The effects of social assistance
interventions on gender, familial and
household relations among refugees
and displaced populations: a review
of the literature on interventions in
Syria, Iraq, Jordan and Lebanon
Pauline Oosterhoff and Raudah M. Yunus
BASIC Research
March 2022
Implemented by
Better Assistance in Crises (BASIC) Research (funded by UKAid) aims to inform policy and programming on how to help
poor and vulnerable people cope better with crises and meet their basic needs through more effective social assistance.
All costs related to BASIC Research are covered by the UK Foreign, Commonwealth and Development Office.
Summary
This literature review aims to explore the evidence on the effects of social assistance on gender, familial, and
household relations and power dynamics among refugees and (internally) displaced populations in Syria,
Iraq, Jordan, and Lebanon. It examines the findings from an intersectional gender perspective allowing the
authors to build on the knowledge of ‘what works’ in interventions in general and hopefully improve gender
equality and social inclusion. Out of 1,564 papers initially identified and screened, 22 were included in the
final stage.
A question that emerged as the papers were analysed was whether the arduous work of targeting individuals
was efficient or necessary, given that the available evidence suggests that beneficiaries generally tend to
share their stipend with other family members for the collective good. Most studies tended to conflate gender
with women and girls – making distinctions between widowed, married, unmarried and divorced women – but
ignoring other dimensions such as class, health status, religion, ethnicity, education, prior work experience,
political affiliation, and civil participation. Many programmes and research fail to disaggregate data.
Social assistance programmes focus on individuals and households, with little attention to the wider context
and overall conflict. Most studies paid negligible attention to familial infrastructures and strategies for
sustainable interventions. Access to, and use of, cash transfers are part of broader familial strategies to
mobilise or increase resources including, for example, (male) migration in pursuit of remittances, or (female)
dependency on ‘community charity’. Short-term cash transfers can, in some circumstances, disrupt
individuals’ and families’ access to more sustainable income or ‘charity’. Thus, important questions are raised
about the purpose of social assistance: does it aim to preserve or transform families through targeting?
About the authors
Pauline Oosterhoff (MA, MPH, PhD) is a researcher and visual artist. She has been a Research Fellow at
the Institute of Development Studies (IDS) since 2014. During this time, she has acted as a team leader,
director, and researcher in over 30 large-scale and boutique programmes, both in South and Southeast Asia
and at a global level. Her research at IDS has focused on sexual and reproductive health and rights, modernday slavery and labour migration, and outbreaks and humanitarian responses examining gaps between
policies, plans and local realities. She believes that consideration of gender, culture and intersectionality has
invaluable ethical and practical benefits for academics, development, and business practitioners. She is
interested in using her technical competences and creativity to innovate research and uptake, pushing
boundaries of participatory, quantitative, qualitative, and visual and performative approaches and techniques.
Raudah M. Yunus (MBBCH, MPH, DrPH) is a public health specialist, researcher, writer, and social
activist. She teaches family health, population dynamics, global health and research methodology at the
Faculty of Medicine, Universiti Teknologi MARA (Malaysia). Raudah actively conducts research in the
field of social gerontology (focusing on elder abuse and neglect), refugee health, and human trafficking.
She was consultant to several research projects on modern slavery and child labour with IDS. In 2018,
she co-founded READ Malaysia (Relief, Education & Development), an NGO that facilitates access of
refugee children to primary education. Her latest projects include developing and testing an online
module on elder abuse among emergency health care workers, designing and testing a mobile app for
dementia caregivers, and setting up a physiotherapy centre for refugees in Klang Valley (Malaysia) as
part of a bigger effort to include this group into the national health agenda.
2
Contents
1. Introduction
4
2. Why is this study needed?
6
3. Methodology
7
3.1 Definition and classification
7
3.2 Inclusion and exclusion criteria
9
3.3 Study limitations
4. Results
10
11
4.1 Overall study quality
14
4.2 Limitations of evidence base
15
4.3 Availability of disaggregated data
16
5. Discussion
17
6. Conclusion
20
7. Recommendations and Implications for future research:
21
References
23
Annex 1: Interventions and outcomes
27
Annex 2: Search strategy
29
Annex 3: Critical appraisal
30
Annex 4: Data extraction and coding tool
31
Annex 5: Overall results/findings
32
3
1. Introduction
Most social protection and social assistance programmes have taken place in stable development contexts,
leaving researchers working in (protracted) conflicts or humanitarian contexts to depend on evidence derived
from the former (de Hoop et al. 2018). Globally, social protection coverage is unequal. In the pre-Covid-19
era, approximately 55 per cent of the world’s population were not covered by any form of social protection
(UNICEF 2020). Given this gap, it is not surprising that social assistance is a growing and rapidly evolving
intervention area in (protracted) emergencies, being targeted at individuals and households including a
growing number of refugees and displaced and stateless persons.
Choices about accessing various social assistance packages are made within political collective contexts,
and nowhere is this more apparent for an individual than within collective familial, household and kinship
units. Registration procedures such as standing in line in public to receive food, cash, access to fee waivers
for schools or free medications can be perceived as shameful for a family, resulting in some members being
allowed or being sent away, while others stay at home (Bila and Egrot 2009). The nature of the social
assistance provided also shapes who in a family is expected to spend time accessing this support. (Patel and
Hochfeld 2011; Zetter and Ruaudel 2018). Some social assistance, such as food or child health, for example,
might be seen to be associated with women’s roles. When refugees do not have the right to work or to
access labour markets in host countries, or when the refugee registration process violates pre-conflict gender
norms about mobility, it affects who in a family will be eligible to access the benefits associated with refugee
status (Zetter and Ruaudel 2018; Harvey, Garwood and El-Masri 2013). Individuals’ choices around
accessing various types of social assistance, as well as control over and use of this social assistance, are
thus made within broader collective familial settings and power relations. These familial and household
politics intersect with the ways systems are organised or with the influence of external actors. Whether and
how interactions between familial, household and system politics result in inclusive or equitable usage of
benefits is hard to predict.
Although there is overlap between households and families, they are not the same. Familial and kinship ties
based on consanguinity (biological) or affinity (i.e. marriage, friendship or other close relations) can reach well
beyond the geographically and administratively bound household or nuclear family unit (Daly et al. 2015; Otto
and Keller 2014). Families and households are already diverse in stable contexts (Cornwall and Jolly 2009;
Oosterhoff, Waldman and Olerenshaw 2014), adapting in composition to meet different needs across a life
course. Displaced and refugee families are no exception. The world’s urban refugee population is not only
increasing but also changing in composition. When forced migration occurs because of climate change or a
(protracted) conflict, the composition, decision-making dynamics and location of households and families as
well as their livelihood options change (McNatt et al. 2018; Akesson and Sousa 2020; Farhat et al. 2018;
Wood et al. 2021).
Social assistance packages that target an individual to include the needs and/or recognise the rights of
vulnerable groups such as ‘person with disability’, ‘female-headed household’ or ‘victim of gender-based
violence’ can intentionally or unintentionally alter power relations and transform the compositions of these
units. In stable ‘welfare’ states, women have historically and typically gained welfare entitlements as a
dependent within a family or a household as a wife or a mother (Lewis 1992). While inclusion and exclusion
criteria might be technical, the allocation of resources is also political. For instance, social assistance can be
provided to preserve family and gender relations or to transform them (Orloff 1996).
When formal governmental and state structures are in disarray in times of crisis, the household and family
end up being the foundation that supports, sustains, endorses or hinders, disapproves and discourages
individuals to act or wait. Intermediaries and social connections outside the household and family can help to
diversify income in a crisis – possibly in addition to cash transfers – but they can also cause tensions or
competition between and within households (Stevens 2016). For example, when refugees are not allowed to
be formally employed, to reduce tensions with local workers, they may seek employment relationships with
intermediaries that can be highly exploitative (Jones and Ksaifi 2016).
4
For policymakers and practitioners providing social assistance, often under very difficult physical
circumstances combined with time pressure, it is hard to assess who needs help or what the (unintended)
effects of interventions might be. Guidance to support humanitarian staff to design, implement and evaluate
gender, age, disability and other intersectional population characteristics is available in many guidelines,
policies and standard handbooks for the sector (Sphere 2018). However, these guidelines tend to be general
and therefore should be contextualised to fit local lived realities, in which social norms and power relations
change even faster due to the need to adjust to a new context.
A crisis can speed up, delay or (re)enforce the norms and material conditions in which choices about access
to available packages continue to be made in conjunction with other livelihood choices of the collective unit.
Decisions on, for example, (1) whether (or not) to and who can migrate for work to generate (international)
remittances, (2) who should stay to work in a public works programme or (3) who should apply for cash
assistance to care for an elderly person depend on what the most powerful figure(s) in the family can accept
and live with. These familial decisions depend also on policy design and shape and affect the implementation
of these policies in practice. Within the family, as well as in wider geographic, identity-based or professional
communities, gender1 and intersectionality 2 shape decisions to seek, consume or provide social assistance.
Although the importance of gender and intersectional characteristics for relevant, effective and efficient
humanitarian programming is widely recognised, little is known about the intended and unintended effects of
social assistance interventions in a crisis. In (protracted) conflicts, the discrepancy between needs and
resources is often acute, thus requiring prioritisation of assistance based on need alone. Groups that are
often deemed to have special needs include pregnant women, mothers of children younger than five years
old, children or people with disabilities. But whether these groups consider themselves as vulnerable or have
control over benefits are part of a different discussion.
While social assistance programmes may be global in terms of universal human rights, the implementation of
universal human rights and responses are shaped by and received in diverse local contexts – a diversity that
is invisible at aggregated programme levels. In local contexts, social support opportunities and structures
provided by religious charities, for instance, might be unknown to global, regional or national social
assistance programmes. Social assistance is provided to individuals to reduce inequities within households,
such as to enable them to care for vulnerable people or to empower or protect them from child marriage
(Malhotra and Elnakib 2021), polygamy or intimate partner violence (Heath, Hidrobo and Roy 2020).
To maximise the efficiency of social assistance interventions, social (dis)advantages must be assessed.
Nevertheless, assessments do not always accurately reflect realities on the ground and sometimes overlook
them. For example, what women consider to be the criteria for self-identification as a female-headed
household can be the opposite of ‘needy’. They can also see themselves as the head, as the leader (Levine
and Mosel 2021). Women who live alone because their husbands migrated may self-identify as being in
female-headed households, but that does not mean they are vulnerable as they might have access to
remittances, despite their limited mobility. Similarly, some pregnant women who receive a small stipend are
reported to share their benefits with the whole family, possibly reducing their share. On the other hand, older
adults tend to be forgotten; families reported that their elderly members often felt like a burden as they
received no benefits and had to rely on others (Oosterhoff 2016).
This paper aims to examine ‘what works’ through a gender and inclusion lens. We explored existing evidence
of the effects of social assistance on gender, familial and household relations among refugees and displaced
populations in Syria, Iraq, Jordan and Lebanon. These four countries were of particular interest due to old
and recent internal and international conflicts in the region that have caused massive displacement and
forced movements among Palestinians, Kurds, Yezidis, Syrians, Iraqis and other ethnic groups. The world’s
largest group of refugees today is from the Syrian Arab Republic, estimated at 6.6 million people, with
Lebanon, Jordan and Iraq hosting a huge number of them (Concern Worldwide 2020). We attempted to find
a middle ground between the need to have a localised and context-specific understanding, and the ability to
1
Gender: the socially constructed roles, behaviours, expressions and identities of girls, women, boys, men and
gender-diverse persons.
2
Intersectionality: other aspects of identity, such as age, ethnicity/nationality, ability and education that impact the
experience of gender and inequality, and access to and control of resources, power and knowledge.
5
compare available evidence of the effects of social assistance interventions on gender, familial and
household relations among refugees and displaced populations in Syria, Iraq, Jordan and Lebanon.
More specifically, this paper attempts to answer the following questions:
1. What is known about the effects of social assistance on gender, familial and household relations among
refugees and displaced populations in Syria, Iraq, Jordan and Lebanon? What is the evidence gap?
