, other assumptions are specific for different numbers/units embedded to fulfill
the activity.
The Proposed Framework
The proposed recommended structural organization in the study should be noted.
Figure 6. 1: 8.1 presents the conceptual structure of the harmonized Uganda NASS and
demonstrates how agricultural statistics should be collected, processed, and disseminated. The
structure establishes UBOS as the body responsible for official agriculture statistics in Uganda.
Two different national-level committees are proposed: the technical committee (five members)
and the coordination committee, which will be composed of one representative from UBOS and
each of the seven sectors of MAIIF (eight members). Most of the costs to drive the agendas of
these committees lie in holding meetings, workshops, and coordination.
District statistics officers are proposed for each of Uganda’s 121 districts (Ministry of Local
Government 2017). At the district level, four people will be assigned to work on agricultural
statistics. The major cost drivers here are the training of personnel and providing them with
transport in the form of a motorcycle. The motorcycle shall be fueled and maintained by the
project.
Data collectors shall be hired at the subcounty level. These will be responsible for collecting
agricultural statistic for all agricultural needs and forwarding the data to the district, who will, in
70
turn, forward them to the zone for further analysis, interpretation, and dissemination. The costs at
this administrative level include transport in the form of a motorcycle and training and retraining
the data collectors. The breakdown of individual costs as summarized in the working
document—cost sheet—is presented in this appendix.
Exchange Rate (Table A2.1: ). The UGX/US$ exchange rate has been ranging between UGX
2,600 (July 2014) to UGX 3,600 (June 2017). An exchange rate of UGX 3,300 per US$ has been
applied to all activities for different years.
Workshops/seminars and meetings (Table A2.2: , Table A2.3: , and Table A2.8: ). Most of
the interventions involve training and sharing knowledge and experiences through workshops,
seminars, and meetings. The assumption is that participants in these activities shall be fully
facilitated with transport, daily subsistence allowances, stationery, a workshop venue, and (in
some cases) a consultant/trainer. The allowances for each have been benchmarked from the
prevailing rates (2017) and the Government of Uganda Revised Rates of Duty Facilitating
Allowances adjusted for by inflation.
Consulting and professional costs. We have costed for consultant fees per hour (Table A2.4: ).
For transparency, and according to the Public Procurement guidelines, the selection criteria for
such professionals require a transparent and fair process. This has also been factored in the costs.
Staffing costs. Some offices require permanent employees. We have budgeted for the selection
of these employees plus their remuneration (salary and benefits). The salary is estimated to be
payable in arrears based on hourly rates (Table A2.5: A2.4). Different rates apply to senior staff
and junior staff. A 22-day month has been proposed.
Office costs. Table A2.6: summarizes office operational costs. Daily costs were accumulated
into monthly totals. They are multiplied by 12 months to derive annual totals.
Advertising. Often, some activities require publication in media (Table A2.6). The newspaper
advertisement prices were derived from the prevailing prices for full-color (or black-and-white),
full, half, or quarter pages in Uganda’s dailies (The New Vision and Daily Monitor). The prices
for radio advertisements, announcements, and other forms were benchmarked from a select
section of Uganda’s radio stations with commendable listenership (CBS, Capital FM, Sanyu Fm,
and Radio 1/Kaboozi).
Capital items. Different long-term assets shall be needed to facilitate work at different levels
and stations. The list in Table A2.7: shows unit costs for each of the assets. Some of the capital
items are office tools and equipment. These include items to be purchased for distribution to
districts and subcounties and those that shall be used in offices.
