Content uploaded by Pius Gumisiriza
Author content
All content in this area was uploaded by Pius Gumisiriza on Jan 18, 2021
Content may be subject to copyright.
Content uploaded by Pius Gumisiriza
Author content
All content in this area was uploaded by Pius Gumisiriza on Jan 18, 2021
Content may be subject to copyright.
Identity at the Margins: Examining Refugee Experiences with
Digital Identity Systems in Lebanon, Jordan, and Uganda
Emrys Shoemaker
Caribou Digital
emrys@cariboudigital.net
Gudrun Svava Kristinsdottir
Cornell Tech
gsk72@cornell.edu
Tanuj Ahuja
Cornell Tech
ta364@cornell.edu
Dina Baslan
Independent Researcher
dina.baslan@gmail.com
Bryan Pon
Caribou Digital
bryan@cariboudigital.net
Paul Currion
Caribou Digital
paulcurrion@gmail.com
Pius Gumisizira
Caribou Digital
piusgumisiriza@yahoo.com
Nicola Dell
Cornell Tech
nixdell@cornell.edu
ABSTRACT
This paper examines refugees’ experiences with and perspectives on
the digital identity systems used by humanitarian organizations to
collect, manage, and share their personal data. Through a qualitative
study with 198 refugees in Lebanon, Jordan, and Uganda, we show
how existing humanitarian identity systems present numerous
challenges for refugees. For example, we nd that refugees have
little to no knowledge of the institutional systems and processes
through which their personal data is managed and used. In addition,
refugees are typically not able to exercise agency with regard to
data that is collected about them (e.g., given choices about the
data collected). At the same time, we show how refugees make
active eorts to negotiate the various identities available to them,
consciously weighing the benets and constraints associated with
dierent statuses to maximize their access to services, eligibility for
employment, and spatial mobility. We use Bardzell’s lens of feminist
interaction design [
8
] to make sense of these ndings and suggest
a path forward that engages refugees in the design of improved
identity management systems.
KEYWORDS
HCI4D; ICTD; refugees; identity; displaced populations.
ACM Reference format:
Emrys Shoemaker, Gudrun Svava Kristinsdottir, Tanuj Ahuja, Dina Baslan,
Bryan Pon, Paul Currion, Pius Gumisizira, and Nicola Dell. 2019. Identity at
the Margins: Examining Refugee Experiences with Digital Identity Systems
in Lebanon, Jordan, and Uganda. In Proceedings of ACM SIGCAS Confer-
ence on Computing and Sustainable Societies, Accra, Ghana, July 3–5, 2019
(COMPASS ’19), 12 pages.
https://doi.org/10.1145/3314344.3332486
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for prot or commercial advantage and that copies bear this notice and the full citation
on the rst page. Copyrights for components of this work owned by others than the
author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specic permission
and/or a fee. Request permissions from permissions@acm.org.
COMPASS ’19, July 3–5, 2019, Accra, Ghana
©
2019 Copyright held by the owner/author(s). Publication rights licensed to Associa-
tion for Computing Machinery.
ACM ISBN 978-1-4503-6714-1/19/07.. .$15.00
https://doi.org/10.1145/3314344.3332486
1 INTRODUCTION
Humanitarian organizations that provide services to refugees across
the world are embracing increasingly sophisticated digital systems
for managing the identities and personal data of the beneciaries
they serve. These digital identity management systems have the po-
tential to provide organizations with simplied reporting, reduction
in fraud, improved service delivery, and increased convenience for
both organizations and refugees. However, identity and personal in-
formation are intrinsically powerful and sensitive topics and, given
the vulnerable nature of the refugees being served, the complexities
and risks associated with these large-scale systems are amplied.
Navigating these complexities requires us to understand the expe-
riences of the individuals that are subjected to these systems, and
balance the operational benets with an empathetic assessment of
how these systems impact individuals’ privacy, dignity, and agency.
This paper contributes a qualitative study examining refugees’
experiences with, and perspectives on, the digital identity systems
that humanitarian organizations use to collect, manage, and share
their personal data. We conducted semi-structured interviews and
focus groups with 198 refugees, in and out of camps, in Lebanon,
Jordan, and Uganda in 2018. We describe the various digital identity
systems that refugees encounter as they work to establish identities
that grant them legal refugee status and entitlement to services
(e.g., food, shelter, health services). We also uncover numerous
challenges that result as individuals struggle to navigate the systems
that organizations use to collect their personal information and
biometric data (e.g., ngerprints and iris scan).