2. Have interventions or evaluations taken intersectionality into account?
3. How do these interventions affect familial, household-level and intrahousehold-level power dynamics?
Our study context/scope is protracted conflict and displacement. We are not directly focusing on recurrent
climate crisis. However, we recognise that there are arguments for including climate change in a broader and
ongoing discussion on the causes of the Syrian civil war (Selby et al. 2017; De Châtel 2014). This is part of a
wider debate that falls outside the scope of this paper. The limited body of evidence on ‘what works’ to deliver
effective and efficient social assistance is one of the rationales for Better Assistance in Crises (BASIC)
Research. Gathering, appraising and examining the latest and most relevant evidence will inform social
protection research and programming, and the metrics to judge the effectiveness of interventions.
2. Why is this study needed?
Gender, familial and intrahousehold relations and power dynamics play a crucial role in determining how
social assistance or social protection benefits (or deprives) individuals. This is because inequity can exist
within a household and living in the same house does not mean all individuals have similar living conditions
or access to resources. Studies have shown that risks, vulnerabilities and capacities are highly gendered
over the life course, and that gender is a key factor that can lead to disadvantage and oppression. Apart from
gender, different forms of discrimination and advantage intersect with other social dimensions and
characteristics such as age, class, race, religion, sexual orientation and special needs. Interpretations of
these characteristics are historically contingent and context specific.
This indicates that the ability to benefit from social protection programmes is influenced by one’s status and
bargaining power in a family or household. Women, children, elderly people and people living with disability
can be unintentionally neglected or may face additional challenges to have their needs recognised and
prioritised. Similarly, an intervention that aims at a specific outcome can cause (negative) unintended effects
when not properly designed, or when social relations and power dynamics governing the lives of recipients
are poorly understood.
The current evidence base on social protection and social assistance, and their impact on gender-related
outcomes and intrahousehold dynamics, has been largely derived from stable development contexts, mainly
in African, Latin American and South Asian countries. In conflict settings in the Middle East and North Africa
(MENA), research on social assistance is relatively young, though steadily growing. Little is known about how
social assistance programmes affect gender, familial and intrahousehold relations and power dynamics
among refugees and displaced populations in countries such as Syria, Lebanon, Jordan and Iraq.
While existing evidence from pre-conflict and development settings can provide an insight, it may not
necessarily be applicable to (protracted) fragile or conflict settings in the Middle East for reasons such as: (1)
geographical and socio-cultural differences; (2) rapidly changing conditions and lack of proper governing
structures; and (3) discrepancies in aims, priorities, design, family compositions and implementation of
existing social protection programmes.
All these complexities point to the importance of taking intersectionality into account in the design and
implementation of social assistance, so that interventions recognise the various social relations of an
individual that affect their vulnerability and resilience to different risks (Newton 2016).
6
3. Methodology
To answer the research questions, a rapid review was conducted according to PRISMA guidelines,3 but
without adhering to the strict requirements of a typical systematic review (Moher et al. 2009). Rapid review is
a type of ‘knowledge synthesis in which components of the systematic review process are simplified or
omitted to produce information in a timely manner’ (Khangura et al. 2012). This method was used mainly to
overcome the need for enormous resources in terms of budget, human resources and time – which may not
suit the needs of decision makers or policymakers – that are commonly encountered in systematic reviews.
There are various approaches to rapid reviews and prior studies showed comparable findings between
systematic reviews and rapid reviews (Cameron et al. 2007; Best et al. 1997; Corabian and Harstall 2002;
Tricco et al. 2015).
The first objective of the rapid review was to determine from the existing body of evidence what is known
about the impact of social assistance programmes on gender, familial and household relations in the four
countries of interest. The second objective was to find out if interventions and evaluations have taken into
account intersectionality perspectives. The third objective was to study how these interventions affect familial,
household-level and intrahousehold-level power dynamics. This was done through a wide search of online
databases and grey literature, followed by multiple steps of screening, selecting relevant papers and
extracting data.
We aimed to examine the findings from an intersectional gender perspective. This would allow us to build on
the knowledge of ‘what works’ in interventions in general and hopefully improve gender equality and social
inclusion. The questions we applied included:
1. Have data been disaggregated by sex and other basic intersectional characteristics such as age, ability,
race and sexual orientation?
2. Does analysis show gender awareness and recognise that gender is a rationale, and is it used in the
methodology?
3. Is there a rigorous analysis of gender relations and how these intersect with other relevant aspects of
identities, and is there an explicit analysis of root causes?
4. Do interventions recognise and aim for structural changes in power relations, norms and policies?
These multiple layers of analysis will help develop research plans that can be taken forward by the BASIC
Research Programme.
3.1 Definition and classification
The definition and taxonomy of social protection in this paper are based on the framework used by the
BASIC Research Programme and proposed by O’Brien et al. (2018) and Carter et al. (2019). While social
protection is a broad term that generally covers three elements (social assistance, social insurance and
labour market programmes), this paper’s main focus is on social assistance. Social assistance is defined as
‘non-contributory interventions (the full amount is paid by the provider) designed to help individuals and
households cope with poverty, destitution, and vulnerability’ (ibid.). In this paper, social assistance is treated
as distinct from social protection in the sense that it is more specific and has a narrower scope. Figure 3.1
illustrates the classification of social protection and types of social assistance.
3
PRISMA is an evidence-based minimum set of items for reporting in systematic reviews and meta-analyses:
http://www.prisma-statement.org/.
7
Figure 3.1: Taxonomy of social protection instruments
Source: Carter et al. (2019). OGLv3.0.
Based on the figure above, there are four types of social assistance, generally defined as follows:
1. Social transfers refer to cash transfers (CTs), vouchers or supplementary food, though sometimes they
may include productive assets (such as seeds, fertiliser, tools) or health and education assets (bednets,
food for severe acute malnutrition, schoolbooks). In emergency situations, social transfers may also
include basic domestic goods (cooking implements and utensils, stoves) and shelter (tents, blankets). CTs
are direct, regular and predictable transfers, increasingly paid through secure electronic systems; for
example, directly into bank accounts and mobile phone accounts or on smart cards. CT types include, but
are not limited to:
a. Unconditional cash transfers (UCTs): money transfers that do not impose specific requirements on
how they are spent or any conditions on when they are received.
b. Conditional cash transfers (CCTs): transfers given with the requirement that beneficiaries meet
certain conditions, such as visiting a health clinic or ensuring children go to school.
c. Multipurpose cash transfers (MPCs) or multipurpose cash assistance (MCA): regular or one-off
cash transfers to cover a family’s needs for food, shelter, education, health, water and sanitation
facilities, and to earn a livelihood (NRC n.d.).
d. Cash plus programmes: schemes that combine CTs with one or more types of complementary
support, based on the understanding that ‘cash alone cannot alleviate non-financial and structural
barriers to improving living standards and well-being’ (Roelen et al. 2017).
e. In-kind assistance: includes school feeding and sometimes take-home rations for children most in
need. School feeding programmes aim to both reduce hunger and improve food security, as well as
increase school attendance and learner performance (HLPE 2012).
It is important to note here that different terminologies are sometimes used for similar interventions across
contexts. For instance, CTs are generally used in social protection settings, while ‘cash-based
interventions’ (CBIs) or ‘cash-based transfers’ (CBTs) are more prevalent in humanitarian settings.
Similarly, transfers with multiple objectives that are linked to additional services or programmes are
referred to as ‘cash plus’ in social protection contexts and ‘multi-purpose cash grants’ (MPGs) in
8
humanitarian settings (Roelen et al. 2018). In this paper, CT and CBI are used interchangeably. Similarly,
use of MPC, MCA and MPG denotes similar interventions.4
2. Public works programmes (PWPs) are activities that entail payment of a wage (in cash or food), often
but not always by the state, in return for labour. The aim is to enhance employment and produce a
physical or social asset, with the overall objective of promoting social protection. They are sometimes
referred to as public employment programmes (PEPs), defined as ‘programmes creating state sponsored
employment which is not market based (known as public works programmes, workfare, welfare to work,
cash for work, employment of last resort, employment guarantee programmes, etc.)’ (McCord 2018).
3. Fee waivers are programmes or interventions that subsidise services for the poor. Examples include
health insurance exemptions, reduced medical fees and education fee waivers.
4. Subsidies are instruments to keep prices low for certain goods and services. Examples include food
subsidies, fuel subsidies, housing subsidies and allowances, utility and electricity subsidies and
allowances, agricultural input subsidies, and transportation benefits.
In this review, we included interventions and programmes that fall under the category of social assistance.
Other types of social protection such as social care, social insurance and labour market interventions were
not included. Likewise, microfinance (microcredit) programmes were excluded from the review, as they are a
type of loan not generally considered as cash-based interventions per se (Berg and Seferis 2015).
3.2 Inclusion and exclusion criteria
As a rule of thumb, for the rapid review phase we included papers, studies, reports and evaluations that
examined the impact or effectiveness of all interventions that fall under the categories of social assistance
among refugees, internally displaced persons (IDPs) and asylum-seekers across the four countries of
interest. More specifically, we applied the selection criteria below.
3.2.1 Inclusion criteria
1. Language and timeframe: publications in English within the past 11 years (from 1 January 2010 to 31
January 2021).
2. Study design/type of study: (a) intervention studies that include randomised control trials (RCTs), quasiexperimental, pre-post study and non-randomised controlled trials; (b) observational studies that include
an intervention, which could be quantitative or qualitative – examples are cohort, case-control, crosssectional or case studies (featuring interviews or focus group discussions); and (c) evaluations of
programmes that can be mid-term or ongoing, and conducted by internal or external parties.
3. Participants/populations: refugees, IDPs, asylum-seekers and vulnerable and poor households in host
communities.
4. Country/setting: Syria Iraq, Jordan and Lebanon.
5. Types of intervention/programme: any forms of social assistance classified in Figure 3.1 and described
in the previous section. These include social transfers, PWPs (e.g. cash for work, food for work), fee
waivers (e.g. exemption from health insurance or school/school-related fees) and subsidies (e.g. food/fuel
subsidies), or any combination of two or more. Social transfers mainly comprise UCTs, CCTs, cash plus
programmes and in-kind assistance. Transfers of basic domestic goods (e.g. cooking implements and
utensils, stoves) and shelter-related interventions (e.g. providing tents, blankets or house
repair/rehabilitation) are also considered part of social transfers.
6. Types of outcomes: the focus was on gender, familial and intrahousehold relations and dynamics.
However, we did not exclude papers that examined other outcomes, as we wanted to examine if an
intersectional lens was applied, and if data were disaggregated (e.g. by gender) in papers that studied the
impact or effectiveness of social assistance in general. Our review included the following outcomes, other
than those related to gender, familial and household relations: (a) poverty and vulnerability; (b) negative
coping behaviours; (c) human capital (e.g. health and education); (d) livelihoods; (e) empowerment of
beneficiaries, especially women and girls; and (e) social cohesion.
4
Studies in this review that used the terms MPC or MPA did not describe additional programmes or services linked
to cash transfers.
9
7. Reviews or systematic reviews: those describing social assistance interventions or assessing their
effectiveness or other relevant systematic reviews were included for citation-tracking purposes.
3.2.2 Exclusion criteria
1. Theoretical or conceptual papers, comments, letters and correspondence.
2. Observational studies that did not include or describe a specific intervention.
3. Studies/papers that described social assistance in any of the four countries of interest without specifically
mentioning refugees or IDPs as beneficiaries.
4. Studies/papers that described broad poverty-reduction programmes without being specific about social
assistance or its impact.
5. Studies/papers that described systems approaches or multi-level initiatives in which social assistance,
while included, was a minor component.
6. Studies/papers that addressed other types of social protection such as social care, social insurance and
labour market interventions.
7. Studies/papers that included interventions that were supportive of, but not specifically categorised as
social assistance, such as vaccination programmes and provision of safe water, hygiene and toilet
facilities.
8. Studies/papers that reported outcomes beyond the scope of this review (see inclusion criteria and Table
A2); where a study assessed multiple outcomes, we only reported relevant ones.
9. Formative or process evaluations that did not explicitly report intervention impacts or effectiveness.
Details on the types of social assistance interventions and outcomes included in this review are presented in
Annex 1; the search strategy is in Annex 2; critical appraisal results are in Annex 3 (and Appendix 1); the
coding tool for data extraction is in Annex 4; and tabulated findings from all analysed studies are in Annex 5.
In this review, we used the Joanna Briggs Institute’s critical appraisal tools to assess the quality of papers
included (Joanna Briggs Institute 2017).
