In the acute stage of a crisis, the priority is to determine the severity, magnitude and expected duration of the problem and the most appropriate response. Information is therefore needed on the extent to which different population groups can meet their immediate food, water, health, shelter and protection needs.
When a longer-term perspective is adopted, additional considerations need to be included in the design of an assessment. More detailed information and analysis are required for a broader nutritional and livelihood security assessment that can help the identification and design of appropriate interventions. This will demand a range of different sectoral expertise and will raise methodological and implementation issues related to the type of emergency situation.
Whatever the characteristics or stage of a crisis, local institutions have expertise, experience and information resources that can contribute to assessing and analysing the food and nutrition situation. These resources need to be sought out, utilized and strengthened during the assessment process.
When designing and planning a needs assessment in crisis situations, specific methodological and logistical aspects need to be considered. Decisions about which methods are most appropriate for a specific situation depend on:
Different agencies and organizations have their own methodologies or guidelines for assessments. Examples of these are: Emergency Needs Assessment (WFP), Household Economy Analysis (SCF-UK), Livelihoods Assessments (Oxfam), and Household Livelihood Security Assessments (Care International). No one method is inherently superior to another. Coordination and networking among the different agencies involved in the assessment with government information activities will help to ensure that methodologies are complementary and that information outputs are consistent and can be used to identify changes in the situation over time.
A number of external factors may also affect the assessment. These include the time available, physical accessibility, logistical and financial resources, the political context, and the security situation. Once these parameters have been discussed, it is necessary to decide what type of data to collect, the sampling approach, and how local groups and different stakeholders should participate.
Planning for a needs assessment requires the following: an analysis of who makes decisions about nutrition interventions; an understanding of what kind of information is needed, and by whom; and the identification of what kind of decisions will be made on the basis of that information, and when the information is needed. The framework developed in Part I can form a basis for discussing the themes to be investigated with potential information users. These themes can be grouped as follows:
how different types of households (e.g. by socio-economic, livelihood or geographical grouping) normally obtain access to food and achieve nutrition security in different seasons;
the risks to livelihoods that threaten household food and nutrition security in the immediate future and the longer term;
how households manage or cope in times of crisis;
people's own perceptions of their food and nutrition problems;
the political context and security situation;
the policy and institutional context and processes that affect household food security and nutrition.
A balance will need to be made among the requirements of different agencies and potential partners, taking into account the time and resources available and the principle of "optimal ignorance". Once the major themes have been identified, core or minimum data needs (quantitative and qualitative) can be specified. Brief 1 provides guidance on developing and preparing checklists for data collection.
When planning a needs assessment it is useful to prepare an analysis plan. This helps to ensure that only data that will be used are collected, and assists forward planning so that the skills, materials, computer equipment and software, etc. required for data analysis are in place. It can also reduce the time lag between collecting the data and making the information available.
Identification of geographical areas and livelihood groupings
In order to provide a basis for geographical targeting, the study area and population must be split up for assessment. This should be based on knowledge of the area or country, preliminary analysis of secondary data and existing vulnerability maps, and information from key informants. A decision will have to be made as to whether data collection is based on administrative boundaries (e.g. districts, provinces); agro-ecological zones, food economies or livelihood systems; or a combination of these. Different social or occupational groups within the different areas will then need to be investigated.
Existing knowledge and secondary information
Even in a crisis situation, knowledge and information related to agro-ecological conditions, natural resource utilization, food production systems, nutritional status and socioeconomic status already exist. Such material can be found in census data, survey reports, descriptive reports, etc. from government departments and national and international NGOs. Information may be available in printed or electronic form, or may have been lost or destroyed. If official records and documents are not available, the same information and complementary knowledge may be accessible through key informants. Using a "nutrition lens", a systematic review and assessment of secondary data related to the themes under investigation will help to refine the initial problem statement, narrow down information gaps and identify where there are additional data needs. As with any source or type of information, it is important to assess the quality and credibility of secondary data through verification and triangulation. Time invested at this stage will avoid the duplication of effort and reduce the cost and duration of primary data collection activities.
