Grounding a global tool—Principles and practice for agroecological assessments inspired by TAPE

Agroecology is widely recognized as able to “…contribute to transforming food systems by applying ecological principles to agriculture and ensuring a regenerative use of natural resources and ecosystem services while also addressing the need for socially equitable food systems within which people can exercise choice over what they eat and how and where it is produced” (High Level Panel of Experts on Food Security and Nutrition [HLPE], 2019, p. 13). It addresses sustainability and operates within local and planetary boundaries. This requires a shift from assessing agricultural performance using a narrow set of indicators (e.g., yield, dollars, and calories) to considering various additional environmental and social indicators. Initiatives that support transitioning to agroecology require a means of assessment for reasons that include diagnosing current status and addressing degradation of nature (López-Ridaura et al., 2002), coanalyzing with farmers the trends of farming practices for making decisions on livelihood and ecological sustainability (Goswami et al., 2017), envisioning possible futures, and characterizing the heterogeneity of socioecological contexts of farming systems in order to match them to a variety of options for farmers (Nelson and Coe, 2014). Assessments can also help determine whether intended changes are taking place in a way that is sustainable and beneficial to all stakeholders (Wiget et al., 2020). Information from assessments can be used to communicate the status of sustainability to stakeholders and build market and consumer confidence that production methods were mindful of the environment (Coteur et al., 2016). Agroecological assessments have been conducted in many parts of the world, but less so in the global south (Schindler et al., 2015).

In 2018, the Food and Agriculture Organization of the United Nations (FAO) was mandated to “…take the lead on developing methodologies and indicators to measure sustainability performance of agricultural and food systems beyond yield at landscape or farm level, based on the 10 elements of agroecology.” A result was TAPE (FAO, 2019). This was based on a review of many existing frameworks, which are discussed in the TAPE documentation, so we will not repeat that here, followed by consultation and some testing. TAPE is based on the 10 Elements of Agroecology (FAO, 2018) and uses them to frame choice of indicators and data collection. It is important to understand that FAO sees TAPE as providing “…a global analytical framework and a supporting database to assess the multidimensional performance of agroecology” (FAO, 2019). Thus, while they acknowledge the role of localized data collection, they envisage some standardization of data to allow global comparative data to be accumulated.

TAPE consists of 3 main steps. Step 0 involves a qualitative description of the context. Step 1, also labelled “Characterization of Agroecological Transition (CAET),” is focused on describing the agroecological state of the food and agriculture system. Step 2 concerns measuring performance of the system, assessing what it is producing and providing as economic, social, and environmental services. In this paper, we are concerned with Step 1¹, the characterization or description of “how agroecological” a system is, and henceforth refer to it as CAET.

CAET characterizes systems by looking at each of the 10 Elements² in turn. For each Element, there are 4 to 6 indicators describing system characteristics. Each of those indicators is defined on a scale from 0 (not agroecological) to 4 (most agroecological). An example is reproduced in Figure 1. The indicators are designed for rapid assessment by anyone with knowledge about the system in question (e.g., farmers, technicians, or others) and does not require direct observation or measurement in the field.

While TAPE has been used for assessments and some results published, there is little literature on experiences from using it or critical assessments of it. The team from FAO that developed TAPE have published an extended description of the methods (Mottet et al., 2020). A recent review of assessment frameworks mentioned TAPE but was not based on field experience (Wiget et al., 2020). TAPE was critically evaluated in France, with the conclusions that it needs adaptation to local conditions and context (Anthonioz, 2021).

The CCRP of the McKnight Foundation supports projects to conduct agroecological research and action. The overall aim is to bring about an agroecological transformation of food and agricultural systems through generating options for smallholder farmers, demonstrating possibilities, providing evidence, and influencing others. Many of the projects use a farmer research network (FRN) approach involving collaboration between farmers, extension staff, scientists, nongovernmental organization (NGOs), and other stakeholders to generate agroecological intensification options that are based on local and scientific knowledge and are responsive to the heterogeneous contexts of farmers. In East and Southern Africa (ESAf), the CCRP works with about 15 projects in Kenya, Malawi, Tanzania, and Uganda. Many of these projects have an interest in making assessments to characterize systems in agroecological terms. In addition, CCRP had provided some support to FAO for the development of TAPE, hence it was natural for the program to look carefully at how the TAPE tool could be used. Thus, when project teams in the region considered making some sort of agroecological assessment, TAPE, and CAET in particular, was among the approaches investigated.