3.3 Study limitations
This paper has several limitations. We did not conduct a full double-blind review in the title/abstract screening
and quality appraisal steps. We also included only English publications, therefore a wealth of information and
evidence captured in other languages – especially Arabic – could have been missed. Systematic reviews are
known to be positivist, favouring certain study designs – RCTs or experimental studies – over other designs.
Our strict inclusion/exclusion criteria may therefore have excluded relevant studies/papers that could have
enriched the body of evidence or given a more comprehensive understanding of how social assistance
influences gender, familial and intrahousehold dynamics.
This review was limited to social assistance interventions as defined and classified by the proposed
framework (Figure 3.1). Other programmes that may be impactful but did not fall under the category of social
assistance – such as provision of non-formal education, or psychosocial support for children, or water,
sanitation and health facilities – were not taken into account, although these could exert similar impacts to
social assistance and could be broadly considered as social protection.
Despite climate change being a cross-cutting theme in the BASIC Research Programme, we did not include
any climate-related perspectives in this paper or attempt to link our findings to climate change adaptation.
This was partly because the selected studies/papers did not provide much (if any) discussion on climate
change. However, we acknowledge the importance of identifying opportunities for social protection to
enhance adaptation and to make social protection interventions more climate resilient (Davies et al. 2008).
10
4. Results
Out of 1,564 papers initially identified and screened, 22 were included in the final stage. Figure 4.1 illustrates
the study selection process.
Figure 4.1: Flowchart of study selection
Source: Authors’ own. Based on PRISMA flow diagram (Moher et al., 2009)
Of those 22 studies, ten were conducted in Lebanon, nine in Jordan, three in Syria and one in Iraq. Two
papers from Syria came from the same programme and most likely used similar participants. One study was
conducted in both Jordan and Lebanon. Figure 4.2 shows the distribution of studies/papers included by
country.
11
Figure 4.2: Number of studies by country
Syria
3
Lebanon
10
Jordan
9
Iraq
1
Source: Authors’ own.
Note: The total is more than 22 because one study was conducted across two countries (Jordan and Lebanon).
With regard to study design, ten were quantitative, nine were qualitative and three used mixed methods.
Among the quantitative studies (that include the quantitative components of mixed-methods papers), one
was an RCT, 11 were quasi-experimental and one was cross-sectional. Of the 11 quasi-experimental
studies, one was an adaptive targeted field experiment, a newly developed method. Others used regression
discontinuity design (RDD), propensity score matching (PSM) and pre-test/post-test. Figure 4.3 shows the
study designs of the included papers (n=22).
Figure 4.3: Distribution of study design
10
9
3
Mixed-methods
Quantitative
Qualitative
Source: Authors’ own.
This choice of study design reflected the types of interventions studied, especially e-cards, or e-vouchers, the
context of lower-middle income host countries and the refugees’ backgrounds. The World food Programme
introduced e-cards in 2013 in Lebanon, allowing hundreds of thousands of Syrian refugees to meet their food
needs and boost the local economy (WFP 2013). Humanitarian staff, governments and donors in the region,
found this a new and promising way of delivering social assistance with potential gender benefits (Oosterhoff
2016). E-cards or smart cards, for example, solved issues of gender-based violence (GBV) faced by (mostly
female) refugees when accessing food, fuel or water in other contexts (Asgary, Emery and Wong 2013).
Cards have been scaled up across the region, encouraging and enabling digital data-driven approaches to
social assistance research.
In terms of the types of social assistance, the vast majority were cash transfers, including UCT, CCT and
MPC programmes. Fewer studies assessed the impact of vouchers (n=1), in-kind assistance (n=1), food
12
subsidies (n=1), PWP (n=2) and cash plus programmes (n=1). The outcomes most studied were those
related to poverty and vulnerability (n=13), followed by human capital (n=8), negative coping strategies (n=8),
social cohesion (n=6), empowerment (n=6) and livelihoods (n=6). All the outcomes on empowerment
focused on women, and some included gender and intrahousehold relations and power dynamics.
Figures 4.4 and 4.5 illustrate the types of interventions and outcomes in this review.
Figure 4.4: Types of interventions
16
2
Cash transfers
1
1
1
1
Vouchers
In-kind
assistance
Food subsidy
Cash plus
(combination)
PWP
Source: Authors’ own.
Note: n>22 because some interventions are studied in more than one paper.
Figure 4.5: Types of outcomes
Social cohesion/relations
6
Women empowerment
6
Livelihoods
6
Human capital
8
Negative coping behavior
8
Poverty and vulnerability
13
Source: Authors’ own.
Note: n>22 because each study can have multiple outcomes.
Among the 13 quantitative and mixed-methods studies, ten examined the impact of cash-based interventions
while the remaining three examined the impact of PWP (cash for work), in-kind food assistance (in
comparison with vouchers) and food subsidy programmes. The outcomes of these 13 studies were poverty
and vulnerability (n=9), negative coping behaviours (n=6), human capital (n=7), livelihoods (n=4),
empowerment (n=4) and social cohesion (n=3).
13
Out of the 22 studies, ten examined the effects or influence of social assistance on empowerment, including
familial and intrahousehold dynamics. Figures 4.6 and 4.7 illustrate the distribution of these studies by design
and components of findings.
Figure 4.6: Number of studies looking into empowerment by design
7
2
1
Quantitative
Qualitative
Mixed-methods
Source: Authors’ own.
Figure 4.7: Number of studies that examined the different components of empowerment,
including gender, familial and intrahousehold dynamics
7
5
3
1
1
0
Intrahousehold
dynamics/tension,
decision-making
power
Confidence,
self-esteem
Gender roles
Change in attitude, Domestic abuse,
norms
violence
Marriage, fertility,
contraception
Source: Authors’ own.
Note: one study can have multiple outcomes, and there can be overlap between one or more indicators.
4.1 Overall study quality
Following quality appraisal, three studies were categorised as being of high quality while three were scored
as low quality. The remaining 16 were of moderate quality. Figure 4.8 illustrates the distribution of quality of
all the studies included in this review.
Among the ten studies that examined empowerment – including gender, familial and intrahousehold relations
– in terms of outcomes, three were low quality, two were high quality and five were categorised as moderate
quality.
14
Figure 4.8: Level of confidence/quality of studies included
16
3
High
3
Moderate
Low
Source: Authors’ own.
During quality appraisal, we observed that the two most common aspects lacking across the majority of
moderately ranked studies were: (1) ability to ensure reliability in outcome measurement; and (2) ability to
show that participants received similar treatment other than the intervention/programme being tested. For the
former, most authors did not provide any information to show that outcomes were measured in a consistent
manner; this required descriptions of either the number of raters, training of raters, the intra-raters’ reliability
or the inter-raters’ reliability. For the latter, proving that participants only received the specific aid
(intervention) under study was difficult due to the crisis settings and the presence of multiple aid and
humanitarian agencies that could have been administering other forms of assistance simultaneously.
Another (less) common ‘deficit’ seen across studies was the failure to clearly describe attrition rates at followup and analyse the differences between those followed up and those who dropped out. Similarly, several
studies did not mention any ethical clearance or attempt to address potential ethical issues during the study,
although this was not part of the components assessed in the checklist. Unfortunately, the only RCT included
lacked various key aspects of robust evaluation. These included inadequate descriptions of statistical
approach, how randomisation was performed and whether outcome measurements were reliable.
4.2 Limitations of evidence base
Despite the geographic scope of this rapid review covering four countries in the Middle East, the number of
studies included in the final analysis was low (n=22). Most of the studies were concentrated in Lebanon and
Jordan; Syria and Iraq were under-represented. This could be due to the time frame of our search, which
corresponded closely with the Syrian refugee crisis. After Turkey, Jordan and Lebanon host the highest
number of refugees and asylum-seekers from Syria, which may have led to a greater number of studies in
these countries.
Similarly, the range of interventions was limited given that most studies examined the impact of cash
transfers (mainly UCTs). Fewer studies assessed the effects of food subsidies, in-kind food assistance (in
comparison with vouchers) and PWPs (cash for work) – the small number did not allow a strong conclusion
to be drawn. Likewise, none of the studies examined the effectiveness of other types of social assistance
such as asset transfers, fee waivers and PWPs other than cash for work.
With regard to outcomes, most studies emphasised poverty, food security, and household expenses and
consumption. Some papers included negative coping behaviours, health and education as outcomes. Fewer
examined the impact of social assistance on empowerment, livelihoods or between different sources of social
assistance, including locally available and internationally funded forms of assistance.
While the overall evidence base of this review was of moderate quality, studies that focused on
empowerment (gender and intrahousehold dynamics) could be considered weak to moderate and lacked
robust findings (only one study was categorised as high quality). In addition to their mixed results, many
15
papers did not disaggregate data according to gender or any other characteristics such as age, ability, race,
religion and so on.
Given that study respondents mostly comprised adult men and women, the underlying assumption seemed
to be that adults are the centre of or key players in a family or household, and that familial or intrahousehold
dynamics revolved around them. Younger age groups – particularly adolescent girls and boys – have been
underrepresented, with their roles and influences on these relationships overlooked. Available research on
familial coping strategies suggests that young people are at risk of being exploited when families are in
distress (Jones and Ksaifi 2016). Similarly, there was little mention of older adults. Where study subjects
were adults (men and women), there was little or no attempt to recognise individuals with special needs such
as disability or chronic health conditions.
4.3 Availability of disaggregated data
Collecting disaggregated data is a condition for an analysis of the impact or effects of an intervention. Many
donors demand sex-, age- and (dis)ability-disaggregated data in their reporting. Humanitarian handbooks
and guidelines have prescribed it for decades, preferably giving both a factual and a narrative analysis. Yet
researchers on social assistance seem not to pay much attention to this.
Only one study had both a narrative description and quantitative sex-disaggregated data. Nine out of 22 had
no sex-disaggregated data. No studies mentioned that there are different ways to fill in or embody gender.
Four studies had a narrative analysis of data collected from men and women, while five studies reported sexdisaggregated data only for one or few selected study outcomes. Similarly, intersectionality or an
intersectional analysis would require disaggregated data. Twelve studies did not disaggregate by other
characteristics such as age. Most studies that looked at empowerment as an outcome of social assistance
focused exclusively on women, without taking into account men, or young girls and boys. This makes it
difficult to analyse gender, intersectionality and power.
While most studies did not provide age group or gender-disaggregated data, two studies focused on children
as their subjects and two included only women. The rest either targeted households or described the number
and/or percentage of males and females in their sample, but not in the outcomes. Two studies provided
gender-disaggregated data with regard to the findings, but only partially. An example was Chaaban et al.
(2020) who examined the impact of MPC on food security, household expenditure, living conditions,
education, employment and health. But the distinction between male and female subjects was reported only
for employment-related outcomes.
Most studies (n=13/22) collected data at household or individual levels, which is not surprising as those are
the levels at which social assistance is provided. But households and individuals are part of wider networks
and broader socio-political, cultural and economic infrastructures. Individuals and households receive, borrow
from and lend money to relatives, neighbours and religious institutes, for example. These linkages, and
resources such as remittances, religious charity or politically motivated social assistance are important in
helping people to make decisions, complement social assistance packages and diminish their dependency
on international or national assistance.
In summary, the evidence base largely comprised moderately ranked studies in terms of the quality that
could be attributed to the findings. In addition, there were significant gaps in relation to the countries covered
(Iraq and Syria were under-represented), types of intervention studied (the focus was predominantly on
UCTs), and effects by gender and age group.
Findings of all the studies included in this review are presented in three tables in Annex 5; Table A5 presents
the findings of all quantitative and mixed-methods studies; Table A6 presents findings of all qualitative
studies; and Table A7 focuses on outcomes related to empowerment, and gender and intrahousehold
dynamics. Tables A5 and A6 show the availability (or lack thereof) of disaggregated data according to gender
and other characteristics such as age, ability, race and religion.
The section below elaborates and synthesises our findings on how social assistance affects gender, familial
and intrahousehold relations and power dynamics according to the research questions.
16
5. Discussion
Question 1: What is known about the effects of social assistance on gender, familial and household relations
among refugees and displaced populations in Syria, Iraq, Jordan and Lebanon?
At the individual level, most recipients – especially women – perceived CTs as having increased their sense
of dignity, self-worth, confidence and decision-making power. CTs were said to cause temporary relief from
stress, restored sense of dignity, and increased self-confidence and self-efficacy (Blackwell et al. 2019).