Quantitative and qualitative data
Both quantitative and qualitative data are required in emergency situations. This implies the need to combine methods and approaches according to: the information needs in a given situation; the time, financial resources and skills available; and the type and level of reliability desired. Two types of survey methods are normally used for assessing the nutrition and food security situation in emergency situations: a statistically representative anthropometric survey can be used to quantify the severity of a nutritional problem (see Brief 2); and rapid appraisal approaches can be used to obtain more in-depth information about the causes of food insecurity and malnutrition (see Brief 3). Both quantitative and qualitative assessments require rigorous and systematic data collection and verification in order to ensure the reliability of the data.
Sampling approaches
Once the general area for the assessment has been identified and decisions about the balance between quantitative and qualitative data have been made, random or purposive sampling should be used to select the sites to be surveyed. If the results of the needs assessment are going to be used to make inferences or generalizations about the total affected population, it is necessary to use a statistically representative sampling technique. In an emergency situation, two-stage cluster-sampling techniques have the advantages that it is not necessary to construct a sample frame (i.e. list all members of the population) and fewer sites need to be visited.
Purposive sampling is most often used in investigations that are collecting qualitative data through open-ended questions, for example, attitudes towards breastfeeding or practices related to diarrhoea management. RRAs/participatory learning and action (PLA) normally use purposive sampling techniques. If purposive sampling is used, local knowledge should be sought to help select the districts and villages to be visited by the assessment team. The locations visited should be typical and represent each of the areas experiencing problems. There are no rules for the size of a purposive sample, but an assessment of the credibility of purposive samples should consider the purpose of the study and the best sampling strategy to achieve that purpose. For some users, random sampling of even small samples will substantially increase the credibility of the results, for example, listing all the households with a certain characteristic (ex-cattle owners) in a particular area and then randomly selecting some for interview. This reduces suspicions regarding why a particular household is selected for interview, and increases the credibility of the results. However, this approach does not permit statistical generalizations.
Team composition
Time spent on selecting and training staff for a needs assessment will be an investment in future multisectoral programming actions and capacity building initiatives. Ideally, staff with local knowledge and a good understanding of the political and broader regional contexts should be borrowed from NGOs or future implementing partners. The team should be as small and unobtrusive as possible. Its composition depends on the particular focus and methodology of the assessment, and should include an appropriate combination of experienced nutritionists, agronomists, water and sanitation specialists, medical professionals and social scientists (e.g. community development workers, gender specialists). All team members should be aware of how their own and others' vested interests (based on ethnic, religious or socio-economic affiliation) may influence or benefit from the flow of aid and affect the team's interaction with local groups.
Accessibility to affected areas and population groups
Isolation, floods, seasonal conditions and the political and security situation may all influence an assessment team's ability to reach particular areas. In some situations, it may be possible to identify appropriate local professional staff who could be trained and equipped to carry out assessments.
Timing, duration and phasing
Normally, as soon as a problem or emerging crisis is "flagged" by information from an early warning system or other formal or anecdotal source, an initial rapid assessment is carried out so that decision-makers can be made aware of immediate needs. At this stage there is often a trade-off between precision and depth of data and the time needed to collect data. Even if there is not sufficient time for a thorough assessment, the initial rapid assessment is crucial in setting the scene for a more in-depth investigation as the second phase of a needs assessment. Food security and nutritional status have seasonal characteristics. The periods when physical access may be more difficult because of rains are also often the periods of greater food and nutrition stress. The period when an assessment is implemented also has implications on the way in which data and information are interpreted in the light of normal seasonal patterns.
Resource constraints
Gathering new information is costly, and budgets for assessment may be low. Budget constraints also affect whether people with appropriate skills can be recruited and whether adequate transport, fuel and equipment are available to implement the assessment. Assessment is an integral part of programme development; carefully managed joint assessments can reduce costs and contribute to multisectoral collaboration. Investment in data collection usually pays for itself in terms of facilitating good planning and the effective implementation of activities.
Local people usually have the best understanding of their own situation and the problems that they face. Communities may have been disrupted, but whether they are displaced or still in their homes, people will try to reorganize themselves to adapt to their new situation. Wherever possible, the role of relief agencies should be to strengthen and build local capacity and facilitate participation and local-level reorganization to lead to sustainable recovery.