The experience with trying to use CAET resulted in lessons that could be useful for others considering using the tool and hence the first aim of this paper is to share those lessons. It is important to recognize that the experiences reported here originate from our intention to use TAPE and CAET for agroecological assessments. We did not set out to develop a comprehensive approach to agroecological assessment nor to do a comprehensive evaluation of CAET. However, reflecting on the experiences and lessons from trying to use CAET led to proposed principles and a generic process that can be used by anyone planning an agroecological assessment of a food and agriculture system. The second aim of this paper is to share those proposals so they can be evaluated and refined by others.

The regional program of CCRP in ESAf supported a series of activities that provided the experience and evidence for us to learn about CAET, learning that is reported in this paper (Figure 2). In this section, we summarize the activities.

• 1. Introducing TAPE

A workshop in Arusha, Tanzania, in July 2019 aimed at strengthening a systems-focus within integrated agroecologically oriented research and action. We introduced agroecology frameworks as means of describing and analyzing properties and processes of an agroecological system and agroecological assessment tools. We included FAO Elements (FAO, 2018), HLPE (HLPE, 2019), Gliessman (2016), and CIDSE (2018) agroecology frameworks, which are well elaborated in the introductory section of TAPE (FAO, 2019) and overviews of TAPE, NaRIA (Natural Resource Integrity Assessment), and RISE (Response Inducing Sustainability Evaluation) as examples of agroecological assessment tools. These tools were discussed in relation to planned project work, the extent to which they are easy to use, enable assessment of the multidimensionality and complexity of agroecology; and allow localization, codevelopment, and participatory field assessment with farmers.

Projects came to the consensus opinion that Step 1 of TAPE, CAET, is a tool that could be potentially applied in the CCRP context and should be investigated further. In a second workshop in Lilongwe, Malawi, in September 2019 projects showed that they were motivated to use CAET not for supplying data to a large database but for characterizing, diagnosing, or tracking agroecology in their project work. This was because it seemed easy and applicable, open to use by nonexperts, including farmers working in FRNs, and would not be too burdensome. All projects agreed to design the assessments jointly, so they could learn from one another.

• 2. Planning for project assessments

The planning of projects’ assessments and sharing of ideas took place through a series of online discussions during October—December 2019. The discussions focused on the following:

• Purpose and objectives of assessments;

• Scale of assessments—farm, landscape, or community;

• Localization of the CAET tool;

• Participation in assessment design and data collection—whom to involve and how;

• Drafting of assessment plans.

One result of this was that projects decided they needed collective field experience to finalize plans and feel confidence in implementing the tool. Hence, we organized a field workshop.

• 3. Field workshop for piloting CAET

The 3-day workshop in Soroti, Uganda, in February 2020 was hosted by one of the CCRP projects (Cereals FRN). Twenty-seven participants from 11 CCRP projects in Kenya, Malawi, Tanzania, and Uganda attended. The CCRP projects all work closely with farmers and farmer communities and have been engaged with them using agroecology for several years. They vary in the extent to which they use systems approaches and envision system transformation, or work in problem-solving through agroecology practices. The participants at this workshop included researchers, leaders of farmer groups, and NGO workers who are part of project teams. Steps followed in the workshop were as follows:

• Reflection on draft CAET plans from individual projects including the broader framing and process, objectives for doing the assessment, conceptual framework and detailed assessment activities.

• Introduction to the Soroti farming-system context by the host project leader, outlining farming practices, main crops, other livelihood practices, markets, culture, and political situation.