Female recipients said they had earned greater respect and a sense of independence as a result of bringing
resources into the household and being able to meet their families’ basic needs (Yoshikawa 2015). Although
these effects may not necessarily be linked to gender or intrahousehold relations, changes at the individual
level can be seen as a starting point that would lead to cascading effects at higher levels of social
interactions.
However, CT recipients reported psychological distress despite receiving CTs. For instance, Falb et al.
(2020) found that a three-month CT had caused symptoms of depression among its beneficiaries; they were
anxious about how they would survive once aid stopped (ibid.). Prior studies have highlighted the role of
duration in affecting CT outcomes and emphasised the need to ensure that interventions were long enough
to achieve more sustainable effects and avoid unintended outcomes (Manley, Gitter and Slavchevska 2012;
UNHCR 2019). In another study that examined the impact of a one-year CT on Syrian refugees in Jordan,
respondents reported a decline in stress and anxiety, which could be attributed to the longer duration and
regularity of aid (Hagen Zanker et al (2017).
Berg, Mattinen and Pattugalan (2013) found mixed reactions and opinions among recipients (men and
women) on how CTs could empower individuals. While some affirmed they felt a sense of dignity because of
receiving aid, others felt that it made no difference given the small amount, and that such assistance could
not solve other key issues and future uncertainty. Similarly, CFW programmes showed mixed effects on
individuals’ opinions and beliefs about gender roles and equality. While Loewe (2020) reported that CFW
contributed to increasing acceptability of female labour participation, Oxfam (2019) found no impact of CFW
on participants’ opinions about gender equality. Subjects were asked if they agreed that men should support
women in household chores, if women could engage in paid work, and whether it was acceptable for men
and women to mix in the workplace. No difference was found in responses to these questions between
treatment and control groups.
At the familial or household levels, there was consistent evidence that CTs led to improved relationships and
a decline in tension and disputes (Lehmann and Masterson 2014). This was mainly attributed to the
immediate economic relief CTs brought, thereby reducing poverty-related stress (Abu Hamad et al. 2017).
However, intrafamilial dynamics with regard to who controlled the cash were complex. Married women largely
perceived an improvement in decision-making power and felt that they made decisions jointly with their
husbands (Blackwell et al. 2019; Pertek 2016; Yoshikawa 2015). But it was not clear what a ‘joint decision’
meant in terms of the weight of each role or opinion. This could mean a light discussion or consultation led by
the husband, or a more detailed joint analysis of facts and action planning.
Women’s experiences in terms of bargaining power and decision-making over CTs and household resources
were diverse. Many recipients interviewed maintained that they were completely independent, while others
had to negotiate with their husbands, although they had access to the money (Yoshikawa 2015). Some
admitted not having any role in financial decisions as they accepted their husband’s position as decision
maker, whether they agreed on how the money was spent. Interestingly, some women reported purchasing
non-essential items – such as cigarettes – for their husbands, as a way of negotiating and mitigating
household tensions, so that they could obtain financial decision-making power (Blackwell et al. 2019).
Likewise, mixed responses were seen among beneficiaries of a cash plus programme (UCT linked to
psychosocial support targeting GBV). While women reported their spouses’ increased openness to dialogue,
thereby allowing the women greater control of household finances, some men expressed negative views.
Where women received CTs, it was associated with husbands’ ‘loss-of face’, fights over how the money
should be spent, projecting a bad image of men in front of their mothers, and an inability to spend the money
17
fairly among families. Thus, women’s role or status did not change because of receiving CTs, and they were
expected to give all or part of the cash to men to manage. Women mostly agreed that the economic relief of
CTs had resulted in a decline in domestic violence (including violence against children), which was supported
by counsellors and cash assistance officers interviewed. Few male recipients, however, concurred with the
link between domestic violence and financial stress (Yoshikawa 2015).
For women living in extended households, power dynamics and relations could be further compounded. In
some instances, CTs were reported to have triggered tensions with mothers-in-law who held more power in
household affairs and assumed role as financial decision makers (Blackwell et al. 2019). This made it more
difficult for women to access money or have a say on how it should be spent, as they now had to negotiate
with both their husbands and their mothers-in-law. It is worth mentioning that even when married couples did
not co-reside with in-laws, that did not mean the in-laws had no influence over their household affairs.
This could also be seen in the case of divorced and widowed women who received CTs. While most of them
expressed satisfaction at having the power and autonomy to make independent financial decisions without
having to negotiate with male family members, a smaller number claimed that male family members and inlaws still played a role, especially when it came to going out to spend the cash (ibid.). However, this did not
necessarily imply an attempt to control women’s movement, as some female recipients reported needing
help with purchasing groceries and other items, which could be due to safety reasons or avoiding suspicion
(ibid.).
Some participants interviewed reported hearing of divorce cases cause by CT programmes targeting single,
divorced and widowed women, but most were anecdotal. Both men and women were said to initiate divorces;
men did so to marry another woman, while women did so to facilitate their access to aid. Respondents
pointed out that the preference to target single women sometimes helped men justify divorces or separations
on the grounds that the women could access cash assistance more easily. However, the link between the
targeting of aid and divorce could not be further determined and needs further research (Yoshikawa 2015).
Rumours about the gendered impacts of CT on divorces or separations may reflect an attempt to make
sense of decisions about CT. Several studies showed that refugees do not always understand how recipients
are selected and how targeting works, which causes tensions.
Question 2: Have these evaluations and/or interventions taken intersectionality into account?
Most studies did not provide disaggregated data according to characteristics other than sex. There was little
or almost no mention of how the outcomes studied affected people with special needs and older adults. Few
studies disaggregated data by age group when the programme focus was on children. Interestingly, less
attention was given to adolescent boys and girls, despite adolescence being a critical period of development
that can be a ‘window of opportunity’ for interventions to offset disadvantage experienced in childhood.
It is worth highlighting that for women marital status was an important additional characteristic that could
influence the impact of social assistance. Beyond marital status and the presence (or absence) of children,
there was no attempt to identify other dimensions or characteristics of women that could affect their
experiences with social assistance and family or household relations.
Syrian refugees or displaced persons were largely portrayed as a homogenous entity; study sample data
rarely acknowledged the differences in their religious identity. While Sunni Muslims make up the majority of
Syrian refugees, there are also Christian, Druze, Ismaili and other non-Muslim minority religious people
(Eghdamian 2015). Little is known about the status and condition of these minorities, and how this aspect
(religious identity) intersects with gender or age to produce extra layers of discrimination or oppression that
may affect their ability to benefit from social assistance.
Staff, especially field staff who play essential roles in the whole system of inclusive social assistance, receive
little if any attention in the research on ‘what works’ in social assistance in (protracted) emergencies. Field
staff are directly affected by intersectional gender norms and practices in refugee and IDP communities, as
well as in their own families, which may or may not be similar. They have, for example, experience getting
access and winning the trust of targeted individuals, families and households. Field staff, whether employed
by the state or an (international) NGO often work long hours on precarious, short-term contracts in harsh
conditions. At the bottom of organisational pecking orders and hierarchies of national and internationally
18
supported social assistance systems, field staff are important gatekeepers for individuals, households and
families, a job which can be both rewarding and stressful. This invisibility, and general lack of recognition of
field staff in much of the research, suggest a lack of empathy for the human beings delivering social
assistance in humanitarian crises and hinders realism about power relations in social assistance
programmes.
In sum, most studies did not take intersectionality into account, other than some mention of gender relations
and power dynamics, or specific needs of children. This could be due to several reasons. First, in
humanitarian contexts, social assistance programmes tend to aim at immediate outcomes such as relieving
food insecurity, improving household expenditure and facilitating access to school or health services.
Outcomes which are long-term and transformative in nature may not be prioritised, unlike in stable
development contexts. Second, individuals’ and households’ diverse needs (e.g. having older persons or
persons living with disabilities) may have been considered at the planning and design phase of interventions
but not necessarily at the evaluation stage.
Question 3: How do these interventions affect familial, household level and intrahousehold level power
dynamics?
Overall, the effects of social assistance – with CTs being studied most frequently in this review – tend to be
mixed, complex and dependent on various factors. These factors included a woman’s marital status, the
presence and influence of other family members (in-laws, brothers, etc.), the amount and duration of aid, and
the existence of complementary interventions (e.g. cash plus). Prior evidence showed that delivery modality,
benefit level and duration are key in addressing gendered risks and vulnerabilities (Camilletti 2021), while
additional programmes or services that serve to strengthen CTs led to better and long-lasting gender
outcomes (Roy et al. 2019).
While there was consistent evidence that women who received CTs felt more confident, independent and
dignified, and assumed greater decision-making power, this was mainly related to household affairs such as
food purchases and health or educational expenses. Whether or not this decision-making power translated
into more equal gender and intrahousehold relations in the long run, or a greater say in other matters, was
unclear. This was in line with previous findings in both development and humanitarian contexts that any
improvement in women’s decision-making capacity because of CTs was largely restricted to the domestic
sphere. In addition, in many fragile contexts, the common belief was that men remained the primary decision
makers (Simon 2019).
Many recipients did not see CTs had any meaningful impact beyond fulfilling basic, immediate needs due to
their small amount (Berg et al. 2013; Abu Hamad et al. 2017). This could explain why, despite multiple claims
that intrahousehold tensions reduced, there was little overall improvement in subjective wellbeing and
relationships with husbands. Prior studies highlighted that men perceived small CT amounts to help women
perform their regular household tasks and were therefore welcomed (Wasilkowska 2012). However,
substantial transfers could pose a threat to traditional male roles, potentially triggering resistance and
dissatisfaction (Brady 2011; Simon 2019; Wasilkowska 2012).
Most interventions in this review were cash based, with fewer studies on other types of social assistance. For
instance, there were only two CFW programmes and both studies examined opinions and attitudes about
gender roles as an outcome, with no inquiry or analysis into how participation in CFW programmes affects
relations and power dynamics at the family or household levels (Loewe 2020; Oxfam 2019). Furthermore,
evidence from these two studies was mixed. Therefore, it is difficult to draw any conclusion on the possible
effects of CFW on gender and intrahousehold dynamics in this context.
In sum, it can be said that social assistance (mainly CTs) exerts positive effects on individual recipients by
giving them a sense of autonomy, confidence and respect. It also often results in reduced tensions in
households due to immediate economic relief and fulfilment of basic needs. This decline in poverty-related
stress may have a positive impact on spousal relationships, but evidence on whether social assistance that
targets women can cause more transformative and sustainable changes in terms of gender equality and
women’s empowerment (beyond the domestic sphere) is limited and mixed. Several factors related to
interventions have been identified as crucial to increasing their effectiveness, such as the amount and
19
duration of aid. It is also important to emphasise that social assistance or CTs alone cannot be relied on to
bring about and sustain meaningful changes. The wider structural contexts, social norms and supplemental
programmes need to be considered along with the core interventions.
6. Conclusion
This literature review aimed to explore the evidence on the effects of social assistance on gender, familial
and household relations and power dynamics among refugees and (internally) displaced populations in Syria,
Iraq, Jordan and Lebanon. While there is no scarcity of research on the effects of crises on the composition,
emotional wellbeing and livelihoods of families and households in the region, we found little evidence on the
specific effects of social assistance on long-term relations and power relations within families and
households, or how social assistance changes livelihood strategies.
A question that emerged as we analysed the selected papers was whether the arduous work of targeting
individuals was efficient or necessary, given that beneficiaries generally tend to share their stipend with other
family members for the collective good. This was indicated in studies such as Lehmann and Masterson
(2014), which reported how CTs improved relationships within households by reducing the number of
disputes between family members, even though beneficiaries did not experience greater subjective wellbeing
compared to non-beneficiaries. Action research on gender mainstreaming in cash transfers showed that
pregnant Syrian women shared their stipend with elderly or otherwise needy family members who lacked
social and financial assistance (Oosterhoff 2016). Families and individuals have their own sense of and
systems for social justice. Questions remain over whether, when and on what grounds international or state
support should aim to change local systems, norms and practices through the distribution of various types of
social assistance. A crisis may be a good opportunity for social change, but who benefits and who loses from
social assistance in the immediate and longer terms?
