Browse by sub-sector
Open all 207 in the libraryCrop Production & Agronomy
Crop yields, area planted and harvested, input use, seed and variety adoption, and on-farm agronomic practices across the production cycle.
Output
Outcome
Climate-Smart Agriculture
Adoption of practices intended to raise productivity, build resilience to climate shocks, and reduce emissions, plus the assessment of farmer resilience itself.
Output
Goal
Agricultural Market Linkages
Farmer access to input and output markets, sales volumes and prices, value chain participation, and gross margins and returns from marketed produce.
Output
Goal
Natural Resource Management
Soil, water, land, and forest condition and use, conservation and restoration practices, and the management of shared or common-pool resources.
Output
Goal
Livestock & Animal Husbandry
Herd size and composition, animal productivity such as milk and offtake, animal health and husbandry practices, and livestock-derived income.
Output
Outcome
Agricultural Livelihoods & Production
Household-level production, food and income from farming, livelihood diversification, and the assets and capacities that underpin agricultural livelihoods.
Output
Outcome
Methods by outcome type
Match the method to the outcome, not to what is easiest to collect.
| Outcome type | Primary method | Observation | Key instruments |
|---|---|---|---|
| Yield & productivity | Crop cut (objective measurement) or farmer recall in a structured survey | Direct measurement required for crop cuts; recall otherwise | Crop-cut protocols, AGRISurvey and LSMS-ISA agriculture modules, FAOSTAT definitions |
| Adoption of practices & inputs | Structured farm and household survey, observation where feasible | Observation preferred for verifiable practices; recall common for input use | LSMS-ISA and AGRISurvey modules, CSA practice checklists |
| Market access & sales | Household and farm survey, trader or buyer records where available | Not required (transaction recall and records) | LSMS-ISA market and sales modules, DCED Standard results measurement |
| Natural resource & land condition | Field measurement, remote sensing, and resource-condition assessment | Direct or remotely sensed measurement required | SEEA accounts, remote sensing for land cover and area, plot-level field measurement |
| Livestock productivity | Structured livestock survey with herd and production recall | Not required (herd and production recall); weighing where feasible | AGRISurvey and LSMS-ISA livestock modules, FAOSTAT livestock definitions |
| Income & resilience | Household income and consumption survey; resilience index measurement | Not required (recall-based modules) | LSMS-ISA income and consumption modules, FAO RIMA, SHARP+, pro-WEAI for empowerment |
Common evaluation questions
The questions that anchor design and data-collection decisions, grouped by cluster.
Crop Production & Agronomy
Did the program change crop yields among participating farmers relative to comparable non-participants?
Evidence: Yields measured the same way at baseline and endline (crop cuts preferred over recall) with a comparison group. Design: Outcome evaluation with comparison group.
Was the intended agronomic package delivered and applied as designed?
Evidence: Farm survey on input use and practices triangulated with program monitoring against the protocol. Design: Implementation evaluation.
Climate-Smart Agriculture
Did farmers adopt the promoted climate-smart practices, and did adoption persist?
Evidence: Practice adoption survey at baseline, endline, and a follow-up point, with observation where practices are verifiable. Design: Outcome evaluation, before-and-after with follow-up.
Did the program change farmer resilience to climate shocks?
Evidence: A recognized resilience assessment (FAO RIMA or SHARP+) applied consistently across rounds. Design: Outcome evaluation.
Agricultural Market Linkages
Did the program change participating farmers’ sales volumes, prices received, or gross margins?
Evidence: Survey of sales and costs at baseline and endline with a comparison group, following DCED Standard logic. Design: Outcome evaluation with comparison group.
Were the intended market linkages established and used by the target group?
Evidence: Program monitoring of linkages and a farmer survey on market participation and channels used. Design: Implementation evaluation.
Natural Resource Management
Did the program change the condition or use of the targeted land, water, or forest resource?
Evidence: Resource-condition measurement (field plots and remote sensing) over time, structured along SEEA accounting categories. Design: Outcome evaluation, before-and-after.
Are resource management practices being maintained by the intended users after support ends?
Evidence: Repeat practice and governance assessment at program end and a follow-up point. Design: Sustainability / ex-post evaluation.
Livestock & Animal Husbandry
Did the program change animal productivity, such as milk yield or offtake, for participating households?
Evidence: Livestock survey with herd and production modules at baseline and endline, with a comparison group. Design: Outcome evaluation with comparison group.
Were improved husbandry and animal health practices adopted as intended?
Evidence: Household survey on husbandry practices triangulated with service and program records. Design: Implementation evaluation.
Agricultural Livelihoods & Production
Did the program change agricultural income or production for target households relative to a comparison group?
Evidence: Household income and production survey (LSMS-ISA style modules) at baseline and endline with a comparison group. Design: Outcome evaluation with comparison group.
Did the program affect women’s empowerment in agriculture within participating households?
Evidence: A project-level empowerment measure (pro-WEAI) applied at baseline and endline. Design: Outcome evaluation.