• Development of protocols for piloting CAET in Soroti. Participants were divided into 3 groups focusing on farm, community, and landscape levels. The group focusing on farm level was made up of researchers, farmers, and NGOs, while the community and landscape groups were made up of community (youth, men, and women), village elders, NGOs, and researchers. Given that the aim of the field exercise was to build experience and confidence in the steps needed in implementing CAET, work in each group was focused on selected Elements not all 10 Elements. The farm and landscape groups first selected the Diversity Element, contextualized indicators and scores, then engaged local stakeholders (farmers, local government administrators, and some NGOs) in making the assessments. The community group planned a participatory process of selecting which Element to assess, localization of that Element, indicators and scores, and conducting the assessment.

• Fieldwork. The Farm level group first selected Element (Diversity), then carried out focus group discussions to adapt indicators and performance scores to systems in their area. This was followed by individual interviews with 5 farmers (including crop and livestock farmers, women, and men) who scored their own farms on the indicators (Figure 3). The Landscape group carried out a field focus group discussion in an upland watershed with 24 people (including farmers from up, mid, and lowlands along with representatives of a local NGO, local government, and the private sector). The Community group held a focus group discussion with 10–15 men and women representing a farmers’ group. They elected to work on Culture and Food Traditions, delving into the meaning of it, the relevance of the CAET indicators and scales and alternatives that would be more informative.

• Reflection on the process. To what extent was it effective? What improvements are needed? How did farmers and other local stakeholders react to participating in the assessment? Was it relevant to them?

Note that the workshop was planned in the expectation that it would give project teams the experience and confidence needed to adapt and pilot the CAET tool to their own context. However, the outcome was a more critical assessment of the applicability of CAET.

• 4. Continued online meetings and one-on-one email exchanges

Projects used the insight gained from the field exercise to revise their individual protocols in readiness for piloting and implementing assessments in their localities. Some projects shared their revisions and plans through email exchanges and online meetings. Relevant literature was also shared. Fieldwork was interrupted by the COVID-19 lockdowns. Online discussions continued by looking for ways in which assessment could inform envisioning of alternative futures and action planning. Participants interested in synthesizing lessons joined a small group to do that.

CCRP projects taking part in the process were Cereals FRN (Uganda), P’KWI FRN (Uganda), Sorghum (Kenya), Sorghum (Uganda), Finger Millet (Kenya and Uganda), FRN-NGO (Kenya), Drylands (Kenya), Agroecology Hub (Tanzania), Multi-Purpose Legumes (Kenya), and Best Bets (Malawi).

Reflecting on the experiences described above allowed us to draw out the following 8 lessons about the value and use of CAET.

• 1. Assessment can be inspired by CAET rather than be a direct application of CAET

Exploring the use of CAET provided inspiration for approaches to assessing the agroecological state of agricultural systems. The remaining lessons below point to reasons why CAET could not be taken as a ready-to-use, universal assessment tool. However, the inspirations we drew are important and summarized as follows.

• It is feasible to look across all (or most) agroecology (AE) elements in a rapid way using methods that are not too complex, burdensome, or knowledge intensive. Before looking at CAET, there was a feeling that the task was daunting. CAET showed that a practical approach is available.

• Simple, broad, rapid methods can be useful for project teams wishing to characterize the agroecological status of systems as part of informing projects where to intervene and whom to involve. They may also set baselines for future tracking of system change.

• CAET provides an opportunity for opening conversations on agroecological transformation by engaging stakeholders, extending discussion beyond the typical limits of agricultural development projects and problem-solving. It can be used to create awareness and initiate engagement with farmers and other relevant stakeholders about agroecological transformation.

As an example, the Best Bets project was inspired by CAET to gather information on knowledge, attitudes, and practices of farmers as well as generate the narratives that influence farmers’ behavior regarding 5 of the 10 agroecological elements (diversity, synergies, efficiency, co-creation and knowledge sharing, and human and social values).

• 2. Objectives of the assessment matter

Assessment or agroecological characterization is not an end in itself. Many projects found it hard to specify the larger goal to which an agroecological assessment would contribute and hence the objectives of the assessment, yet every aspect of the assessment design depends on that. After reflection, project teams gave the following overall objectives for their assessments:

• Integrating agroecology for sustainable food systems and improved landscape resilience (Drylands).