Encouraging people to identify their own problems and constraints and helping them to plan locally feasible solutions are more likely to meet real needs. These will be more effective and likely to achieve outputs that can be continued with minimal external inputs. A participatory process can also foster the progressive reorientation from dependence on external assistance to local initiative and self-help, thus bridging the gap between relief and rehabilitation and leading to sustainable improvements in household food security and nutrition.
Participation is not a homogeneous entity - different levels and types of participation are required and feasible, depending on the objectives and context. While the principle or right of beneficiaries and civil society to participate in development and humanitarian interventions is recognized, there are obstacles to its practical implementation. It is often not clear who holds the rights (individuals or groups) and who the duties (national government and/or international bodies), and to what extent participation and the mechanisms through which it is implemented are legally binding. It is necessary to assess who the different stakeholders are (local people, CBOs, local interest groups, government technical departments, local administrations, potential implementation partners/donors) and their roles in respecting, protecting and fulfilling the right to food.
Woman preparing grains for milling
FAO/23822/D. Minkoh
In the acute stages of an emergency, the principle of participation has to be balanced with the need for a timely response. In the very first phase after a disaster, there may be little time to engage in participatory techniques, which can be time-consuming and assume that people have a certain degree of control over their own situation and future. In some emergency contexts it may even be dangerous to try to incorporate participatory practices, as this can lead to severe abuses of relief and the strengthening of powerful groups and individuals at the expense of the weak. However, in many contexts, participatory approaches can be used cautiously. Representative structures (reflecting the local context with respect to gender, ethnicity and socio-economic status) that acknowledge but are not dominated by local political power dealings need to be identified and strengthened, or established. The context of relationships among State structures, local structures and affected population groups also needs to be recognized. When planning for a needs assessment, local people and civil servants with technical skills and knowledge in health, nutrition, agriculture, communication and facilitation should be actively sought out, trained and included in the assessment team. During the design and planning of interventions, feedback and consultation with key local people and government departments will help to inform decisions about the mechanics of implementation, for example, the timing or location of service provision. The prioritization and implementation of interventions require a broader type of participation to ensure that there is an understanding of the reasons and criteria for selection and targeting. This kind of participation will also increase transparency and accountability in the allocation of assistance. Wider mobilization and sensitization are required for some interventions, for example, immunization campaigns. In comparison, participation in an intervention that provides small grants for particular livelihood activities will be more limited.
Identifying and understanding different perceptions
Women's opinions should not be overlooked as women play a key role in procuring and preparing food (including collecting fuelwood and water) and in caring practices. In war situations, women see the impacts of conflict every day, and they have to deal with some of its effects in terms of access to water resources, markets, etc.
Older people's perceptions and views are important, as they make key contributions to the livelihoods of their families and communities through generating cash income, facilitating access to work and contributing to child care. They may also be in a better position to understand changes in traditional care provision practices, and how drought, economic migration and war have affected family structure and livelihood systems. Children and youth can be particularly vulnerable in conflict situations, and appropriate techniques for understanding their needs and views on the future should be used in any assessment.
Box 21 shows how perceptions change according to age, gender and length of residence in Angola.
Box 21: Perceptions by age and gender among IDPs in Angola
"Displacement affected the different age groups and the sexes differently. Some young people were born in the places of refuge and some arrived there when very young so do not remember the things that happened. Older people do remember and their memories are negative: deaths of immediate family members, hunger, illness or rape. These recollections create difficulties in adapting to their places of refuge." "Men older than 18 have quite detailed memories ... women of this age group have more selective memories, and focus on the problems that their own families experienced ... old men and women are cautious about stating their opinions, but their memories are strongly marked by feelings of loss: cattle, goods, strength, values, power, dignity, and status. Their age means that they have little hope of reconstructing their lives before they die, and this forms part of their sense of loss...." |
Source: F. Andrade. 2001. A life of improvisation! - Displaced people in Malanje and Benguela. In P. Robson, 2001.
How can people participate?
Locally based institutions (government and non-governmental) will already have information and experience, and within the affected population there will be professionals who can be included as team members or key informants in an assessment.
Depending on the methods used for the household food security and nutrition assessment, the perceptions and opinions of different groups within the affected population can be gathered through liaison with existing representative bodies, CBOs and focus groups. Focus group discussions can be centred on different livelihood/occupational activities, such as fishers or market traders, but should also take into account age and gender and any existing ethnic differences.