For example, we show how individuals have little to no knowl-
edge or visibility into the institutional processes through which
their personal data is managed and shared, including which organi-
zations have access to their data. This lack of transparency often
results in confusion, disappointment, and anxiety about the safety
and privacy of their information. Despite these concerns, refugees
are desperate for assistance and often have no choice but to provide
organizations with all of the information that they request.
In addition, refugees are rarely oered the opportunity to exer-
cise agency with regards to data that is collected on them (e.g., given
choices about what data is collected or how it is shared). Never-
theless, many refugees make active eorts to negotiate the various
COMPASS ’19, July 3–5, 2019, Accra, Ghana E. Schoemaker et al.
identities available to them, consciously weighing the benets and
constraints associated with dierent statuses (such as registered vs.
unregistered) in eorts to maximize their access to services.
Moreover, despite acknowledging that they do not know the
criteria that organizations use to assess their vulnerability and eli-
gibility for assistance, we nd that many refugees actively work to
redene their family structure in ways they perceive will maximize
the services they receive, such as selectively registering only the
woman and children. However, such actions can impact household
power dynamics, since women registered as heads of households
are given control over the family’s resources (e.g., food and shelter)
in communities where traditionally men are the decision-makers.
We use Bardzell’s lens of feminist HCI [
8
] to further analyze and
make sense of our ndings. For example, we discuss how current
identity management systems target a “universal” refugee and fail
to account for dierent points of view or partial perspectives that
may exist in dierent contexts. As a result, individual refugees
are unable to negotiate exibility in the nature of their identity or
their eligibility for services. This further results in refugees striving
to (re)dene themselves in ways that better t existing identity
systems, especially if it may result in additional benets.
We discuss how challenges with existing identity management
systems impact the broader ecologies in which they are deployed,
such as how diculties obtaining and renewing identities restrict
refugees’ mobility or access to basic services essential for survival.
We also highlight important information and power asymmetries
between refugees and the organizations tasked with advocating for
them, with organizations possessing a signicant amount of power
over refugees, who are desperate and have no choice but to comply
with any demands made of them by organizations. In light of these
insights, we suggest a starting point for engaging refugees in the
design of improved identity management systems, which may lead
to benets for both humanitarian organizations and refugees.
2 RELATED WORK
The concept of identity and its importance in the social, cultural,
and historical lives of humans has been discussed at length in psy-
chology and the social sciences (e.g., [
29
,
31
,
34
,
53
,
54
]), and has
been shown to have direct relationships to social, cultural, and his-
torial contexts [
9
]. Theoretically, ‘identity’ refers to social category,
dened by membership rules and socially distinguishable features
that a person takes a special pride in, or views as unchangeable
but socially consequential [
20
]. The theoretical concept of iden-
tity has also been critiqued for being ‘asked to do too much’ [
12
]
and as a result may have limited analytic utility. Brubaker and
Cooper [
12
] argue for more specicity and application of the con-
cept, while Whitley et al. [
54
] argue for greater attention to the
interplay between the distinct phenomena of information systems,
identication, and identity. In our work, we focus on the identi-
cation of individuals as refugees by humanitarian organizations
and institutions, whilst being sensitive to the broader concept of
identity as a theoretical social category.
Given the strong inuence identity has in dening social and
cultural milieus, it becomes especially crucial to study identity in
the context of refugees and displaced populations. Although there
have been abstract mentions in the literature of the linkage between
identity and refugees [
37
,
38
], there is little literature available that
discusses the changing nature of refugee identities in the digital age.
With the global proliferation of digital technologies, digital identity
is now fundamental in dening a refugee’s status, especially in
camps. Although prior research has explored digital identity in
general (e.g., [
3
,
30
,
33
]), very few empirical studies specically
focus on digital identities in the context of refugees.
Our research also contributes to a growing body of work examin-
ing the role of the computing community in responding to refugee
crises [
2
,
41
,
43
,
52
]. For example, shelter dynamics is an important
topic that is closely related to our work on identity and registration
systems, since shelters are often assigned using information pro-
vided to these systems. Sabie et al. [
35
] used ethnographic methods
in Syrian and Iraqi camps to explore cultural and social dynamics
around shelter. Building on this, Nabil et al. [
32
] did eldwork in
Jordan that shows how decorating shelters provides an escape from
the camp and compensates for loss of identity, home, and leisure.