• Understanding how agroecology is being used and influencing farm productivity (Finger Millet).

• Increasing farmer understanding of agroecology so they can engage in a FRN more effectively (PKWI FRN).

• Increase farmer involvement in developing and using agroecological interventions (Multi-purpose Legumes).

• Broaden membership of an agroecology sharing platform (Agroecology Hub Tanzania).

• Strengthen stakeholders’ knowledge of agroecology for sustainable agricultural productivity (Sorghum Uganda).

The origins of these objectives were not traced, but all projects have distinct histories that have contributed to them. Each project also has its own vision of agroecology transitions and the way in which these different objectives contribute to them. Each of these broad objectives probably requires assessments that differ in their details. For example, increasing farmers’ understanding requires the assessment to be integrated with some farmer learning activities. This is not needed for the objective of understanding how agroecology is currently used.

In many cases, an assessment will actually have multiple objectives, with different parties being interested for different reasons. For example, a researcher may be interested in comparison across different contexts while farmers want to understand their own situations; an NGO wants to plan interventions and a donor is hoping to set priorities for investment. There will thus be trade-offs to make between these different objectives, and the result may be a decision that multiple assessments using different methods are needed.

None of the project teams set out with objectives that included contributing to internationally comparable data on the status of agroecology transitions, the core aim of TAPE. Thus, it is not surprising that we found CAET a good source of inspiration for assessments rather than a tool ready to use in each case. Incentives for contributing data to open sharing platforms are complex but often not in the interests of those doing the work in the field (de Lima et al., 2022).

• 3. Trade-off between breadth and depth of assessment

There is a trade-off between breadth and depth when doing an assessment. CAET is broad, covering 10 Elements of agroecology, each of which is itself broad. For example, under the element “Diversity” in CAET is an assessment of the diversity of crops, animals, trees, activities, products, and services of the farming system. However, there is little depth, as each is assessed on a simple 0–4 scale. In some cases, a single one of these scales covers multiple aspects that could be separated to add depth of understanding. For example, the “Crop diversity” indicator goes beyond diversity per se and refers also to local adaptation of crops and use of cropping arrangements such as polyculture. A deeper characterization of “Diversity” could include: type and relative abundance of the specific crops, diversity of other components of the farm or landscape; description of where/who has or maintains diversity; functions and values attributed to diversity; historical trajectories and current trends in diversity, and so on. Much the same is true for every one of the 10 elements and indicators therein.

The broad characterization of CAET has the virtue of “taking a wide-angle picture” of the system, and this is well adapted to FAO’s aims. However, it is less well adapted to the objectives of the teams reporting here. It could be appropriate for the Tanzania Agroecology Hub that wanted to compare different parts of the country, with an aim of raising discussion about needs and differences. However, for most of the teams, the broad characterization generates information they already know about their locations and contexts, and their specific objectives require more depth. Increasing the depth of assessment would require more resources since tools need developing and testing, data collection will be more intense and require data collectors to have higher levels of knowledge and skills, and interpreting data would be more challenging. In addition, we noted that:

• Breadth is likely to be most relevant when making assessments across a wide geographical area and objectives include describing the differences within the area.

• A broad assessment using standard tools is likely to be extractive since it will not be tuned to needs of stakeholders in each location.

• The limited scope of a broad characterization means that relevant details of interest to any stakeholders cannot be explored.

• A deeper study will need adaption to the specific local conditions, as well as the precise objectives of the exercise.

• A broad assessment could be the first step of an iterative process, informing more in-depth studies that follow the initial broad assessment. To some extent, this is the way CAET is designed to be followed by TAPE Step 2. However, there is still a breadth/depth trade-off to consider as TAPE Step 2 is also generic.

It is the nature of such trade-offs that decisions on design of an assessment will be based on the objectives as well as values, knowledge, and resources of those involved. The Best Bets project chose to assess just 5 of the 10 Elements so that they could go deeper into these 5. A cost of that decision is that interactions with aspects of other Elements will be missed, but it is interactions that make characterization at system level essential. It is tempting to place different system characteristics in some sort of hierarchical sequence and assess the “most important” first. It is common to characterize those factors directly related to productivity and livelihoods ahead of social and environmental factors. However, this contradicts a core principle of agroecology that production cannot be separated from social and environmental characteristics of systems, but both depends on them and influences them.