Continuing an iterative process
The participation of potential stakeholders (interest-based groups, local institutions, government departments, implementation agencies, donors) will contribute to an iterative consultation process throughout the programme cycle. Dialogue within communities can contribute to the establishment or re-establishment of social capital in terms of bonds among individuals, trust, social fabric, common social norms and defined roles.
Potential implementation partners and donors should be involved in identifying information needs that are appropriate to their respective technical areas or sector interests, and should be kept informed of preliminary findings. If donor organizations have been engaged during the needs assessment process, they will have more confidence in its findings. At a later stage, the findings and recommendations should be discussed with these stakeholders as part of the negotiations regarding funding and implementation modalities. Broad participation can also contribute to consensus building and the responsible use of information for advocacy purposes.
Constraints to participation
A number of difficulties can arise when trying to increase local participation in needs assessments. These include raised hopes and expectations, bias in information, which can compromise the independence of an agency, and the additional time and resources needed for a more participatory approach. These are discussed in the following paragraphs.
Raising hopes and expectations. The history of previous emergency and development interventions will influence the hopes and expectations of affected populations. Clear oral and written information about the purpose and possible outcomes of the needs assessment should be made available to different local leadership structures (religious, administrative, political, CBOs), as well as to the population groups covered directly or indirectly by the assessment.
Identifying biases. Biases can emerge when only certain geographical, political, socioeconomic, ethnic, age and gender groupings are reached. Biases can result in severe abuses of relief, and can strengthen the powerful at the expense of the weak. Respondents are unlikely to underestimate their own needs for potential emergency assistance. The potential biases in different sources of information and the implications for power relationships should be taken into consideration in both the choice of methods and the selection of team members for needs assessments. Early feedback on preliminary assessment findings to different groups at different levels is not only good survey protocol, but can also act as an important verification tool to identify biased information. Using a participatory approach in an inappropriate context can endanger the impartiality and neutrality of a needs assessment.
Additional costs and time. The use of a livelihood approach and greater local participation is costly in terms of the time and resources required for needs assessment. At the beginning of an acute crisis, these may not be available on the side of either the affected population or the organizations wishing to undertake the assessment. A livelihood-based needs assessment also has implications in terms of the type and duration of interventions that will emerge. Interventions that protect and promote good nutrition through strengthening livelihoods need to start earlier, to have broader target groups and to last for longer than interventions that focus on immediate life-saving needs.
See Brief 3 for specific techniques that can be used to encourage participation in the assessment and design of interventions.
The technical briefs on developing checklists (Brief 1) and different survey and assessment methods (Briefs 2, 3, 4 and 5) provide practical guidance for implementing assessments. The data collected then need to be analysed to provide the following information:
the extent (numbers affected), severity (food and income deficit) and likely duration of food insecurity for different population groups and livelihood systems, in order to establish whether or not food aid is required;
the extent, severity, distribution and causes of acute malnutrition;
the short- and long-term risks associated with vulnerability to food and nutrition insecurity in different livelihood systems;
the groups (by physiological, socio-economic and residential status) that are most vulnerable to food and nutrition insecurity;
opportunities and constraints in each livelihood system (including the political and security situation, the policy environment and resource use and control by gender and age status, and institutional capacities);
local priorities and expectations.
In assessments using RRA/PLA methods, data analysis normally needs to be undertaken at the same time as data collection. This helps to create a picture of the situation and identify gaps and inconsistencies in information that can be resolved while fieldwork is in progress. Part I of this document presented livelihood, food security and nutrition concepts and interrelationships that can be explored during data analysis. Data can be structured, evaluated and synthesized by using matrices for each thematic area under investigation and comparing the situation for each livelihood and/or socio-economic group. Differences by age and gender should also be noted. Box 22 gives an example of what an assessment report could cover.
Box 22: Example of report content
If food aid is considered an appropriate response, include a clear statement of the precise nature, purpose and duration of any food aid intervention. This should be justified on the basis of the above data and analysis. |
Source: Sphere Project, 2004.
When it is difficult to obtain quantitative data, analysis can compare how the current crisis situation is different from a normal or baseline year, or the previous year. It is important to identify trends in food availability and access, changes in sources of food and income, or the stage in a sequence of coping/crisis strategies. In conjunction with the suggestions for different checklists in Brief 1, the following subsections provide a basis for analysing the data obtained.