With regard to whether intersectionality was taken into account, most studies tended to conflate gender with
women and girls – making distinctions between widowed, married, unmarried and divorced women – but
ignoring other dimensions such as class, health status, religion, ethnicity, education, prior work experience,
political affiliation and civil participation. Given the large body of research on the importance of these social
characteristics in understanding an individual’s socioeconomic and political status, this is a clear research
gap. Whether this research gap should be addressed to improve programme effectiveness depends on how
and whether it aims to transform social relations and promote equity.
Humanitarian principles and standards should apply to both practitioners and researchers working in
humanitarian contexts. This means that power hierarchies within organisations and social assistance
systems should be acknowledged with an eye to promoting the socio-political and economic sustainability of
programmes. Gender-, age- and disability-disaggregated data are supposed to be part of development and
humanitarian interventions, and are a condition for intersectional gender analysis. Recognition of power,
gender and intersectionality can be seen as being at odds with neutrality. Yet not recognising power
dynamics based on gender and other intersecting characteristics can expose organisations and programmes
to accusations of (re)enforcing inequalities either in their own organisations or in the communities they serve.
This is an almost intractable dilemma. We found that data in the selected papers were frequently not
adequately disaggregated, which seriously restricted our ability to analyse in depth the effects of these
interventions on gender and other social relations. It also shows a lack of awareness of how gender and
social inclusion or exclusion work. Failure to disaggregate data – which can be due to time, financial or other
resource constraints – exposes organisations to risks and questions around the accountability, effectiveness
and sustainability of social assistance.
Gender is different to sex. Studies in this area look at the relationships between men and boys and women
and girls as well as people with fluid gender identities. All papers included for analysis in this review that
studied empowerment as an outcome looked only at women, suggesting an underlying assumption that only
this group needs to be empowered, without taking men or characteristics such as ability, religion and age into
20
account. This demonstrates a limited understanding of the multiple dimensions of power and their interaction
with gender.
Beyond social assistance programmes’ focus on individuals and households, there was little discussion
about the wider context, despite the glaring situation and issues related to displacement in the region.
Whether this is a technical or political choice, it risks making families implicitly responsible for problems that
are well outside of their authority. Although these studies took place in protracted emergencies, and three out
of the four sovereign states lack control over their territory, the studies paid remarkably little attention to the
overall conflict. It is unclear what the reasons for this were. It could reflect a humanitarian principle or the
desire to be ‘neutral’, but allocation of scarce resources in a poor context affects power relations, which are
not neutral, but rather intersectional and gendered at multiple levels.
Although there is evidence that social assistance alone rarely, if ever, covers all the expenses of a family, and
is time bound, most studies paid negligible attention to familial infrastructures and strategies for sustainable
interventions. Access to, and use of, cash transfers are part of broader familial strategies to mobilise or
increase resources including, for example, (male) migration in pursuit of remittances, or (female) dependency
on ‘community charity’. Short-term cash transfers can, in some circumstances, disrupt individuals’ and
families’ access to more sustainable income or ‘charity’. Thus, important questions are raised about the
purpose of social assistance: does it aim to preserve or transform families through targeting; and how do
families perceive this and why?
Lastly, it is important to highlight that the geographical scope of this review means that while some gender
norms may be unique to the MENA region, these norms and relations can change over time. Refugee status
affects pre-conflict gender relations in multiple ways (Harvey et al. 2013). While modesty about the
predictability of human relations is wise, social assistance programming should acknowledge that gender and
intersectionality shape access to and control over the use of social assistance at the individual, household,
familial and wider community system.
7. Recommendations and Implications for future
research:
Based on the findings – and research gap – in this review, we offer several recommendations for
humanitarian actors, service providers, researchers and policymakers:
1. Support a context-specific and historical contingent understanding of gender and intersectionality in social
assistance systems. Use the experiences of field staff and refugees to inform rights-based strategic
management in action-oriented participatory research focused on inclusive social assistance systems.
Clarify, at all levels, whether the social assistance programme aims to achieve equity or equality of
individuals, households, families or social groups, or a combination of these, and what that means in
practice as part of expectation management and programme sustainability.
2. Recognise and research what formal and informal social assistance programmes and mechanisms are in
place in recipient communities and how these interact with social assistance programmes by the state or
international actors. Support reliable, forward-looking, strength-based approaches to social assistance –
formal or informal – that recognise and encourage people’s contributions, solidarity and aspirations,
including those of field staff.
3. Apply humanitarian principles and guidance on gender and intersectionality to research and evaluations
on lived individual and collective experiences in local contexts along the humanitarian-development
nexus. Acknowledge power differences within families, households and organisations delivering social
assistance.
4. Support mixed-methods research, including use of (disaggregated) monitoring, evaluation, accountability
and learning (MEAL) data on social assistance combined with qualitative, participatory and creative
methods to capture diverse and changing perspectives between places and across life courses.
21
5. Promote the inclusion and participation of men and boys – along with women and girls – in programmes
that aim to empower individuals or families and improve gender equity. Similarly, include the perspectives
of those with special needs such as the elderly and people living with disability when exploring the effects
of social assistance on relations and power dynamics.
6. Research familial short- and long-term strategies for sustainable income for families to meet their
expenses in contexts with limited access to (formal) employment and (safe) credit.
We also propose the following research questions for the BASIC Research Programme:
1. What are the effects of state-led and (international) NGO social assistance interventions on familial and
household relations among refugees and (internally) displaced populations in conflict and climate-related
crises?
2. Can an intersectional, adaptive and action-oriented approach to social assistance systems in conflicts in
the MENA region and beyond strengthen inclusion and accountability? What can be useful indicators at
familial, household and organisational levels?
3. How do government and (international) NGO staff who deliver social assistance and operate in the field
deal with diversity, intersectionality, gender and familial obligations within their own organisations?
4. Do ideas about inclusion and exclusion among field staff align with those of management and recipient
populations (i.e. the target individuals, households and families)? How are contradictions or conflicts about
targeting and tailoring resolved at different levels? Is this satisfactory and for whom?
5. What are the opportunities and costs for organisations to reach the most marginalised and vice versa?
What are some of the main trade-offs in extending coverage?
6. What are the existing familial or community infrastructures, strategies and informal support used by
refugees and IDPs to secure and sustain livelihoods? How can social assistance leverage these
resources, while taking into account equity, gender and intersectionality?
To answer these questions, we propose mixed-methods research that combines desk research on human
resource policies and MEAL systems with field observations and action research. Researchers should spend
time with families and communities who receive targeted social assistance to improve understanding of how
various households and families decide to access and use social assistance. Use of (stratified) random
samples from lists of beneficiaries will reduce bias. A comparative study looking at government, international
NGO and local (e.g. religious) social assistance systems will help improve understanding of the costs and
benefits of providing assistance to marginalised groups. The action research should be broadly framed by
academic evidence and integrate existing operational evidence generated by MEAL systems, for example.
A mixed-methods approach with participatory and other research techniques and methods (e.g. analysis of
secondary data sets, literature reviews of policies, archival materials, literature, media, in-depth interviews,
focus group discussions) could have different components that build on each other and leave flexibility, which
would make most sense. Participatory techniques could involve participant observation, mapping, ranking,
performative methods, diary drawing, photography or journaling.
For the action research, stakeholders should formulate questions that resonate with their lived experiences to
produce new evidence and insights that are locally relevant. Researchers could facilitate development of a
research design to answer practitioner- and beneficiary-led questions. The research techniques and methods
could be narrative, visual, quantitative or qualitative, depending on the research questions. The action
research could be an analysis of data generated by mapping, ranking, participatory observations, immersion,
statistical analysis or literature review. Participants should be involved in the development of the research
design, data collection and analysis and uptake, not just in the data collection or attending a feedback
session, for example. For the uptake of research evidence, commitment from the leadership of organisations,
and possibly funders, is necessary as this involves the allocation of time and scarce (human) resources.
22
References
Abu Hamad et al. (2017) A Promise of Tomorrow: The Effects of UNHCR and UNICEF Cash Assistance
on Syrian refugees in Jordan, Overseas Development Institute (ODI)
Akesson, B. and Sousa, C. (2020) ‘Parental Suffering and Resilience among Recently Displaced Syrian
Refugees in Lebanon’, Journal of Child and Family Studies, 29(5): 1264–1273
Asgary, R.; Emery, E. and Wong, M. (2013) ‘Systematic Review of Prevention and Management
Strategies for the Consequences of Gender-based Violence in Refugee Settings’, International Health,
5(2): 85–91
Bastagli, F.; Samuels, F.; Stavropoulou, M.; Tukmani, N.; Abbani, H and Plank, G. (2021) World Food
Programme Multi-purpose Cash Assistance in Lebanon: Protection Outcomes for Syrian Refugees,
Overseas Development Institute (ODI)
Battistin, F. (2016) Impact Evaluation of the Multipurpose Cash Assistance Programme, Beirut: Lebanon
Cash Consortium
Berg, M. and Seferis, L. (2015) Protection Outcomes in Cash Based Interventions: A Literature Review
Danish Refugee Council, UNHCR, and ECHO
Berg, M.; Mattinen, H. and Pattugalan, G. (2013) Examining Protection and Gender in Cash and
Voucher Transfers, World Food Programme (WFP) and UNHCR
Best, L.; Stevens, A. and Colin‐Jones, D. (1997) ‘Rapid and Responsive Health Technology
Assessment: The Development and Evaluation Process in the South and West Region of England’, J
Clin Effec., 2:51–6
Bila, B. and Egrot, M. (2009) ‘Gender Asymmetry in Healthcare-facility Attendance of People Living with
HIV/AIDS in Burkina Faso’, Social Science & Medicine, 69(6): 854–861
Blackwell, A., Casey, J., Habeeb, R., Annan, J. and Falb, K. (2019) ‘Women’s Status and Qualitative
Perceptions of a Cash Assistance Programme in Raqqa Governorate, Syria’, Gender & Development,
27(2): 253–271
Boston Consulting Group (2017) Food-restricted Voucher or Unrestricted Cash? How to Best Support
Syrian refugees in Jordan and Lebanon?, World Food Programme
Brady, C. (2011) Walking the Talk: Cash Transfers and Gender Dynamics, Concern Worldwide and
Oxfam GB
Cameron, A.; Watt, A.; Lathlean, T. and Sturm, T. (2007) Rapid versus Full Systematic Reviews: An
Inventory of Current Methods and Practice in Health Technology Assessment, ASERNIP-S Report
Number 60, Adelaide: Australian Safety and Efficacy Register of New Interventional Procedures –
Surgical (ASERNIP-S)
Camilletti, E. (2021) Social Protection and its Effects on Gender Equality: A Literature Review, United
Nations.
Caria, S.; Gordon, G.; Kasy, M.; Quinn, S.; Shami, S.; and Teytelboym, A. (2020) An Adaptive Targeted
Field Experiment: Job Search Assistance for Refugees in Jordan (2020). CESifo Working Paper No.
8535, SSRN
Carter, B.; Roelen, K.; Enfield, S. and Avis, W. (2019) Social Protection Topic Guide. Revised Edition,
K4D Emerging Issues Report 18. Brighton: Institute of Development Studies
Chaaban, J.; Salti, N.; Ghattas, H.; Moussa, W.; Irani, A.; Jamaluddine, Z. and Al-Mokdad, R. (2020)
Multi-purpose Cash Assistance in Lebanon: Impact Evaluation on the Well-being of Syrian Refugees,
American University of Beirut Press
Concern Worldwide US (2020) The Ten Largest Refugee Crises to Follow in 2021
Cluver, L.D.; Orkin, F.M.; Boyes, M.E. and Sherr, L. (2014) ‘Cash Plus Care: Social Protection
Cumulatively Mitigates HIV-risk Behaviour among Adolescents in South Africa’, Aids, 28, S389–S397
23
Corabian, P. and Harstall, C. (2002) ‘Rapid assessments provide acceptable quality advice’, Annu Meet
Int Soc Technol Assess Health Care Int Soc Technol Assess Health Care Meet,18: Abstract 70
Cornwall, A. and Jolly, S. (2009) ‘Sexuality and the Development Industry’, Development 52.1: 5–12
Daly, M.; Bray, R; Bruckauf, Z.; Byrne, J.; Margaria, A.; Pecnik, N. and Samms-Vaughan, M. (2015)
Family and Parenting Support: Policy and Provision in a Global Context, UNICEF Office of Research –
Innocenti
Davies, M.; Leavy, J.; Mitchell, T. and Tanner, T. (2008) Social Protection and Climate Change Adaption,
Report prepared for the Commission on Climate Change and Development, Brighton: Institute of
Development Studies
De Châtel, F. (2014) ‘The Role of Drought and Climate Change in the Syrian Uprising: Untangling the
Triggers of the Revolution’, Middle Eastern Studies, 50(4): 521–535
De Hoop, J.; Morey, M.; Ring, H.; Rothbard, V. and Seidenfeld, D. (2019) ‘Min Ila’ Cash Transfer
Programme for Displaced Syrian Children in Lebanon (UNICEF and WFP) Impact Evaluation Endline
Report, Innocenti Research Report, UNICEF.