• 4. Grounding the tool is necessary

The generic indicators, scales, and scores of CAET need grounding in local reality before they can be used in assessment. This is briefly mentioned in Mottet et al. (2020). A consequence of this localizing or grounding of the tools is that data from different locations may only be comparable in a qualitative way. We describe examples of what this means and why it is necessary below.

• If the intention is to use data provided by farmers or other stakeholders, rather than from the assessor’s observations, then the concepts and indicators have to be translated into terms they use and understand. For example, the farmers in Soroti associated “food sufficiency” with having a large cassava garden and assessed a household’s food sufficiency in terms of cassava production.

• The scales used in CAET indicators need to be interpreted and specified in the local context before they can be used consistently. For example, the scale for “Tree diversity” goes from “0–No trees” to ”4—High number of trees (and/or other perennials) of different species integrated within the farmland.” This cannot be used until the meaning of “High” is determined. A well-treed farm in the humid tropics has a very different numbers of trees from a well-treed farm in a dry area. What is the basis for determining “High” in a specific context? It could be the highest seen on any farm in the area, an expert assessment of what is possible under current policies (e.g., on land tenure or timber trade), or an expert opinion of what it could be under more favorable policies. Alternatively, “High” could be defined by farmers’ own desires or targets.

• Parts of the agri-food system that are important locally but are not considered in TAPE may need to be added. For example, among the farmers in Soroti, fish and fishing are important. Fish contribute to dietary quality, and fishing interacts with agriculture in multiple ways. Hence, it would be natural to include fish and fishing in any assessment.

• What farmers consider to be “good” or “bad” may not exactly align with what the FAO tool implies. For example, farmers in Soroti considered some “new” foods as good for diets and nutrition (such as the recently introduced fruit trees jackfruit, avocado, and banana), yet the FAO tool seems to imply that “traditional food” is “good.” Similarly, farmers may not regard indigenous knowledge as highly as the tool implies. In reference to “Culture and food traditions,” the response about diet was associated with linkages to markets and connection to a cash economy. The time and labor involved in traditional cooking were seen as narrowing foods to those that are easy to handle and transport.

• Indicators that have a utility value are easier for farmers to score. For example, numbers and diversity of fruit trees were easier to score than trees in general. Indicators such as “trees for soil fertility and medicinal value” were difficult to score, perhaps because some trees are not associated with these services.

• Some of the indicators and scales used in CAET make assumptions that may not be relevant locally. For example, the indicator on “Management of Soil Fertility” uses a scale from “0–Synthetic fertilizers are used regularly on all crops and/or grasslands (or no fertilizers are used for lack of access, but no other management system is used)” to “4–No synthetic fertilizers are used, soil fertility is managed only through a variety of organic practices.” The assumption is that there is a gradient characterized by the points 0–4 on the scale. However, there can be a gradient of use of synthetic fertilizers and, orthogonal to this, a gradient from soil degradation to soil regeneration. Regeneration is not necessarily or functionally connected to using organic practices. A localized version may replace this one index by two, one assessing the use of synthetic fertilizers and the other assessing how effective organic management practices are.

• We also noted the need to understand whether to describe the current status of an indicator or the perceived trend over time. A status that still qualifies for a score of 4 could have been even better some time before and could be on a degradation trajectory.

• 5. Participatory adaptation of tools is useful and insightful

The grounding or local adaptation of CAET as an assessment tool can be done in a participatory way, allowing stakeholders, particularly farmers, to have a significant role and influence in the process. There are 2 reasons for doing this. First, participation is a value held by many who believe farmers and communities have a right to help determine those things done for them. Second, participation of stakeholders in design will increase the relevance and quality of the objectives and data collection. Both these reasons are captured by the agroecology principle of co-creation, itself part of the assessment in CAET.