Analysing the wider political, economic and institutional social context and its impact on food and nutrition security
How do macroeconomic conditions and policies affect food security and nutrition? Are there structural factors that underlie the current crisis? Will the current crisis affect the macroeconomic situation?
How does the current situation affect trade and food marketing activities (locally, nationally, regionally); for example, sanctions, blockades, bans on exports, import tariffs, restrictions on the movements of traders?
How might the political and social context affect or constrain the assessment process, or the selection of interventions and implementation mechanisms (e.g. targeting)?
What government departments, NGOs and CBOs are working in food and nutrition activities? How do their institutional capacities influence the effectiveness of actual or potential interventions?
Analysing the asset base of different livelihood groups:
How do households obtain and use assets (natural, economic, physical, social, institutional) in normal times, and how has this changed as a result of the crisis?
Are there differences by household, livelihood grouping and geographical area?
Analysing food security
Food availability:
If available, macro-level data such as food balance sheets can be used to determine overall national shortfalls, and may help to give a dimension of the overall food gap problem. However, such data do not reflect the needs of vulnerable households.
Meteorological data, in conjunction with land under cultivation and crop yield estimates for key crops, can provide information on production differences across geographical areas and between years.
The contribution that second plantings or dry season production in wetland areas make to food availability should be included.
The contribution of root crops to staple food production should be estimated. There are inherent methodological problems associated with calculating the production of crops with multi-year growing cycles, such as cassava. There can also be difficulties in estimating the production of crops that can be harvested on a continual basis, such as sweet potatoes.
In pastoralist-based economies, trends in livestock health and production and offtake (including distress sales) can be analysed.
If reliable population data are available, per capita kilocalorie availability values can be calculated for districts and socio-economic and livelihood groups for comparative purposes.
Seasonal food availability calendars can be used to identify seasonal shortfalls for specific foods, and the possible risks of specific nutrient deficiencies.
Data on household food stocks that are collected through either physical inspection or the household's own estimates can be used to approximate the likely duration of the hungry season. Both these methods are difficult in emergency situations, where food stocks may be vulnerable to theft and households have a vested interest in reporting low stocks.
The food availability situation should be compared with those of the previous year and of a normal year.
Analysing food access:
It is necessary to determine whether there has been a change in the amount of food obtained from the usual principle food sources, what the shortfall is, and whether households have been able to cover the deficit through other means of obtaining food.
If staple food production has been reduced through drought or the effects of conflict, there is a need to analyse whether households have been able to turn to other sources of food (purchase, wild foods, gifts from kin and neighbours), or to expand existing employment opportunities or take up new forms of employment to provide cash for food purchases.
Market prices (for cereals, livestock, essential non-food items) and casual wage rates can be used to calculate the purchasing power of a normal daily wage, exchange ratios and the local values of rations. Prices fluctuate seasonally if there is significant local agricultural production. Other fluctuations may occur in situations where parts of the population are dependent on food rations and there are ruptures in the food aid pipeline, and/or security conditions affect normal transport routes.
The proportion of expenditures spent on food can be compared across different groups. Interpretation should take into account that better-off groups may spend a high proportion of their expenditures on food in order to diversify their diets.
Do any types of households or individuals have difficulty obtaining physical access to markets?
Analysing food utilization:
Meal frequency of different household members (e.g. adults and children under five years).
Frequency of consumption of specific foods. This type of data can be analysed and interpreted by using simple scoring systems, and can indicate the presence of micronutrient-rich foods, the absence of which could indicate micronutrient deficiencies.
The availability of and distance from water and fuel sources can be related to their effects on food preparation and women's time.
Availability and utilization of storage and processing equipment, and its suitability if new foods have been introduced through the ration or are being cultivated.
Analysing the social and caring environment:
Availability and use of essential non-food items: soap, blankets, cooking utensils, clothing and footwear. Impact on individual and household dignity and well-being.
Availability and use of latrines can be assessed in relation to the incidence of diarrhoea, intestinal worms, etc.
Changes in caring practices such as breastfeeding, in the composition of complementary foods and in family and food hygiene practices can signal threats to nutrition security.
Levels of basic literacy and nutritional knowledge.