Connection and community-building within refugee communi-
ties has also received recent attention in the literature. For example,
Almohamed and Vyas [
4
] use ethnographic research to identify
themes around social isolation, cultural backdrops, and the role of
technology in the lives of refugees and asylum seekers. Aal et al.
expose challenges with physical and social infrastructure through
analyzing a refugee camp computer club [
1
], while Schmitt et al.
examined community-level physical access divides in camps [
36
].
Xu et al. conducted a study to promote participatory community
building via mapping technologies [
58
], and further discussed data-
driven community development in follow-up work [57].
Other research highlights vulnerabilities associated with reset-
tlement [
5
] and the role of digital technologies in helping refugees
to integrate into new environments and attain independence [
7
,
26
].
In addition, work has examined the security threats and vulnera-
bilities associated with refugees in the digital tech space [
13
,
16
], a
topic that our work touches on as well. A variety of strategies have
also been used to help refugees rebuild social capital and a sense of
community in displaced environments [
6
], including community
radio [27], co-located social media platforms to help refugees nd
their voices [
55
], human-in-the-loop messaging to reduce language
barriers [
11
], leveraging social ties to bridge the social and economic
divide by forming computer clubs in Palestinian camps [
61
], and
utilizing the importance of food in camps to develop social capital
[
21
]. This research is orthogonal to our work, which focuses more
on how refugees interact with organizations and the digital systems
that organizations use to make refugees legible to the organization.
Another set of papers have worked to understand refugees’ ac-
cess to or usage of health services in camps. Talhouk et al. [
42
] used
focus groups to identify contextual and cultural factors that can
inform the design of digital technologies to support refugee ante-
natal care. Carrying this forward, they then explored the concept
of refugee-led community radio shows to deliver health informa-
tion and discussed community dynamics in utilizing technology
for health [
40
]. Focusing more directly on technology access and
use, research has explored refugees’ use of the Internet and mobile
communication devices, both in adult populations [
56
] and with
youth [
60
]. Ya et al. studied how Syrian youth at the Za’atari Camp
in Jordan use information and mobile technologies in their daily
Identity at the Margins COMPASS ’19, July 3–5, 2019, Accra, Ghana
lives [
59
]. Fisher et al. conducted participatory design workshops
at a Syrian refugee camp in Jordan to learn how youth help others
to use technology [
22
]. Boulus-Rødje et al. worked with children
in Palestinian refugee camps and explored issues around digital
fabrication in developmental and educational settings [39].
Our research contributes to this literature with the rst study
of refugees experiences with, and perspectives on, the digital sys-
tems that humanitarian service providers use to render people
legible to their organizations. We highlight numerous challenges
that refugees navigate as they struggle to maximize the services
that they are entitled to, discussing how information and power
asymmetries negatively impact refugees’ ability to exercise agency
and control over their personal information and identities.
3 RESEARCH METHODS
Research Sites:
We conducted our IRB-approved study with 198
refugees in Lebanon, Jordan and Uganda, in and out of camps, in
2018. We chose Lebanon and Jordan because they have absorbed
one of the refugee movements as people have ed Syria, yet oer
dierences in terms of identity systems in use. As of December 2017,
Lebanon had 1,018,057 registered refugees and asylum-seekers,
predominantly in the Bekaa Valley but also in urban and peri-urban
areas around Beirut [
47
]. We spoke with refugees in the Bekaa
Valley and in central and peri-urban Beirut. Lebanon is distinct
because most refugees are in informal tented settlements, and the
registration and data management systems have reected this.
In 2018 Jordan had the second highest share of refugees compared
to its population in the world, 89 refugees per 1,000 inhabitants [
46
].
Although the majority of Syrian refugees in Jordan live in urban
areas, 126,490 refugees live in camps, with Zaatari and Azraq, the
two largest, hosting 78,554 and 41,089 respectively [
50
]. We spoke
with refugees in both these camps and in the capital, Amman, in
February 2018. Jordan is distinct for its highly developed camps
and the use of advanced biometric technologies.