DFID (2014) Assessing the Strength of Evidence: How to Note: A DFID practice paper, ALNAP
Doocy, S.; Tappis, H.; Lyles, E.; Witiw, J. and Aken, V. (2017) ‘Emergency Food Assistance in Northern
Syria: An Evaluation of Transfer Programs in Idleb Governorate’, Food and Nutrition Bulletin, 38(2): 240–
259
Eghdamian, K. (2015) ‘Refugee Crisis: Syria’s Religious Minorities Must Not be Overlooked’, The
Conversation
Falb, K.L.; Blackwell, A.H.; Stennes, J. and Annan, J. (2020) ‘Cash Assistance Programming and
Changes Over Time in Ability to Meet Basic Needs, Food Insecurity and Depressive Symptoms in Raqqa
Governorate, Syria: Evidence from a mixed methods, pre-posttest’, PloS one, 15(5), e0232588.
Farhat, J.B.; Blanchet, K.; Bjertrup, P.J.; Veizis, A.; Perrin, C.; Coulborn, R.M., … and Cohuet, S. (2018)
‘Syrian Refugees in Greece: Experience with Violence, Mental Health Status, and Access to Information
During the Journey and While in Greece’, BMC Medicine, 16(1): 1–12
Hagen-Zanker, J., Ulrichs, M., Holmes, R., & Nimeh, Z. (2017). Cash Transfers for Refugees: The
Economic and Social Effects of a Programme in Jordan. London: Overseas Development Institute and
Aya Consultancy.
Hagen‐Zanker, J.; Ulrichs, M. and Holmes, R. (2018) ‘What are the Effects of Cash Transfers for
Refugees in the Context of Protracted Displacement? Findings from Jordan’, International Social
Security Review, 71(2): 57–77
Harvey, C.; Garwood, R. and El-Masri, R. (2013) Shifting Sands: Changing Gender Roles among
Refugees in Lebanon, Oxfam International
Heath, R.; Hidrobo, M. and Roy, S. (2020) ‘Cash Transfers, Polygamy, and Intimate Partner Violence:
Experimental Evidence from Mali’, Journal of Development Economics, 143, 102410.
HLPE (2012) Social Protection for Food Security. A Report by the High Level Panel of Experts on Food
Security and Nutrition of the Committee on World Food Security, Rome: Food and Agriculture
Organization.
Joanna Briggs Institute (2017) Critical Appraisal Tools
Jones, K. and Ksaifi, L. (2016) Struggling to Survive: Slavery & Exploitation of Syrian Refugees in
Lebanon, Freedom Fund.
Khangura, S.; Konnyu, K.; Cushman, R.; Grimshaw, J. and Moher D. (2012) ‘Evidence summaries: the
evolution of a rapid review approach’, Syst Rev., 1:10
Lehmann, M.C. and Masterson, D.T. (2020) ‘Does aid reduce anti-refugee violence? evidence from
Syrian refugees in Lebanon’, American Political Science Review, 114(4): 1335–1342
24
Lehmann, M.C., and Masterson, D.T. (2014) An impact evaluation of the 2013–2014 winter cash
assistance program for Syrian refugees in Lebanon. International Rescue Committee. Levine, S. and
Mosel, I. (2021) ‘Economics, Social Status and Gender Relations: What Makes Households ‘Femaleheaded’ in Somalia? Lessons from a Rapid Learning Exercise’, HPG Briefing Note, London: ODI
Lewis J. (1992) ‘Gender and the development of welfare regimes’, Journal of European Social Policy,
2(3):159–173
Lombardini, S. and Mager, F. (2018) Livelihoods in the Za’atari Camp: Impact Evaluation of Oxfam’s
Cash for Work Activities in the Za’atari Camp (Jordan), Oxfam
Loewe, M. (2020) Community Effects of Cash-for-work Programmes in Jordan: Supporting Social
Cohesion, More Equitable Gender Roles and Local Economic Development in Contexts of Flight and
Migration, DIE Studies, 102Malhotra, A., & Elnakib, S. (2021). 20 years of the evidence base on what
works to prevent child marriage: A systematic review. Journal of Adolescent Health, 68(5), 847–862.
Manley, J.; Gitter, S. and Slavchevska, V. (2012) How Effective are Cash Transfer Programmes at
Improving Nutritional Status? A Rapid Evidence Assessment of Programmes’ Effects on Anthropometric
Outcomes, Social Science Research Unit, Institute of Education, University of London
McCord, A. (2018) Linking Social Protection to Sustainable Employment: Current Practices and Future
Direction, Social Protection for Employment Community (SPEC)
McNatt, Z; Boothby, N.G.; Al-Shannaq, H.; Chandler, H.; Freels, P.; Mahmoud, A.S., … and Zebib, L.
(2018) Impact of Separation on Refugee Families: Syrian Refugees in Jordan
Moher, D.; Liberati, A.; Tetzlaff, J.; Altman, D.G. and Prisma Group (2009) ‘Preferred reporting items for
systematic reviews and meta-analyses: the PRISMA statement’, PLoS medicine, 6(7): e1000097
Newton, J. (2016) Making Social Protection Gender Sensitive for Inclusive Development in Sub-Saharan
Africa, INCLUDE Platform
NRC (n.d.) Cash and Vouchers, Norwegian Refugee Council (NRC)
O’Brien, C.; Scott, Z.; Smith, G.; Barca V.; Kardan, A.; Holmes, R.; Watson, C. and Congrave, J. (2018)
Shock-Responsive Social Protection Systems Research: Synthesis Report, Oxford: Oxford Policy
Management
Oosterhoff, P.; Waldman, L. and Olerenshaw, D. (2014) Literature Review on Sexuality and Poverty
Oosterhoff P. (2016) Innovations in Lebanon: Gender Mainstreaming from the Ground-up Knowledge
Sharing Workshop Emerging Themes and Good Practices for Gender-sensitive Food Security
Programming, Workshop Report – World Food Programme, Brighton: IDS
Orloff, A. (1996) ‘Gender in the welfare state’, Annual Review of Sociology, 22(1): 51–78
Otto, H. and Keller, H., (2014) Different Faces of Attachment: Cultural Variations on a Universal Human
Need, Cambridge University Press
Oxfam (2019). Livelihoods in the Za’atari Camp: Impact evaluation of Oxfam’s Cash for Work activities in
the Za’atari camp (Jordan). Oxford, UK.
Patel, L. and Hochfeld, T. (2011) ‘It buys food but does it change gender relations? Child Support Grants
in Soweto, South Africa’, Gender & Development, 19(2): 229–240
Pertek, I.S., (2016) Conditional Cash Project for Vulnerable Syrian and Jordanian Children in Irbid,
Jordan, Islamic Relief
Phadera, L.; Sharma, D. and Wai-Poi, M. (2020) Iraq’s Universal Public Distribution System: Utilization
and Impacts During Displacement, World Bank Policy Research Working Paper, (9155)
Roelen, K.; Devereux, S.; Abdulai, A.G.; Martorano, B.; Palermo, T. and Ragno, L. P. (2017) How to
Make ‘Cash Plus’ Work: Linking Cash Transfers to Services and Sectors, UNICEF
Roy, S.; Hidrobo, M.; Hoddinott, J. and Ahmed, A. (2019) Transfers, Behavior Change Communication,
and Intimate Partner Violence: Postprogram Evidence from Rural Bangladesh, Review of Economics
and Statistics 101(5), 865–877.
25
Samuels, F.; Bastagli, F.; Stavropoulou, M.; Tukmani, N.; Abbani, H. and Plank, G. (2020) World Food
Programme Multipurpose Cash Assistance in Lebanon: Social Cohesion and Stability Between Syrian
Refugees and Host Communities, ODI for the Cash Monitoring, Evaluation, Accountability and Learning
Organisational Network (CAMEALEON)
Selby, J.; Dahi, O.S.; Fröhlich, C. and Hulme, M. (2017) ‘Climate change and the Syrian civil war
revisited’, Political Geography, 60: 232–244
Simon, C.A. (2019) The Effect of Cash-based Interventions on Gender Outcomes in Development and
Humanitarian Settings, United Nations
Slater, R., and Sabates-Wheeler, R. (2021): Defining Social Assistance for the BASIC Research
Programme: Internal Guidance Note
Sloane, E. (2014) The Impact of Oxfam’s Cash Distributions on Syrian Refugee Households in Host
Communities and Informal Settlements in Jordan, Oxfam
Sphere (2018) Sphere Handbook: Humanitarian Charter and Minimum Standards in Humanitarian
Response, Practical Action
Stevens, M.R. (2016) ‘The Collapse of Social Networks among Syrian Refugees in Urban Jordan’,
Contemporary Levant, 1(1): 51–63
Tricco, A.C.; Antony, J.; Zarin, W.; Strifler, L.; Ghassemi, M.; Ivory, J., … and Straus, S.E. (2015) ‘A
Scoping Review of Rapid Review Methods’, BMC Medicine, 13(1): 1–15
UNHCR (2019) Cash Assistance and the Prevention, Mitigation and Response to Sexual and Genderbased Violence (SGBV): Findings from Research in Lebanon, Ecuador and Morocco
UNICEF (2020) UNICEF’s Social Protection Response to COVID-19, New York: UNICEF
Wasilkowska, K. (2012) Gender Impact Analysis: Unconditional Cash Transfers in South Central
Somalia, Somalia Cash Consortium
WFP (2013) WFP Launches E-Cards For Syrian Refugees In Lebanon With Mastercard’s Support, 11
October (accessed 25 March 2022)
Wood, A.L.; Ansah, P.; Rivers III, L. and Ligmann-Zielinska, A. (2021) ‘Examining Climate Change and
Food Security in Ghana through an Intersectional Framework’, The Journal of Peasant Studies, 48(2):
329–348
World Vision (2018) Impact of Multi-purpose Cash Assistance on Child Labour among Syrian Refugee
Children in Bekaa, Lebanon
Yoshikawa, L. (2015) Integrating Cash Transfers into Gender-based Violence Programs in Jordan:
Benefits, Risks, and Challenges. Improving Cash-based Interventions Multipurpose Cash Grants and
Protection, London: UNHCR and International Rescue Committee
Zetter, R. and Ruaudel, H. (2018) ‘Refugees’ right to work and access to labour markets: constraints,
challenges and ways forward’, Forced Migration Review, (58): 4–7
26
Annex 1: Interventions and outcomes
Table A1: Types and examples of interventions
Types of interventions
Components/examples
Social transfers
1. Cash transfers (UCT, CCT, MPC, cash plus)
2. In-kind assistance/asset transfers
3. Vouchers
Public works programmes
1. Cash for work
2. Food for work
3. Voucher for work
Fee waivers
1. Exemption from health insurance/health service fees
2. Exemption from/reduction of school/school-related fees
Subsidies
1. Food subsidies
2. Fuel subsidies
27
Table A2: Types and indicators of outcomes
Types of outcomes
Indicators
Poverty and vulnerability
1. Household income or expenditure on essential items (rent, food, utilities,
transportation, etc.)
2. Food consumption/food security
Negative coping behaviour
1. Debts/borrowing
2. Child labour
3. Early/forced marriage (of children, especially daughters)
4. Diet restriction
5. Eating elsewhere
6. Selling of productive asset
Human capital
1. Education (enrolment, attendance, school performance)
2. Health (health status, health expenditure, health service use, access to health
services, physical & mental wellbeing, dietary diversity)
Livelihoods
1. Employment/job search
2. Job retention
3. Job-related skills
4. Savings/investment
Empowerment
1. Decision-making power and autonomy
2. Confidence, self-esteem, dignity
3. Gender roles
4. Change in attitude/norms
5. Intrahousehold dynamics
6. Domestic abuse/violence
7. Marriage/fertility/contraception
Social cohesion
1. Sense of belonging
2. Trust
3. Community engagement
4. Social relations/tension
5. Violence/harassment
There can be overlap between different indicators, depending on how they are defined. This table is meant as a guide for this review, not as a definitive
way of classifying the outcomes of social assistance in general. Social cohesion generally refers to refugee-host relationships, but it can also mean
relations within or between refugee communities.