Table A2.1: Exchange Rate
UGX Units US$
Exchange rate 3,300 — 1
Table A2.2: Workshop, Seminar, and Meeting Costs
71
Workshop Costs UGX Unit US$
Transport refund 200,000 To and fro 60
Subsistence (residential)
Kampala 330,000 Per night 100
Other places — Per night —
Teas — Per day —
Lunch
Kampala — Per meal —
Other places — Per meal —
Dinner
Kampala — Per meal —
Other places — Per meal —
Water — Per day —
Stationery and printing 20,000 Per person 6
Coordination and mobilization 500,000 Per day 150
Subsistence (nonresidential)
Senior officers 150,000 Per day 50
Junior officers — Per day —
Hire of venue (100+)
Kampala venues 2,500,000 Per day 760
Other venues 1,000,000 Per day 300
Hire of venue (small numbers)
Kampala venues 500,000 Per day
Other venues 200,000 Per day
Rapporteur 200,000 Per day 60
Meeting costs
Transport refund 200,000 Per day 60
Teas 15,000 Per day 5
Water 10,000 Per day 3
Stationery and printing 10,000 Per person 3
Coordination and mobilization 200,000 Per day 60
Hire of venue
Kampala — Per day —
Other places — Per day —
Rapporteur 200,000 Per day 60
Table A2.3: Consulting Costs
Consulting Costs UGX Unit US$
Expert selection process (designing the EOI, placing advertisements, screening,
selection, and contract award)
5,000,000 One-off 1,500
Professional fee 1,650,000 Per day 500
Reimbursables (including stationery and reporting) 3,000,000 Lump sum 900
Table A2.4: Staffing Costs
Staffing Costs UGX Unit US$
Recruitment costs (including selection and meetings) 5,000,000 Lumpsum 1,500
72
Staffing Costs UGX Unit US$
Advertising (print media) 5,000,000 Per advertisement 1,500
Induction 1,000,000 Lumpsum 300
Salaries
Senior positions (50,000/hour; 22 days) 8,800,000 Per month 2,600
Junior positions (30,000/hour; 22 days) 5,280,000 Per month 1,600
Contracted staff (10,000/hour; 22 days) 1,760,000 Per month 500
Driver, security, administration assistant, and so on 880,000 Per month 260
NSSF (10%)
Senior positions 880,000 Per month 260
Junior positions 528,000 Per month 160
Contracted staff 176,000 Per month 50
Driver, security, administration assistant, and so on 88,000 Per month 26
Staff feeding
Kampala 20,000 Per day 6
Other places 15,000 Per day 5
Medical insurance per year 2,000,000 Per person 600
Workers' compensation per year 70,000 Per person 20
Unemployed data collectors' wages/month 500,000 Per person 150
Note: NSSF = National Social Security Fund.
Table A2.5: Office Costs
Office costs UGX Unit US$
Office equipment (see Table 3.8)
-
Rent – Kampala 1,500,000 Per month 450
Rent - other places 1,000,000 Per month 300
Stationery
Internet 2,000,000 Per month 600
Communication 1,000,000 Per month 300
Website/technology 1,500,000 Per month 450
Transport per person (10,000/day) 220,000 Per month 70
Fuel cost per liter 4,000 Per liter 1
Fuel (20 L/day/vehicle) 1,760,000 Per month 500
Fuel (5 L/day/motorcycle) 440,000 Per month 130
Servicing and vehicle repairs 900,000 Per month 270
Other repairs and maintenance 1,000,000 Per month 300
Cleaning 300,000 Per month 90
Table A2.6: Advertising Costs
73
Advertising UGX Unit US$
Print media
Full page black and white 12,000,000 Per unit 3,600
Half page black and white 6,000,000 Per unit 1,800
Quarter page black and white 3,000,000 Per unit 900
Television per advertisement 1,000,000 Per unit 300
Press conference 5,000,000 Per unit 1,500
Radio per advertisement 50,000 Per unit 15
DJ mentions 80,000 Per piece 24
Spot advertisement 80,000 Per piece 24
Talkshow
30 minutes 2,000,000 Per piece 600
1 hour 3,500,000 Per piece 1,000
Online advertisement
Digital 60,000 per day 18
Uganda Business Directory per year 2,500,000 Per annum 760
Websites for other entities (50,000/day) 50,000 Per month 15
Social media advertisements (50,000/day) 70,000 Per month 21
Table A2.7: Office Equipment Costs
Capital Items UGX Unit US$
Motor vehicle - double cabin (Hilux 2017) 163,879,332 Per unit 49,660
Motorcycles (Suzuki 2017) 26,400,000 Per unit 8,000
Motorcycles (Bajaj 2017) 4,000,000 Per unit 1,200
Laptop (Dell) 3,000,000 Per unit 900
Desktop (Dell) 2,000,000 Per unit 600
Printer (heavy duty) 2,000,000 Per unit 600
Office chairs 550,000 Per unit 170
Waiting chairs 850,000 Per unit 260
Office table 1,500,000 Per unit 455
Boardroom table 5,000,000 Per unit 1,500
Boardroom chairs 3,000,000 Per unit 900
File cabinets 1,000,000 Per unit 300
Website/portal design 2,000,000 Per unit 600
Table A2.8: Meeting and Workshop Costs
Trainings UGX Unit US$
Facilitator 500,000 Per day 150
Transport refund: trainees 200,000 Per person 60
Transport refund: visiting officers 300,000 Per person 90
Transport refund: facilitator 300,000 Per person 90
Per diem 350,000 Per person/day 100
Stationery 20,000 Per person 6
Training material 20,000 Per person 6
Certificates 10,000 Per person 3
Venue 1,000,000 Per day 300
Mobilization and coordination 500,000 Per day 150
Rapporteur 200,000 Per day 60
Technology 2,500,000 Lump sum 760
Other logistics 2,000,000 Lump sum 600
74
Top Related