A high proportion of orphan-headed households or of elderly/disabled people living without family support can indicate a breakdown in social networks.
Types of households and individuals (by socio-economic and livelihood group, gender and age) resorting to coping and crisis strategies. How does the use of these strategies affect short- and long-term food, nutrition and livelihood security? A growing proportion of household members engaged in activities that are likely to lead to deteriorating health status indicates increasing vulnerability.
Interpretation of anthropometric data
Anthropometric data can determine the existence and extent of poor nutrition. It can also predict how the problem will evolve if no action is taken, as anthropometrics provides strong predicators of morbidity and mortality. However, there are no universally applicable rates to define the severity of a situation, and so trends in anthropometric indicators are the most useful way of interpreting them. In crisis situations, rates of malnutrition and mortality may not increase until the problem is very severe, and only certain indicators of nutritional status are sensitive to short-term changes in food intake. Wherever possible, data should be disaggregated by age and social or occupational groups. See Brief 2 for further guidance on the interpretation and use of anthropometric data.
The use of contextual data: morbidity and mortality data. In order to determine the causes of malnutrition for different socio-economic groups, areas or age groups, trends in nutritional status should be compared with:
morbidity and mortality patterns and rates (particularly recent outbreaks of measles and diarrhoeal disease and estimated prevalence of HIV/AIDS and TB);
the proportion of the population with access to water and sanitation;
immunization coverage and access to health services;
breastfeeding and infant feeding practices;
the food security situation;
recent population movements.
Analysing who is vulnerable
A picture of what types of households are more vulnerable to food and nutrition insecurity can be developed by using data disaggregated by livelihood system and socio-economic status. An examination of intra-household food allocation (by age and gender) and caring practices will help identify individuals who are at greater risk of malnutrition.
Chronically resource-poor households. These households have few possessions or assets and less resilient livelihood strategies, and they are more vulnerable to food insecurity in crisis situations. Not all poor households are at risk of malnutrition. Some may have a combination of good sanitation and environmental conditions, be able to manage their limited resources effectively, or have cultural practices that help to protect members who are at increased risk because of their physiological status (children under five years, pregnant or lactating women, and the chronically sick).
Some situations may have led to the widespread loss, theft or destruction of household productive and liquid assets, so that previously rich households are suddenly as vulnerable to food insecurity as poor households. However, richer households may also have less obviously visible assets such as skills, social standing and kinship networks that enable them to recover more quickly.
Social factors. Conflict situations and the HIV/AIDS epidemic can result in households and individuals (e.g. orphans, single male youths, elderly people) losing their kinship networks. While many cultures try to ensure that these individuals are cared for within the extended family network, this too may have broken down or, in the case of orphans, vulnerable individuals may be disinherited of land and other assets.
In conflicts that have an ethnic or religious dimension, certain households may be discriminated against or marginalized because they belong to or have married into the wrong group.
Vulnerability at different stages of the life cycle. Nutrient requirements differ at different periods in the life cycle (pregnancy, lactation, infancy, childhood, old age), and some deficiency diseases are more prevalent in particular population groups.
The disabled, mentally ill and chronically sick. Some people may have physical, sensory or mental disabilities that make them more vulnerable than others to the safety and health hazards of disasters. They can have specific needs in emergency situations; for example, shelter allocation close to service facilities, aids to compensate for their disabilities.
Gender considerations. In crises, particularly those involving conflict, a large proportion of households are without men. This may be because male household members have migrated in search of employment, have joined warring factions, or are missing or killed. Women therefore often have to cope alone with increased responsibilities for both productive activities and household chores, which limits their possibilities for involvement in public works. The physical safety of women is also often threatened by the presence of land mines, increased lawlessness and instability. Women are often deliberate targets of rape and violence in war situations, or they need to negotiate entitlements in exchange for sexual favours. Girls also shoulder increased physical and psychological burdens, through taking on additional household duties, missing out on education and schooling opportunities, and having to accept involuntary early marriage.
The most vulnerable members of society are the physiologically vulnerable individuals in socio-economically deprived households. In conflict situations, particular ethnic groups may be at risk of additional harm. While the physiologically vulnerable can be readily identified, and specific interventions targeted to meet their particular needs, the identification of households at risk for socio-economic reasons is more difficult. No single criterion can automatically be applied, and measures to address vulnerability will depend on each particular situation. Identifying or targeting particular ethnic groups may also increase their vulnerability.