28
Annex 2: Search strategy
Online searches were conducted of both academic and grey literatures, supplemented by citation tracking of
all relevant systematic reviews. The searches involved three academic databases, seven online research
databases/libraries, 11 government agency/donor/NGO/UN websites, and four systematic review databases.
Table A3: List of databases
Academic database
WOS, Scopus, ProQuest Dissertations and Theses Global
Online research databases
1. 3ie Impact Evaluation Repository
2. BRIDGE Global Resources
3. J-PAL Evaluations
4. OECD Working Paper Series
5. University of California Center for Effective Global Action
6. IDS Publications
7. Migration Research Hub
Government agency/donor/
NGO/UN websites
1. IOM publications/evaluations
2. Development (DFID) Research for Development Outputs
3. UNHCR research/publications/evaluations
4. UNICEF Data and Evaluation Database
5. UNFPA Evaluation Database
6. UNDP Evaluation Resource Centre
7. UN Women Gender Equality Evaluation Portal
8. USAID publications/evaluations
9. Overseas Development Institute (ODI)
10. World Bank Open Knowledge Repository (OKR)
11. OXFAM Policy and Practice
Systematic review databases
1. 3ie Systematic Review Database
2. EPPI Systematic Reviews
3. Cochrane Database of Systematic Reviews (CDRS) Cochrane
4. Campbell Collaboration
A set of keywords was selected and finalised after conducting a quick scoping to identify the different terms
used to describe the target populations (refugees and IDPs) and types of interventions in social assistance.
In academic databases, searches were run on title/abstract/keywords using Boolean operators and filters on
language and timeframe (year of publication). For non-academic databases and websites, similar keywords
were used but they were modified according to what the function permitted. For instance, some websites did
not allow combination of two or more terms but provided an option to search for studies/papers by region or
country. In that case, country names were inserted – one by one – and all studies/papers relevant to the four
countries of interest were screened. Citation tracking was performed on all relevant systematic reviews to
identify potential studies.
29
List of keywords:
1. (Refugees) OR (Internally Displaced Persons) OR (IDP*) OR (asylum) OR (migrants)
2. (Social protection) OR (social assistance) OR (social safety) OR (social transfer*) OR (public work*) OR
(fee waiver*) OR (fee-waiver*) OR (subsid*) OR (cash transfer*) OR (cash assistance) OR (voucher*) OR
(in-kind) OR (scholarship*) OR (stipend) OR (school feeding) OR (benefit*) OR (cash plus) OR
(graduation) OR (productive inclusion)
3. (Famil*) OR (household*) OR (kinship) OR (household dynamics) OR (gender)
4. (protracted) OR (emergency) OR (conflict*) OR (fragile) OR (crisis) OR (humanitarian)
5. (Lebanon) OR (Syria) OR (Jordan) OR (Iraq) OR (Arab) OR (middle east*)
6. Search results were exported to Endnote software and copied to Microsoft Excel. After removing
duplicates, title and abstract screening was performed. Where abstracts did not give a clear indication of
whether to include or exclude studies/papers, full texts were retrieved. This was followed by full-text
screening. For a small number of papers that could not be accessed, the authors were contacted.
Annex 3: Critical appraisal
The Joanna Briggs Institute (JBI) critical appraisal checklists were used to assess the included studies.
These tools were developed by the Faculty of Health and Medical Sciences at the University of Adelaide. The
JBI critical appraisal checklists cover a wide range of study designs, and comprise a series of questions,
each with four response options: ‘yes’, ‘no’, ‘unclear’ and ‘not applicable.’ For each question, one point was
allocated for ‘yes’ and no points were allocated for ‘no’ and ‘unclear’.
For RCTs, there are 13 items on the JBI checklist. We categorised quality as ‘high’ (11–13 points), ‘moderate’
(7–10) and ‘low’ (less than 7). For non-randomised trials (e.g. quasi-experimental studies), there are nine
items, therefore we categorised a paper’s quality as ‘high’ (8–9 points), ‘moderate’ (5–7) or ‘low’ (less than 5).
For cross-sectional design (analytical studies), there are eight items, so quality was categorised as ‘high’ (7–8
points), ‘moderate’, (5–6) or ‘low’ (less than 5).
For appraisal of qualitative studies, three items from the original JBI checklist were slightly modified, but
retained the total number of items (n=10). We categorised quality as ‘high’ (8–10 points), ‘moderate’ (5–7) or
‘low’ (less than 5). For mixed-methods studies/evaluations, appraisal was conducted separately; JBI tools for
quantitative (based on the specific design) and qualitative research were used to assess the quantitative and
qualitative components of the same study, respectively.
It is important to note here that the quality ascribed to a study was based on the best design it could offer and
may not be appropriate for comparison across study designs. For example, a well-designed case-control
study could be categorised as ‘high quality’, while a poorly designed RCT could be categorised as ‘low
quality.’ However, that does not imply that the evidence derived from the case-control study is more robust
than that derived from the RCT, given that the former is inferior to the latter in the hierarchy of evidence and it
cannot infer causality.
The scoring system applied was not based on JBI guidelines, rather it was employed to simplify the overall
judgment process. The complete checklists (and results) of quality appraisal are available in Appendix 1.
30
Annex 4: Data extraction and coding tool
Data were extracted using a piloted form that captured the following information: author, year of publication,
study design, population, setting (country), study objectives, type of intervention (including amount/value,
duration and mode of delivery), types of outcomes and main findings. The coding tool used for data
extraction included items illustrated in Table A4.
Table A4: Coding tool for data extraction
Item
Findings
Date of extraction:
Author (year):
Study design
● Quantitative/qualitative/mixed methods
● Quantitative: which type?
Population
● Refugees/IDPs
● Vulnerable households among host communities
● Any specific groups (women, children, etc.)?
Setting/country
● Iraq/Jordan/Lebanon/Syria
● Can be multiple countries
Types of intervention(s)
● Social transfers – which type?
● PWP
● Fee waivers
● Subsidies
Description of intervention
● Value/amount
● Duration
● Mode of delivery
Types of outcome(s)
● One or more outcomes as listed in the table under inclusion/exclusion criteria
Main findings
(relevant to outcomes of interest)
● Quantitative – impact/effectiveness
● Qualitative – provide description
Effect sizes
● For quantitative studies
Notes/other findings
● Other important points to highlight (e.g. major limitation, etc.)?
Quality appraisal
● High/moderate/low
Sex-disaggregated data
● Yes/no
Impact on gender, familial and
● Describe
intrahousehold relations and dynamics
31
Annex 5: Overall results/findings
Tables A5, A6 and A7 show the findings of all studies analysed in this review.
Table A5: Findings of quantitative and mixed-methods studies (n=13)
Author
(year)
Study design
Sample/
population
Types of social
assistance
Sex-disaggregated Intersectional data
data
(disability, age,
religious
background,
sexual orientation)
Level of analysis
Findings (impact/effectiveness)
(individual,
household, family,
community,
province, state)
Falb et al. Mixed methods: 456 women
pre-test/post(2020)
(displaced
test and
Syrians from
interviews
Raqqa City,
Syria)
UCTs of USD76/month None (subjects were Disability mentioned Individual and
at demographic level household levels
for 3 months
all women).
Chaaban Regression
et al.
discontinuity
(2020)
design (RDD)
MPC of USD173.5/
month for 3 varying
periods:
11,457
households
Significant reductions in food
insecurity
No change in serious household
needs and daily stressors
Increases in symptoms of depression
(Syrian refugees
in Lebanon)
<12 months
12 months
Described for one of Partially described Individual and
for age and disability household levels
the outcomes
(impact of social
assistance on
employment among
males vs females)
Lower food insecurity with longerduration MPC
No impact on household rent
expenditure, but improved access to
drinking water
Increased school enrolment
>12 months
Positive impact on job searches for
men
Improved mental health and access
to primary health care
Lehmann RDD
and
Masterson
(2020)
1,300
households
(Syrian refugees
in Lebanon)
Cash transfer (CT) of
USD100/month for 6
months using ATM
cards (during winter)
None
None
Individual,
household and
community levels
Cash transfers did not increase
hostility towards refugees
32
Author
(year)
Study design
Lehmann RDD
and
Masterson
(2014)
Sample/
population
Types of social
assistance
Sex-disaggregated Intersectional data
data
(disability, age,
religious
background,
sexual orientation)
1,360 Syrian
refugee
households in
Lebanon
CT of USD107/month None
for 4 months, preceded
by USD147 in the first
month – using ATM
cards
Level of analysis
Findings (impact/effectiveness)
(individual,
household, family,
community,
province, state)
Described at
Household level
demographic level
for age and disability
Increased household expenditure on
basic items (not enough to meet
winter needs)
No impact on debt reduction, but
decreased intrahousehold disputes
(Nov.–Mar., during
winter)
Reduced child labour
Increased access to school
Improved relations with community
members
Caria
et al.
(2020)
Adaptive
targeted field
experiment
De Hoop Mixed methods
et al.
Quantitative:
(2018)
geographic
RDD
Qualitative:
key informant
(KII) interview
and FGD
One-off labelled CT of
USD92 using ATM
(Syrian refugees cards
and Jordanians)
in 3 cities:
Amman, Irbid,
Mafraq
3,770 subjects
1,440
households,
targeting Syrian
refugee children
in Lebanon
None
Described based on Individual level
nationality (Syrians
vs Jordanians)
CT that lasted for the
Described for
Described according Individual and
duration of school year, several outcomes
to age groups
household levels
using ATM cards
including health,
time use,
Value: USD20/month
psychosocial
(younger children),
wellbeing and
USD45/month (older
education (sexchildren)
disaggregated data
in Appendix E)
Subsequently, value
decreased to
USD13.50–USD20/
month
Significant impacts on refugee
employment and earnings, 2 and 4
months after treatment
Increase in job retention and hourly
wages
Positive impact on children’s food
consumption
Improved physical health for younger
children (5–9 years)
Reduction in child work for older
children (10–14 years)
Increased education-related
expenses
Little/no impact on economic
wellbeing and food consumption at
household level
33
Author
(year)
Study design
Phadera, Propensity
Sharma score matching
and WaiPoi (2020)
Sample/
population
Types of social
assistance
Sex-disaggregated Intersectional data
data
(disability, age,
religious
background,
sexual orientation)
8,615
households in
Iraq (including
IDPs)
Universal food subsidy None
programme
None
Level of analysis
Findings (impact/effectiveness)
(individual,
household, family,
community,
province, state)
Household level
Improved food security
Decreased odds of poverty
Increased total food and non-food
consumption
No/little impact on subjective
wellbeing
Increased daily calorie intake
Abu
Hamad
et al.