Practical advantages of developing tools using participatory methods include bringing in local knowledge and perceptions, creating ownership of the process, enabling future self-assessment, and generating data that is relevant and useful to the context. However, the teams in our study noticed that, “There is a limit to participation.” For example, we engaged farmers in developing the scales for indicators, choosing the definitions and descriptions of scores. This was insightful, farmers were engaged, and it was effective. However, it was also very slow, and the time involved should be negotiated with participants. We could not see how it could practically be done for the large number of indicators in CAET. We also noted that participation in developing the tools of assessment can be confused with participation in assessment itself.

Some of the project teams had increasing farmers’ understanding of agroecology as one of the objectives of their planned assessments. It is likely that the step of codesign and adaptation of any tools could serve an important pedagogical function itself. However, that was not investigated during our field workshop because of its limited scope and the primary aim of giving project teams experience.

The Best Bets team engaged farmers in participatory tool design and had previously worked with those farmer groups. The social capital that already existed between farmer groups and researchers meant that farmers were willing and interested in engaging in the piloting activity despite there being no immediate benefit to them. However, the project found that this might lead to design of indices that tend to favor knowledge on topics researchers have previously discussed as farmers believe they already understand researchers’ interests.

As with any other participatory research and development, skilled field facilitation is essential, paying attention to language and voice, power, and interests.

• 6. Landscape-level assessment is feasible

CAET is envisaged as a tool for assessment at farm level, with TAPE Step 1 and Step 2 data collection described as farm and household surveys. However, other scales of assessment, such as landscape, community, and territory, can also be important. This is hinted at in the TAPE documentation with the statement “Step 1 consists of characterizing the level of transition to agroecology of agricultural systems (e.g., farms, households, communities/territories)” (FAO, 2019, p. 16). Assessment at landscape level is important and we realized that CAET could inspire landscape-scale assessment and investigated that with farmers. The need for landscape-scale assessment in locations of the project teams is, in part, driven by the small size of farms—often <1 ha. This means that many of the important interactions in the system occur across farms and the agroecological system must be characterized by taking them into account. In addition, parts of the landscape outside farm boundaries can play an important part in the farm and household system. These nonfarm elements of landscapes include wetlands, forests, bushland, and grazing areas.

The experience of both research teams and farmers in Soroti suggests that some of the agroecology elements and indicators in CAET only make sense when considered at landscape or community level. Most obvious were aspects of governance, along with synergies due to those interactions between farm and nonfarm areas and activities. However, farmers also found difficulty with concepts under Culture and Food Traditions. While the “Food security” indicator could be assessed at household level, “use of traditional knowledge and abilities in food preparation” required community groups. The stakeholders with an interest in landscape assessment will probably be broader than those for farms so should be included in an assessment.

We noted that along with other frameworks (Musumba et al., 2017; Wiget et al., 2020), ecological processes and environmental services which are important to farms and operate at landscape scale, such as pollination, pest occurrence, and soil and water movement, are not included in CAET. Also, experience with landscape-level assessments shows the importance of mapping as a tool to understand spatial layout and interactions in systems. Both these are examples of the need for a landscape-level characterization to go deep rather than remain at the broad level of CAET (Lesson 3).

• 7. Piloting is an important stage of deploying methods

The concept of “piloting” a research method or tool should be familiar to any field researcher. We found that it is helpful even if the research tool has previously been developed, tested, and used, such as CAET. In our case, careful piloting provided the following:

• It gives researchers a better understanding of CAET and useful insight into its application in ways that make sense in the context and to participating farmers and local stakeholders.

• It enables researchers to get prepared better by localizing the tool, making the assessments and interpreting results.

• Projects can scrutinize their own rationale for making the assessments and what might be the motivation for participation of local stakeholders.

• It showed that there is scope for getting broader answers than the tool is asking for and prepares researchers for dealing with such information. Farmers’ responses to questions do not always tally with what can be observed in the field. For example, when asked about trees in the landscape, farmers sometimes neglected to mention naturally occurring trees.

• Important insights not captured by CAET may be exposed. For example, we found that indigenous practices were being replaced due to a changing culture and cash economy. Farmers clearly identified “the need for cash” as the major threat to a healthy diet, whereas a project team had assumed that, in this case, cash contributed to having a healthy diet.

