Download Chapter

Chapter 3 Tools and approaches

Targeting

Photo:

Targeting is used to direct limited humanitarian resources to those who need them most. It is closely associated with response analysis, context analysis and assessments and profiling, which are discussed in other sections of this GPR.

This section defines urban targeting, identifies urban targeting challenges and discusses targeting methods. The section ends with the steps involved in urban targeting approaches. An example is given of using cash-based indicators for urban targeting.

3.7.1 Defining targeting and targeting challenges

Targeting can be defined as ‘the process by which individuals or groups are identified and selected for humanitarian assistance programmes, based on their needs and vulnerability. It is a way to focus limited resources on those within the population that would most benefit from support’.+G. Smith, L. Mohiddin and L. Phelps, Targeting in Urban Displacement Contexts: Guidance Note for Humanitarian Practitioners (London: IIED, 2017) (http://pubs.iied.org/10826IIED).

Targeting in urban areas is fraught with complexity. Vulnerability may well be hidden – those seemingly living well may be in chronic debt, or unable to sustain livelihoods (see the example from Syria in Box 3.10). Conversely, those who may appear to be vulnerable may well not be (a single-headed, unemployed household may be receiving remittances, for instance). Accurate baseline vulnerability data may not be available, which can exacerbate the challenge of distinguishing between those with humanitarian  needs  from those who are in chronic need (for example the chronically poor). In adhering to a strict interpretation of humanitarian need (in order to ‘focus limited resources on those within the population that would most benefit from support’, as noted in the definition above), one review+R. Patel et al., What Are the Practices to Identify and Prioritize Vulnerable Populations Affected by Urban Humanitarian Emergencies? A Systematic Review Protocol of Methods and Specific Tools Used to Target the Most At-need Individuals, Households and/or Communities in Urban Crises (Cambridge, MA: Humanitarian Evidence Program, Tufts University, 2016) (https://policy-practice.oxfam.org.uk/publications/what-are-the-practices-to-identify-and-prioritizevulnerable-populations-affect-605166). recommends that clear targeting objectives are essential: ‘Many high-risk and rapidly growing urban environments are characterized by widespread need and endemic problems and deficits of development in the absence of any acute crisis. Targeting vulnerable populations in such urban areas affected by crises cannot be an open invitation to permanent missions or demand that all pre-existing needs or deficits of development are met. Clear objectives and exit strategies must be employed’.+Ibid., p. 32.

A 2017 guidance note, Targeting in Urban Displacement Contexts,+Smith, Mohiddin and Phelps, Targeting in Urban Displacement Contexts. makes the  following useful observations concerning  urban  targeting.  First,  that  targeting  is  imperfect:  ‘all [targeting activities] will generate errors of inclusion and exclusion’. Second, that targeting requires trade-offs, for example in cost and coverage, and in the time required to undertake targeting and the quality of information gathered. Third, that it is important to remain pragmatic – ‘practitioners should select the mechanism that allows for the rationing and prioritisation of assistance to meet needs as quickly, fairly and transparently as possible’. Fourth, to use mixed methods – ‘given the scale of need and the limitations of each targeting mechanism, it is considered best practice to use more than one targeting mechanism in combination so as to reduce errors and further prioritise resources’.

Concerning displacement situations, one study on urban targeting+Ibid. identifies the following challenges:

  • The complex nature and heterogeneity of urban vulnerability – there is no easy distinction between ‘the vulnerable’ and those who are ‘not vulnerable’: most households may be considered vulnerable to varying degrees
  • Accurate data can be limited, biased or non-existent, for instance where a municipality has poor records or there have been recent population movements or rapid population growth. Records concerning informal settlements in particular may be missing or inaccurate.
  • Extensive chronic poverty – for example, many people live below the Sphere minimum standards, for instance regarding water availability (see Section 3.1 on standards) even without a crisis.
  • Targeting and assessments may be open to manipulation (see Section 2.2 on corruption).
  • Urban populations are often fluid, for example because of seasonal rural-to-urban migration, intra-urban migration and rapid population

3.7.2 Methods for targeting

A systematic review looking at best practice in urban assessments+Patel et al., What Are the Practices to Identify and Prioritize Vulnerable Populations Affected by Urban Humanitarian Emergencies? identified the following methods for targeting vulnerable urban populations affected by conflict, disasters and displacement.