(2017)
Mixed methods 2,114
households
Quantitative:
(survey), 432
cross-sectional
persons in
(survey)
qualitative phase
Qualitative: KII (most were
Syrian refugees
and FGD
in Jordan)
Described based on Individual and
CTs and food vouchers Described for
various outcomes
age and disability for household levels
CT
including educational some outcomes
expenditures, control
1) UNHCR: USD86–
over spending,
USD455/month
school enrolment,
2) UNICEF CCG:
social participation
USD28/child or max.
and psychosocial
USD106/family
wellbeing
WFP voucher – value
was USD10–20/month
Little/no impact on employment
Decreased harmful coping strategies
Improved access to housing
Enhanced food consumption
Improved spending on schooling and
academic performance, but no
impact on enrolment
Increased spending on child health,
but not on adult health
Improved intrahousehold
relationships, but limited/no impact on
social participation
34
Author
(year)
Study design
Sample/
population
Battistin
(2016)
RDD
508 subjects
Types of social
assistance
Sex-disaggregated Intersectional data
data
(disability, age,
religious
background,
sexual orientation)
MCA of USD174 using Partially described None
(distinctions made
ATM cards
(261 treatment vs
between households
247 control)
headed by males vs
females)
Syrian refugees
in Lebanon
Level of analysis
Findings (impact/effectiveness)
(individual,
household, family,
community,
province, state)
Household level
Increased household expenditures
(mainly for food and gas)
No impact on dietary diversity
Decreased borrowing (of food) and
rental debt
Increased psychosocial wellbeing,
sense of security and trust
Increased happiness but also more
financially stressed
No impact on child labour or school
enrolment
More likely to rely on work, as
opposed to negative coping
strategies such as debt, remittances,
gifts and sale of assets or food
World
Vision
Lebanon
(2018)
Quasiexperimental
study
Boston
RCT
Consultin
g Group
(2017)
360 Syrian
children in
Lebanon
MPC of USD175/month None
for at least 3 months
Children divided into Individual and
3 age groups
household levels
No impact on child labour, school
enrolment and participation in house
chores
Reduced workplace risk/increased
sense of safety
1,848
households in
Jordan and 1,275
households in
Lebanon (all
were Syrian
refugees)
CT vs food-restricted
value vouchers and
choice for 4–8 months
Cash value was
USD28/person/month
Partially described
(for food spending
decisions and
household
responsibilities)
None
Household and
community levels
(effects of
programmes on
local economy)
Cash leads to food security that is
superior or equal to food vouchers
Cash is more effective in challenging
contexts (when food security is lower)
Cash enables greater purchasing
power and choices of food
No difference in expenditure on other
needs
35
Author
(year)
Study design
Sample/
population
Types of social
assistance
Sex-disaggregated Intersectional data
data
(disability, age,
religious
background,
sexual orientation)
Level of analysis
Findings (impact/effectiveness)
(individual,
household, family,
community,
province, state)
Doocy
et al.
(2017)
Pre-test/posttest study
981 vulnerable
households in
Idleb, Syria
(including IDPs)
In-kind food
commodities, food
vouchers, and
unrestricted vouchers
None
Household and
community levels
(effects on local
market prices)
Oxfam
(2018)
Quasiexperimental
study (PSM)
None
In-kind food was more effective in
improving household food security
and access to food compared to food
vouchers or unrestricted vouchers
Value of unrestricted
vouchers was
USD70.20/month for 11
months
Both in-kind food and food vouchers
were more effective in improving food
consumption score than unrestricted
vouchers.
Food voucher was
USD42.50–159/month
for 24 months
Largest reduction in negative coping
strategies was among in-kind food
recipients
1,136 Syrian
Cash for work
refugees living in programme
Za’atari camp,
Semi-skilled workers:
Jordan
USD1.50/hour
Skilled to technical
workers: USD1.50–
2.50/hour
None (but opinions None
on gender equality
and equal access to
work were
measured)
Individual and
household levels
Increased household income/
expenditure
Improved self-esteem (more for semiskilled workers)
No impact on perceptions/attitudes
towards gender roles
No impact on new skills acquisition
36
Table A6: Qualitative studies that explored the effects of social assistance (n=9)
Author
(year)
Tool and
sample
Types of social
assistance
Blackwell In-depth
UCTs of USD76/month
interviews (IDIs) for 3 months
et al.
(2019)
with 40 women
(displaced
Syrians from
Raqqa city in
Syria)
SexIntersectional
disaggregated data (disability,
data
age, religious
background,
sexual
orientation)
None (all
subjects were
women)
Described
according to
marital status
Level of analysis: Findings
individual,
household,
family,
community,
province, state
Individual,
household and
community levels
Mixed impacts – both positive and negative
Individual level: temporary relief from stress; increased
confidence and self-efficacy
Household level: increase in joint decision-making; no change
in relationship with husband; improved relationship with
extended family; worsened relationship with mother-in-law;
stopped getting financial support from family
Community level: worsened tension with community;
community jealousy; stopped getting financial support from
community; increased social capital from reducing debts;
women able to stop working thus reducing community stigma
HagenZanker
et al.
(2018)
IDIs and FGDs
involving 140
Syrian refugees
in Jordan
UCTs of USD110–220/
month for at least 1 year
using ATM card
None
None
Individual and
household levels
Respondents reported:
● Improved access to shelter/rent and utility
● Reduction in stress and anxiety
● No significant influence on health care and education
● Little/no influence on access to employment
● Greater tendency to keep children in school (instead of work)
Samuels
et al.
(2020)
IDIs and FGDs MPC of USD332/
Described in a
household/month for 12 narrative
involving 254
Syrian refugees months using e-card/ATM manner
and Lebanese
(host community
members)
Age (disability
Individual,
status mentioned household and
as a
community levels
consideration
while capturing
response)
Mixed results on impact on interactions and social cohesion:
● Effects are least/minimal on Syrian refugees living in informal
tented settlement and Syrian women
● Respondents reported improved trust and relationships with
Lebanese landlords and increased access to health care
37
Author
(year)
Tool and
sample
Types of social
assistance
SexIntersectional
disaggregated data (disability,
data
age, religious
background,
sexual
orientation)
Bastagli IDIs and FGDs MPC of USD332/
Described in a Age
household/month for 12 narrative
et al. ODI involving 254
Syrian refugees months using e-card/ATM manner
(2021)
and Lebanese
(host community
members)
Level of analysis: Findings
individual,
household,
family,
community,
province, state
Individual,
household and
community levels
MPC shapes the outcomes of interest in the following ways:
● Decreased pressure to borrow money and enabled debt
repayment
● Facilitated school attendance
● Mixed findings on child work/labour
● Decreased early marriages
● Reduced intrahousehold tension and violence
● Empowered women/female recipients
● Enabled recipients to avoid exploitative work or work less in
poor working conditions
Pertek
(2016)
FGDs and IDIs
involving 26
Syrians
(refugees) and
Jordanians
CCTs (amount and
duration not specified)
Described in a None
narrative
manner
Individuals and
household levels
● Shift in traditional gender roles – women became more selfreliant and independent
● Women reported feeling burdened by work overload
● Women became more vocal and their opinions were sought,
thus increasing joint decisions
● Improved family cohesion and relationships
Yoshikaw In-depth
a (2015) interviews, KIIs
and FGDs
involving 72
Syrian refugees
in Jordan
Cash plus (UCTs and
psychosocial support)
UCT value: USD170–
254/household/month for
6 months
Described in a None
narrative
manner
Individual and
household levels
● Reduced household tension and domestic violence
● Women felt more empowered, respected and valued
● Change in intrahousehold dynamics and power relations
● Improved psychosocial wellbeing for recipients and family
● Improved relations with host community
● Impact on resilience and domestic violence more significant
when cash combined with other services, rather than cash
alone
38
Author
(year)
Tool and
sample
Types of social
assistance
SexIntersectional
disaggregated data (disability,
data
age, religious
background,
sexual
orientation)
Sloane
(2014)
IDIs with 8 Syrian UCTs for 3 months, using Described in a
refugee families ATM card
narrative
in Jordan
manner
In Balqa: JOD80 +
JOD15/family member/
month (max. JOD185)
In informal settlements in
Amman: JOD20/
member/month
None (but
selection of
subjects was
based on levels
of vulnerability
including health
status and
disability)
Level of analysis: Findings
individual,
household,
family,
community,
province, state
Household level
and community
levels
IDIs and FGDs
Population:
Syrian refugees
in Jordan and
vulnerable
Jordanians
Berg et al. KIIs and FGDs
(2013)
involving 124
participants in
Jordan
Cash for work (CFW)
programme
● Increased food consumption and diversity
● Increased health care expenditure for children
● Limited impact on debt repayment
● Unclear impact on child labour
● Decline in intrahousehold tension
(max. JOD140)
Loewe
(2020)
● Moderate positive effects on expenditure and standard of
living
● Potential increase in tension with host community
Described in a None
narrative
manner and
quantitatively
Individual,
CFW affects social cohesion and economic opportunities by:
community and
municipality (local ● Increasing sense of belonging and horizontal trust among
programme participants (especially among Syrians toward
government) levels
Jordanians)
● Influencing acceptability towards female labour participation
● Improving skills, attitude and networking
UCTs and vouchers
For Iraqis: USD105–420/
family using ATM
For Syrians: USD70–168/
family using ATM
None
None
Individual and
community levels
● Increased sense of dignity, but no other impacts on
empowerment
● Did not cause jealousy/increase in tension with host
community
● Increase in house rent (by some landlords)
39
Table A7: Findings on the effects of social assistance on empowerment or those related to
gender, familial and intrahousehold dynamics.
Author
(year)
Study design,
country
Lehmann and RDD, Lebanon
Masterson
(2014)
Abu Hamad
et al. (2017)
CTs improved relationships within households. The number of disputes between
household members in the treatment group was significantly lower than that of control
group. However, cash beneficiaries did not report better subjective wellbeing
compared to non-beneficiaries.
Mixed methods, Cash assistance improved intrahousehold relationships by reducing poverty-related
Jordan
stress but was too low in value to allow participation in social or recreational events
and activities. Adolescents reported greater improvement in peer interactions than
adults because of cash – largely because of improved access to pocket money.
Oxfam (2018) PSM, Jordan
Blackwell
et al. (2019)
Findings
CFW programme was found to have no impact on participants’ opinion of/belief about
gender equality.
Qualitative, Syria Impact at individual level: temporary relief from stress; restored sense of dignity;
increase in self-confidence and self-efficacy; sustained stress over future.
Impact at household level: increase in joint and individual decision-making; unchanged
or improved relationship with husband; improved relationship with extended family;
worsened relationship with mother-in-law; cessation of financial support from family.
Impact at community level: worsened tension with community and jealousy among
neighbours; cessation of financial support from community; increased social capital
from reducing debts; women able to stop working thus reducing community stigma.
Yoshikawa
(2015)
Qualitative,
Jordan
Impact of cash plus (CT + GBV programme): most women receiving CT spoke of
feeling strong, confident, respected, and independent as a result of bringing resources
into the household. Additional GBV initiatives made women felt more knowledgeable
and skilful. Men participated less in psychosocial activities due to perceived stigma or
embarrassment. Some women reported that the programme increased openness to
dialogue among their spouses on how to spend household finances, allowing the
women greater control of the household finances. However, this change in dynamics
did not last beyond the period of assistance. Women perceived CT to reduce
household tensions, as well as domestic violence against women and between
parents and children. Counsellors and cash assistance officers corroborated this link.
In contrast, only a few men drew the link between domestic violence and financial
problems. Some male participants expressed negative views about women receiving
CT, while other interviewees said CT reduced immediate risks of abuse and
exploitation, by allowing them to move away from a landlord or avoid going to an aid
agency, where a staff member made ‘unreasonable’ demands.
Loewe (2020) Qualitative,
Jordan
CFW facilitates employment for women and contributes to increasing acceptability of
female labour participation. Most participants showed a positive opinion/attitude
towards female participation, but reaction among non-participants was mixed.
Berg et al.
(2013)
Qualitative,
Jordan
Mixed effects: some recipients felt more dignified, while others reported no difference
due to the small amount and that CT could not solve other key issues and their future
uncertainty.
Bastagli
(2020)
Qualitative,
Lebanon
Reduced intrahousehold tension; female recipients reported feeling more empowered.
Pertek (2016) Qualitative,
Jordan
Women became more vocal, self-reliant and independent. They also found that their
opinions were more sought after, thus increasing joint decisions in the household.
Family cohesion and relationships improved, but some women felt burdened by work
overload.
Sloane
(2014)
Receiving social assistance reduced intrahousehold tension.
Qualitative,
Jordan
40
Acknowledgements and Disclaimer
This document was developed by the Better Assistance in Crises (BASIC) Research programme. BASIC is implemented by
the Institute of Development Studies (IDS), the University of Sussex and the Centre for International Development and
Training, funded by UKAid from the UK government. The views expressed in this document are entirely those of the authors
and do not necessarily represent views or policies of the UK governments official policies.
© IDS copyright 2022. Copyright in the typographical arrangement and design rests with IDS.
This publication (excluding the logos) may be reproduced free of charge in any format or medium, provided that it is reproduced
accurately and not used in a misleading context. The material must be acknowledged as IDS copyright with the title and source
of the publication specified.
Published by IDS.
Suggested citation
Oosterhoff, P. and Yunus, R.M. (2022) The Effects of Social Assistance Interventions on Gender, Familial, and Household
Relations Among Refugees and Displaced Populations: A Review of the Literature on Interventions in Syria, Iraq, Jordan, and
Lebanon, BASIC Research Working Paper 11, Brighton: Institute of Development Studies, DOI: 10.19088/BASIC.2022.011
41