• Teams were able to make the data collection process more acceptable to farmers. For example, the draft tools from the Best Bets project had 11 pages of questions. Piloting showed this as unacceptably long and suggested ways of shortening it. That team also reported how piloting had revealed inconsistencies in answers when the same question was asked by different interviewers, prompting them to investigate why and fix it.

Some of these lessons about piloting were a surprise to us as we had expected researchers to be familiar with them from other contexts. Indeed, one researcher participant said, “Piloting was an eye-opener.” The message is that piloting is necessary, even if the tools have been used in other contexts and researchers are experienced in that context. This prompted us to write a short guide to piloting survey tools (Coe, 2020).

• 8. The AE element of “circular and solidarity economy” was least familiar

Project teams engaged in the exercise had all been exposed to and worked with agroecology ideas for a number of years. However, the concept of “circular economy” was often only vaguely understood and confusing. This limited teams’ ability to think critically about how it should best be included in an assessment and the relevance of the indicators used in CAET. This is an important gap as economic considerations are often among those given high-priority farmers.

The lessons described above point to the need for some further guidelines on making agroecological assessments or characterizing the agroecological state of systems since, while there are important ideas to be gleaned from TAPE, it is not a readymade approach or set of tools to use in every situation. The same will be true for any other “readymade” assessment tool that involves processes, indicators, and tools that have been defined without reference to the specific context and objectives. Hence, we propose principles (Box 1) and a process for operationalizing (Figure 4) them that should be broadly applicable.

We have headed this section “Emerging …” because these are proposals based on the lessons learnt and reported here. Both the principles and process for putting them into practice need validation and revision.

We formulated the principles in Box 1 based on the lessons reported, which provide the justification and explanations behind them, along with principles of agroecology. Principles 1–3 concern the overall style and purpose of the assessment and the way it connects to other activities or processes. Principles 4–6 guide choices about the content of the assessment. Principles 7–9 are about building on experience of others while adapting to the needs and capacities of those with an interest in the assessment. Principle 10 describes the iterative process needed to complete the design and implementation of the assessment.

We suggest a general stepwise process that has broad applicability, builds on the lessons, and can be seen as a means of operationalizing the principles. The process is much the same as for other research and data collection exercises but laying out the steps explicitly for an agroecological assessment may help users, particularly nonresearchers, to plan effective assessments. There are 6 steps in the process (Figure 4) and here we summarize each. Some of these steps are already included in TAPE to some extent, but we want to describe the process that needs to be followed for any agroecology assessment, whether it follows TAPE closely or not.

Agroecological assessment is not an end in itself but should be part of a larger process. TAPE is part of FAO’s effort to support comparable data collection across countries (FAO, 2019). An alternative is provided by Levard et al. (2019), which aims to provide evidence of effectiveness of development interventions. Biovision’s ACT tool (https://www.agroecology-pool.org/methodology/) aims to assess the agroecological content of a policy or project. All these objectives are different, so the tools developed to support them, while related, are also different. Hence the first step has to be to define the purpose of the proposed agroecological assessment within broader project, community, or intervention objectives. These objectives, and the context in which the assessment will take place, should define all other aspects of the method. The first principle (Box 1) is a reminder that the objectives should be negotiated between all those who have an interest in it. If the trade-offs involved, for example between local relevance and large-scale comparison, cannot be resolved, then different assessments will be needed to meet the needs of each. Principle 3 is a prompt to look for synergies between the objectives of the assessment and other objectives of those involved.

Among our teams, PKWI FRN sought to support farmers’ transition to a more resilient system based on agroecology principles. They work with farmer groups and decided that an agroecological assessment of their members’ own farms should be integrated with farmer training on agroecology principles, to help make these explicitly understood.

The assessment needs an explicit conceptual frame. TAPE uses FAO’s 10 Elements, but there are other frameworks that might be more suitable in some situations. Some are reviewed in table 1 of the TAPE document. Other approaches could be based on agroecology principles (Wezel et al., 2020) or practices, either of which could be used as the basis of assessment. Adopting an explicit framework reveals the logic of the assessment, allows one to determine whether it is complete, and makes connections with other work. Hence, this step starts with choosing or developing a framework for assessment, with the former strongly recommended.