As of 28 February 2018 there were 1,444,873 refugees and asy-
lum seekers in Uganda, with 21% in Yumbe, the location of Bidi
Bidi camp [
51
]. We spoke with refugees in Bidi Bidi in March 2018.
Uganda is distinct because the government, through its Oce of
the Prime Minister (OPM) [
44
], has taken the lead in registering
and managing refugee data. It also has one of the most progressive
policies towards refugees, granting them legal recognition and giv-
ing each refugee household ownership of a small parcel of land.
Recruitment:
Participants were recruited through a combination
of invitation by partner organizations, personal networks, and
snowball sampling [
23
]. Save the Children, an international NGO,
provided logistical and initial recruitment support, inviting ben-
eciaries to participate. We also drew on our research assistants’
personal networks to access community-based organizations active
in supporting refugees. At all sites we used snowball sampling, [
23
]
asking participants to introduce us to people they knew.
Qualitative Methods:
We used a combination of semi-structured
interviews and focus groups to engage with participants (see Table
1). Our question guides were structured around topics identied
in the literature, stakeholder interviews, and pilot research. We
All Lebanon Jordan Uganda
Total participants 198 65 58 75
Men 86 27 16 43
Women 112 38 42 32
Interviews 39 9 4 26
Focus groups 23 9 8 6
Table 1: Summary of participants and research methods
sought an understanding of: refugees’ experiences using identity
credentials before displacement; their rst experience of registra-
tion and using new identity credentials; using refugee credentials
to obtain services; everyday experiences of proving who you are;
and managing dierent identity credentials. We also probed partici-
pants’ understanding of how organizations manage their data, and
the barriers and workarounds used to maximize access to services.
Interviews and focus groups took place in a mix of locations,
including NGO oces in camps, homes, food distribution points,
and shops. We began by explaining the purpose of our research
and obtaining the participant’s consent. Sessions lasted approxi-
mately one hour and were conducted in the participant’s choice of
English or Arabic. The majority of participants in Lebanon and Jor-
dan (approx. 80%) chose Arabic, while participants in Uganda were
more evenly split between both English and Arabic. Sessions were
audio recorded with permission from participants. In a few cases,
participants declined to be audio recorded and so instead we took
detailed handwritten notes. Participants were not compensated.
Participants:
Table 1 provides some demographic information
about our participants at each research site. In total, we spoke with
198 refugees: 86 men and 112 women. Almost all of our participants
had access to a mobile phone, either owning their own device or
sharing a device with family members and/or friends. The most
common applications that participants reported using were Face-
book and WhatsApp, which participants ubiquitously described
using to keep in contact with family members or friends, both in
the host country and in other countries (e.g., back in their home
countries). In addition, mobile money applications were commonly
mentioned as being useful for sending and receiving money.
Data Analysis:
Audio recordings were professionally translated
into English (if necessary) and transcribed. We then analyzed the
transcripts and our copious eld notes thematically [
10
], beginning
with a comprehensive reading of the transcripts during which we
identied initial codes. We performed multiple passes over the data
to iteratively rene codes and ensure they accurately represented
the data. Coding was performed by two research assistants with
input, supervision, and discussion with the broader group after each
iteration. Our nal set of 57 codes were formalized in a codebook
that was used to perform a detailed analysis of all transcripts. Exam-
ples of codes include biometric data,UN card renewal, and political
persecution. We clustered related codes into high-level themes that
represent our prominent ndings. Examples of themes include pri-
vacy concerns,challenges with registration, and lack of transparency.
COMPASS ’19, July 3–5, 2019, Accra, Ghana E. Schoemaker et al.
We emphasize that the nature of our data is qualitative, not quan-
titative, so we do not report on raw numbers of participants who
made certain statements in our data and analysis.
4 ESTABLISHING DIFFERENT IDENTITIES
When individuals ee their own country and enter that of a host
government, they also enter a network of organizations and sys-
tems that are part of the humanitarian response. The organizations
and systems that make up the humanitarian response labels and
classies people in order to make them legible to the systems em-
ployed by organizations to manage the provision of services. The
labels and classications are used to organize and ration the provi-
sion of scarce resources, according to diverse and complex criteria.