The standards that matter
Agriculture has a layered standards ecosystem built around statistical survey systems and a few specialized indices. Start with FAOSTAT for definitions and national-level agricultural statistics, use AGRISurvey (the 50x2030 Initiative) or LSMS-ISA modules for household and farm data so your numbers stay comparable, and pick measurement methods deliberately: crop cuts for objective yield, recognized resilience tools (FAO RIMA, SHARP+) for resilience, pro-WEAI for empowerment, the DCED Standard for market and value chain results, and SEEA for natural resource and land accounting. Because yield, income, and adoption are hard to attribute, document the measurement method and the comparison design alongside the indicator.
FAOSTAT ↗
FAO corporate statistics database with standardized definitions and country data for crops, livestock, inputs, and land use.
LSMS-ISA (World Bank) ↗
Living Standards Measurement Study Integrated Surveys on Agriculture: validated multi-topic household survey modules with a strong agriculture focus.
50x2030 Initiative / AGRISurvey ↗
Multi-partner program (FAO, IFAD, World Bank) standardizing farm and household agricultural surveys, building on AGRISurvey and LSMS-ISA.
FAO Climate-Smart Agriculture Sourcebook ↗
FAO reference framework defining climate-smart agriculture and guidance on assessing productivity, resilience, and emissions practices.
FAO RIMA ↗
Resilience Index Measurement and Analysis: FAO quantitative methodology for estimating household resilience to food insecurity, usable for impact evaluation.
FAO SHARP+ ↗
Self-evaluation and Holistic Assessment of climate Resilience of farmers and Pastoralists: a participatory survey tool for assessing climate resilience.
WEAI / pro-WEAI (IFPRI) ↗
Women’s Empowerment in Agriculture Index and its project-level version (pro-WEAI), a standardized measure of women’s empowerment in agriculture.
DCED Standard for Results Measurement ↗
Practical framework for measuring results in private sector and market systems development, including agricultural value chains.
SEEA (UN) ↗
System of Environmental-Economic Accounting: the international statistical standard for natural capital, land, and ecosystem accounting.
Common mistakes to avoid
Mixing crop-cut and farmer-recall yields without saying which
Objective crop cuts and recall-based yields are not interchangeable and can differ substantially. State the measurement method, since survey systems like AGRISurvey and LSMS-ISA distinguish them explicitly.
Claiming a yield or income increase without a comparison design
Yields and incomes move with weather, prices, and seasons regardless of a program. Attributing change requires a comparison group or another credible design, not a single before-and-after figure.
Using non-standard indicator wording
Reworded production indicators break comparability with FAOSTAT, AGRISurvey, and LSMS-ISA. Use established statistical definitions for area, yield, production, and herd measures.
Counting attendance or distribution as adoption
Training attendance and input distribution are outputs. Adoption requires a separate measure of practice use, verified by observation where the practice is observable.
Treating resilience as a generic score
Resilience is a defined construct in tools like FAO RIMA and SHARP+. Use a recognized resilience methodology rather than an ad hoc index if you report resilience as an outcome.
Reporting natural resource condition without a defined measurement basis
Land and resource condition needs a stated measurement basis (field plots, remote sensing, or SEEA accounting categories), not unspecified qualitative claims about improvement.
Frequently asked questions
How should I measure crop yield for a donor proposal?
State the method explicitly. Objective crop cuts give the most defensible yield estimate but are costly; farmer recall is cheaper but less accurate and tends to be reported differently. Survey systems like AGRISurvey and LSMS-ISA treat these as distinct measures, so name which one you use and apply FAOSTAT definitions for area, production, and yield so your figures stay comparable.
Can I attribute a yield or income increase to my program?
Not from a single before-and-after number. Yields and incomes change with weather, prices, pests, and the season independent of any program. To attribute change you need a comparison group or another credible design (for example, matched non-participants), and the same measurement method at each round. Otherwise, report the change as observed without claiming the program caused it.
What is the difference between adoption and reach?
Reach counts who you trained or what you distributed; those are outputs. Adoption measures whether farmers actually use the promoted practice or input, ideally verified by observation where the practice is observable in the field. Many plans conflate the two, which overstates results. Measure delivery and adoption separately.
How do I measure resilience or empowerment credibly?
Use a recognized methodology rather than a custom score. FAO RIMA estimates household resilience to food insecurity and is designed to support impact evaluation; SHARP+ is a participatory self-assessment of climate resilience for farmers and pastoralists. For empowerment in agriculture, pro-WEAI is the standardized project-level measure. Applying these consistently across rounds keeps results interpretable and comparable.
Which framework fits market and value chain work versus natural resource work?
For market systems and value chain results, the DCED Standard for Results Measurement is the recognized framework and emphasizes documenting the results chain and attribution. For natural resource and land outcomes, SEEA provides the international accounting structure, and condition is best measured through field plots or remote sensing rather than qualitative claims. Match the framework to the outcome type rather than forcing one across both.