Targeting by displacement or disaster-affected status, i.e. IDPs or refugees  versus  host populations. This approach can be problematic as it may create resentment, for instance among people not assisted. Evidence suggests that both host and refugee populations should be included given the prevalence of extensive chronic urban poverty, which often leads to similar vulnerabilities among both groups.+See for example Under Pressure: The Impact of the Syrian Refugee Crisis on Host Communities in Lebanon, World Vision, 2013 (https://assets.worldvision.org.uk/files/1413/7363/8729/Lebanon_Report_UNDER_PRESSURE_WV.pdf).

Using locally derived assessment tools, incorporating local voices, leading to context-specific indicators that may be more precise and accurate. Recognising the complexity and heterogeneity of urban areas, the review concludes that ‘Evidence suggests locally contextualized tools may represent best practice going forward’.+Patel et al., What Are the Practices to Identify and Prioritize Vulnerable Populations Affected by Urban Humanitarian Emergencies?, p. 31. It is noted, however, that this takes time and is resource-intensive. One example of such an approach+A. Wirtz et al. (2013) ‘Development of a Screening Tool to Identify Female Survivors of Gender-based Violence in a Humanitarian Setting: Qualitative Evidence from Research among Refugees in Ethiopia’, Conflict and Health 7(1), 2013, cited in Patel et al., What Are the Practices to Identify and Prioritize Vulnerable Populations Affected by Urban Humanitarian Emergencies? is from Ethiopia, where a screening tool was developed to rapidly identify female refugee survivors of GBV in order to target services better. The tool was developed through interviews and focus group discussions.

Targeting by demographic category, such  as  gender,  age,  ability  and  ethnicity.  While  there  are benefits of such an approach in terms of transparency and ease, entire groups who may be vulnerable can be overlooked. For example, ‘In Syria, unaccompanied Iraqi men living in Damascus were not considered for resettlement, despite lacking family or local networks’.+S. Haysom and S. Pavanello, Sanctuary in the City? Urban Displacement and Vulnerability in Damascus, HPG Working Paper (London: ODI, 2011). There is also a risk of being perceived as partisan. One study from Somalia reported that ‘Targeting is another extremely difficult exercise in Mogadishu. The Somali clan and sub-clan system is such that aid organisations run the risk of being seen as the enemy by one side if they provide aid to another, making needs-based targeting and the allocation of aid a risky endeavour’.+F. Grünewald, ‘Aid in a City at War: The Case of Mogadishu, Somalia’, Disasters 36(1), 2012.

Using pre-existing administrative data. The risk here is that data may be inaccurate, destroyed or biased. ACAPS notes that ‘The civil wars in Sierra Leone and Liberia, and the 2010 Haiti earthquake and 2008 Typhoon Fengshen in Manila, saw the destruction of physical data’.+ACAPS, Rapid Humanitarian Assessments in Urban Settings, Technical Brief, 2015 (www.alnap.org/resource/20125), p. 20.

Self-targeting, wherein people identify themselves as vulnerable, and, for example, may  seek help at an office. Drawing on examples from Syria and Somalia, the review found that ‘self-targeting is unlikely to reach the most vulnerable who may wish to remain hidden and proved expensive and difficult to maintain long term or to transition to local authorities’.+Patel et al., What Are the Practices to Identify and Prioritize Vulnerable Populations Affected by Urban Humanitarian Emergencies?. In DRC, UNHCR and partner agencies used self-identification as one targeting approach for girls engaging in survival sex.+This example is from Smith, Mohiddin and Phelps, Targeting in Urban Displacement Contexts, p. 46. Programme staff made contact with girls who were known to frequent specific areas. Staff encouraged girls to introduce them to their friends; support included protection, health and economic help.