Some categories, such as specic nationalities, genders or ages are
categorized as priorities for the provision of services, and dierent
organizations have dierent priorities. For example, UNHCR, the
UN’s Refugee Agency, serves all refugees and has a legal mandate to
recognize refugee status. By contrast, UNICEF, the United Nations
Children’s Fund, prioritizes young people, while the priorities of
many non-governmental organizations (NGOs) are set by donors.
As organizations interact with individual refugees, they create
records and issue credentials in order to identify and authenticate
that people are who they say they are. Through the journey of
interacting with these diverse organizations and services, individ-
ual refugees collect many dierent identity credentials, from the
UNHCR registration document that identies them as a recognized
refugee to temporary project-related credentials issued by NGOs.
Refugees frequently end up being identied by multiple organiza-
tions, possessing and managing multiple identity credentials, with
dierent identities holding dierent functions and levels of impor-
tance in the refugees’ lives. Common to many of these systems
that record an individual’s data is that they are increasingly digital,
with biometric and biographical data recorded in computerized
systems. We describe refugees’ dierent identity credentials and
the processes for obtaining them before discussing the challenges
refugees face managing these credentials.
4.1 Registration with UNHCR
Our ndings show that the most important identity credential for
most refugees is a UNHCR identity document, which typically
consists of a physical ID card or piece of paper with a barcode. This
credential is obtained through ocial registration with UNHCR
and provides legal recognition of refugee status. It is a material
symbol of the refugee’s legal status and entitlement to benets,
and is critically important for establishing identity and accessing
services from multiple organizations. For example, the World Food
Program and other NGOs that provide services on behalf of UNHCR,
distribute assistance to refugees based on their registration status
with UNHCR. As such, a person’s ability to obtain food, shelter,
health, and other essential services is strongly tied to their UNHCR
registration card. As one participant said,
“The [UNHCR] card is the very important one. At the
distribution centers it’s what they use to give you a
ration card. Secondly, when you go to the hospital, and
thirdly it shows that you are a refugee.” (Focus group,
Uganda, Woman)
As this quote suggests, the UNHCR registration card is also im-
portant because it conrms the refugee’s right to be in the host coun-
try, which is particularly important in countries, such as Lebanon
and Jordan, that limit refugees mobility, access to employment, or
state services. Many participants said that their UN registration
document was very important to carry with them and how it was
often inspected by security forces who stop refugees and demand
they prove their status and identity. One participant said,
“In Syria we never had to carry any documents with us.
Right now, in Lebanon, the situation is very dierent.
Having to carry IDs at all times causes stress because
if the military catches us without IDs, we could be in
trouble.” (P8, Lebanon, Woman).
Interestingly, our analysis revealed that this legal status can
impact men and women dierently, with women often possess-
ing greater mobility than men because security forces stop and
interrogate men more than women. One participant explained,
“When the military come to the house, they don’t talk
to women, only men.” (Focus group, Lebanon, Man)
We learned from our participants that registration with UNHCR
takes place in dedicated centers equipped with private interview
spaces and biometric scanning equipment. Each refugee participates
in an in-person interview that can take up to a few hours. If the
refugee has a family, the whole family is registered as a single
unit, with one adult designated as head of the household. During
registration, participants said that they were asked many personal
questions about life in their home country, reasons for becoming a
refugee, experiences getting to the camp, etc. A participant said,
“[The UNHCR] asks about the experience you had, how
you were in Syria, what you used to do, and what is the
reason that made you leave to here for refuge, and these
things. We gave them a summary, how we arrived, and
how we arrived to them.” (Focus group, Jordan, Man)
Many of these questions aim to enable the UNHCR and other
aid organizations to perform vulnerability assessments, in which
refugees who are deemed to be more vulnerable or high-risk may
receive additional assistance. These vulnerability assessments are
based on complex criteria that are largely opaque to refugees, who
are often nervous about the interview because they fear making
mistakes and failing to get access to much needed services, a chal-
lenge we discuss in detail in Section 5.3. The data collected about the
family is usually stored in proGres [
49
], the identity management
software used by the UNHCR.
In addition to collecting the refugee’s biographical information,
the UN representative also records their biometric data, including
ngerprints, iris scans, and facial images, which are typically stored
in the UNHCR’s Biometric Identity Management System (BIMS)
[
45
], which integrates with proGres. In Jordan, biometric data was
collected by Irisguard, a private biometrics company that provides
their technology to the humanitarian sector.