In our work, most project teams took inspiration from TAPE and based their assessment on the 10 Elements, but other frames could be defended.

Definition of the scope includes the spatial and time scales, extent of coverage, and the sources of information. The objectives may call for work at national scale or in a single community or landscape. The objectives may require a single snapshot view of the current situation, include historical data, or require collection of longitudinal data through repeated assessments. The objectives may require coverage of all land in the study area or only that which falls in defined categories, such as smallholder farmland. TAPE describes data collection through key informants and farmer surveys. Other possible sources include other stakeholders, measurement by technicians, collective participatory assessment, self-assessment, or use of secondary data such as remotely sensed layers. Principles 4 to 6 are particularly relevant to this step.

In our case, projects’ areas of interest were chosen before they considered doing an agroecological assessment and those generally determined the locations for the planned work. In all cases, attention was to landscapes dominated by smallholder farms though most have landscape elements such as wetlands, bushland, and forest that influence and interact with the farming systems.

The details of the assessment include choosing indicators and measurement tools, who will collect data and how that will be managed, timing, budgets, and the many other aspects of practical field data collection. The principle of co-creation means these are jointly determined with stakeholders and that is likely to include negotiation. For example, while researchers might seek time-consuming detail, farmers may only be willing to take part if the activity does not take too long. Negotiation is needed to meet everyone’s interests. Principles 8 and 9 are applicable in this step.

This step puts Principle 10 into practice. The importance of testing or piloting data collection tools were discussed above. Testing needs to cover all aspects of the study, including, for example, data capture and processing. Testing or piloting is only valuable if there are then time and resources available for revising, and perhaps retesting tools and processes, before the main data collection activity starts.

Put the assessment plan into practice, monitor how well it is going, and make adjustments as needed.

Reflecting on and sharing results should be part of any research and data collection activity. At this stage, the assessment informs the larger process within which it sits, and the requirements of that process should determine what happens next.

It is important to be able to describe or characterize the agroecological state of food and agricultural systems. TAPE Step 1, or CAET, was designed to do that. Our experience using it in the context of projects working with smallholder farmers in ESAf shows that CAET provides inspiration and ideas for how such assessments can be done. It provides a tool that is broad, covering all 10 Elements of agroecology, and practical, allowing assessments to be done by nonexperts in a reasonably short time. However, its generality makes it poorly suited to the specific purposes and contexts of stakeholders interested in a particular instance of an assessment. The approach, methods, and tools of these assessments can take inspiration from TAPE but need to be developed for their specific task and tested to ensure they are fit for purpose. The process of localizing and contextualizing the design of the assessment will inevitably lead to trade-offs and compromises between the needs of different interest groups. We propose 10 principles and a generic process for navigating this assessment design process. These need validating and, if practitioners find them useful, we can expect them to evolve and be refined over time.

The authors would like to thank all colleagues, communities, and farmers engaged in the processes described in this paper for giving their time and sharing their ideas.

The work described in this paper was supported by the Collaborative Crop Research Program (CCRP) through grants from the McKnight Foundation.

The authors declare no competing interests.

Conception, review, approval of submitted manuscript: All authors.

Drafting: SN, RC.

How to cite this article: Namirembe, S, Mhango, W, Njoroge, R, Tchuwa, F, Wellard, K, Coe, R. 2022. Grounding a global tool—Principles and practice for agroecological assessments inspired by TAPE. Elementa: Science of the Anthropocene 10(1). DOI: https://doi.org/10.1525/elementa.2022.00022

Domain Editor-in-Chief: Alastair Iles, University of California Berkeley, Berkeley, CA, USA

Guest Editor: Nils McCune, Agroecology and Livelihoods Collaborative (ALC), Plant & Soil Science Department and Environmental Program, University of Vermont, Burlington, VT, USA

Knowledge Domain: Sustainability Transitions

Part of an Elementa Special Feature: Principles-based Approaches in Agroecology