Interpreting risks to food and nutrition security
It is also necessary to analyse the data in terms of threats to long-term food and nutrition security and what the opportunities are for strengthening the capacity to deal with these threats. The framework presented in Part I outlines potential risks; these should be specified for each livelihood system, along with how a shock would affect the livelihood activities of different socio-economic groups and whether this would increase the vulnerability to food and nutrition insecurity and/or poverty. For each livelihood system it is also important to analyse how positive attributes can be supported and strengthened and how leverage points (e.g. existing programmes and institutional capacity) can be used to mitigate and/or reduce future vulnerability to food insecurity. A strengths, weaknesses, opportunities and threats (SWOT) analysis is a useful tool for organizing and analysing information in terms of understanding the strengths and weaknesses for each livelihood group and the opportunities and threats coming from outside the group. By carrying out a SWOT analysis for each livelihood group, common factors can be identified that can lead to the development of a "win-win" intervention strategy, or potential tensions can be identified where conflict management or consensus building is required. See Brief 4 for guidelines on how to carry out a SWOT analysis.
Identifying priority problems
The above analysis based on a nutrition and livelihood framework will provide an understanding of the key problems and characteristics (asset base, demographics, nutritional well-being, coping strategies) for different population groups.
Different tools are available to assist the prioritization of key problems. A "problem tree" can be developed for each livelihood group, showing the cause and effect linkages and the interrelationships among problems (see Brief 5 for guidance on developing and using a problem tree). For each problem, it is necessary to specify which type of household (e.g. single-headed, with small cultivated areas, IDPs, returnees) or individual (e.g. elderly, orphans, physically disabled, widows) is most affected by these problems. It is also necessary to consider how gender and age differences and relations influence the impact of these problems. Such analysis will show that diverse actions are required, depending on how a shock affects different aspects of livelihoods and on the coping ability of the population group(s). It is then possible to plan and select the most appropriate interventions.
Figure 2 provides an example of a problem tree for the food and nutrition problems identified in Lunsar town, northern Sierra Leone. The country had seen a protracted and vicious civil war for more than five years. Peace had been brokered, and democratic elections had taken place in March 1996. The assessment took place in July 1996.
Figure 2: Example of problem tree
Source: Kate Godden, Children's Aid Direct, Sierra Leone. Personal communication.
Casley, J. & Lury, D.A. 1981. Data collection in developing countries. Oxford, UK, Clarendon Press.
FAO. 1993. Guidelines for participatory nutrition projects. Rome.
Patton, M.Q. 1990. Qualitative evaluation and research methods. Second edition. Sage Publications.
Pretty, J.N., Guijt, I., Thompson, J. & Scoones, I. 1995. Participatory learning and action. A trainer's guide. International Institute for Environment and Development (IIED) Participatory Methodology Series. London, IIED.
Scrimshaw, S.C.M. & Hurtado, E. 1987. Rapid assessment procedures for nutrition and primary health care. Anthropological approaches to improving programme effectiveness. UCLA Latin America Center. Reference Series Volume 11. Tokyo, United Nations University (UNU), UNICEF.
Seaman, J., Clarke, P., Boudreau, T. & Holt, J. 2000. The household economy approach. A resource manual for practitioners. Save the Children Development Manual No. 5. SCF-UK.
Sphere Project. 2004. Sphere Handbook. Revised edition. Humanitarian Charter and Minimum Standards in Disaster Response. Geneva. Available at: www.sphereproject.org/handbook/index.htm.
Young, H. 1998. Food security assessment in emergencies: theory and practice of a livelihoods approach. In Young et al., 2001.
Young, H., Jaspars, S., Brown., R., Frize, J. & Khogali, H. 2001. Food security assessments in emergencies: a livelihoods approach. Humanitarian Practice Network (HPN), Overseas Development Institute (ODI) Paper No. 36. London, ODI.
Technical briefs
Brief 1. Development and testing of checklists
Brief 2. Anthropometric surveys
Brief 3. Rapid appraisal techniques
Brief 4. Developing a SWOT analysis
Brief 5. Developing and using a problem tree
Meals being distributed at a feeding centre
FAO/23435/M. Bleich