Community-based targeting, wherein communities themselves identify  who  is  vulnerable.  Effective participatory assessments involve affected populations learning about needs and capacities alongside agencies. For example, following Typhoon Haiyan the NGO ActionAid undertook PRA exercises to enable communities to identify their needs and present them to NGOs.+Sanderson and Delica Willison, Philippines Typhoon Haiyan Response Review. Community-based targeting must be well-planned, with adequate training provided. Problems can arise when this is not the case. One programme by Oxfam and Concern in Nairobi, for example, relied on local community members – community health workers – to identify beneficiaries. A validation survey found substantial evidence of inclusion error, indicating that the health workers did not correctly identify the most vulnerable households. Problems included a preference for including friends or relatives and a lack of incentives to make the extra effort to uncover every vulnerable household in their area, as they did not receive compensation.+I. MacAuslan and M. Farhat, Review of Urban Food Security Targeting Methodology and Emergency Triggers. Final Report, Oxfam, 2013, cited in Smith, Mohiddin and Phelps, Targeting in Urban Displacement Contexts, p. 52.

Area-based targeting, where neighbourhoods  or  wider  administrative  areas  are  selected.  Criteria for this may vary, such as levels of damage following a disaster or conflict, pre- existing activities by aid agencies (which may want to continue working  in  the  same area), negotiations within coordination meetings to assign activities and level of legality (host governments may for instance not give permission for agencies to work in informal settlements). Area-based approaches are discussed further in Section 3.2.

For further discussion of these methods, see R. Patel et al., What Practices are Used to Identify and Prioritize Vulnerable Populations Affected by Urban Humanitarian Emergencies? A Systematic Review, Humanitarian Evidence Programme (Oxford: Oxfam GB, 2017) (https://fic.tufts.edu/wp-content/uploads/Urban-Humanitarian-Action-Systematic-Review.pdf)

Cash programming is discussed further in Section 3.3.

Box 3.12 Using cash-based indicators for urban targeting

CaLP’s Urban Toolkit on cash programming notes that good practice in urban targeting ‘requires clear definition of urban-specific vulnerability criteria, a selection process that prioritises the neediest families, and a verification process that can ensure that exclusion and inclusion errors are corrected transparently and quickly’.+Cross and Johnston, Cash Transfer Programming in Urban Emergencies.

Given the critical role that markets play in meeting the needs of urban residents, chronic and widespread poverty and its link with vulnerability and the growing use of cash and market-based approaches in humanitarian response, income- or poverty- related measures might be particularly useful targeting criteria. Various indicators may be used, each with their own pros and cons:+MacAuslan and Farhat, Review of Urban Food Security Targeting Methodology and Emergency Triggers.

• Livelihoods and income. Income is critical in urban areas but hard to measure directly, hence the use of proxies. Questions on type of employment are more likely to succeed and are often useful. Questions on debt are important but can be unreliable and sometimes ambiguous.

• Expenditure. This is highly relevant information but hard and time-consuming to collect. Proxies are easier.

• Assets and housing. This information can be relatively easy to gather and reliable because it can be verified by targeting teams, but may not be well correlated to poverty following an emergency (therefore reducing the usefulness of proxy means tests).

• Receipt of assistance from formal or informal sources. This information is usually highly relevant but can be difficult to interpret in contexts where informal sharing of resources is common.

Finally, regardless of poverty measures, the coping strategies used by individuals and households may be a better reflection of their vulnerability, but may not be captured by the measures listed above. A scale based on locally contextualised coping strategies may allow for a more holistic assessment of vulnerability and thus guide targeting.

 3.7.3 Steps in urban targeting

The guidance note Targeting in Urban Displacement Contexts+Smith, Mohiddin and Phelps, Targeting in Urban Displacement Contexts. provides principles for targeting criteria and decision-making tools. Figure 3.8, from the guidance note, illustrates five steps in urban targeting in relation to the project management cycle.

In summary, good targeting involves trade-offs, including time, affordability, quality of data and achieving something ‘good enough’ for informed programming. Transparency concerning ‘who is getting what and why’ is critical.