When we asked participants for their opinions regarding bio-
metric data collection, we discovered that most refugees view the
process positively since it allows organizations to control the dis-
tribution of services and curb fraud. A participant in Jordan said,
Identity at the Margins COMPASS ’19, July 3–5, 2019, Accra, Ghana
“My personal take on the procedure is very positive,
because there are thousands and thousands of refugees
here, so you need a solid identication system to identify
them all accurately. So in terms of documenting the
identities in the computer system, the eye scan is the
most reliable .. . simply because it is the one thing that
you cannot forge. Any documents can be forged, the eye
scan no.” (P34, Jordan, Woman)
Although we found that participants generally viewed the link-
ing of services like food distribution to people’s biometric data as
a positive step, biometric verication was not able to solve all of
the fraud or corruption issues that our participants described. For
example, our data revealed how refugees believe some people who
are responsible for distributing food to communities steal a portion
of the food for themselves. One participant said,
“The ones who ooad [the food trucks]. They are just
taking [it] and we are left with little food . . . now that’s
a big problem .. . they take two kilos from each person’s
12 kilograms [of food].” (P100, Uganda, Woman)
Finally, although the vast majority of our participants viewed
biometric data collection as a positive step, a small number did
express concerns about the biometric systems. These concerns
most often stemmed from rumours and perceptions rather than
issues like data sensitivity. For example, one participant said,
“I heard stories that the eye scanner might steal my
eyes, is that true? Many people are afraid of it.” (P170,
Uganda, Woman)
4.2 Registration with other organizations
In addition to UNHCR, people also register with a wide range of
other organizations in an eort to access desperately needed ser-
vices. These services are often refugees only source of support,
particularly in countries refugees have limited access to labor mar-
kets or state welfare services. One participant said,
“Only people who are here illegally don’t approach
any NGOs for services, they live a very rough life in
makeshift shelters, cold weather, no food or supplies.”
(Focus group, Lebanon, Man)
Some participants described how they registered with any orga-
nization that they came across, sometimes losing track of all the
organizations that they registered with. One participant explained,
“For me, I registered in so many organizations that I
do not even remember any of their names. Whenever
we found or heard of an organization we went and
registered every time.” (P184, Lebanon, Woman)
Although registering with multiple organizations required peo-
ple to provide the same information over and over again, they said
they did not mind the repetition if it led to assistance. One said,
“We don’t get tired [when they ask the same questions
over and over] because all we want is to get help. If help
can come, no problem.” (Focus group, Uganda, Woman)
However, our analysis showed that registering with so many
organizations led to confusion and frustration among the refugees.
In many cases, refugees were unable to associate specic services
received with the organization responsible. They also frequently
confused dierent organizations and did not understand how the
organizations made decisions about who to help. A participant said,
“I registered with a Danish organization and received
nothing, not even bed blankets. I registered and went
through a whole process and when I thought I was -
nally getting something, they informed me that I did
not make the grantee list.” (P59, Lebanon, Woman)
As this quote suggests, refugees often lack knowledge or aware-
ness of the assessment criteria used by organizations to apportion
scarce resources, leading to disappointment and frustration when
organizations who recorded their information failed to deliver any
benets. Nevertheless, many participants still registered with any
organization that might oer assistance. One participant said,
“At the beginning other NGOs showed up and registered
refugees with the promise of assistance. Most of them
didn’t deliver on promises. Refugees lost trust in them
and they have suspicions of corruption in the humani-
tarian system, where they think NGOs take [refugee]
names for their numbers and funding but don’t deliver
to refugees. However, they still register with anyone who
oers help, like a drowning person hanging to a straw
in the ocean.” (Focus group, Lebanon, Man)
5 CHALLENGES MANAGING IDENTITIES
Having described the dierent types of identities that refugees cre-
ate as they try to obtain assistance from humanitarian organizations,
we now discuss additional challenges that our analysis surfaced as
refugees work to establish and manage these credentials.
5.1 Registration is time-consuming & laborious
A big issue for participants was that the process of registering with
organizations was time-consuming and laborious. Many partici-
pants described that it could take months or even years to register
with the UNHCR, with delays arising due to a variety of reasons. For
example, in Lebanon and Jordan, obtaining refugee status was com-
plicated by the host government’s restrictions around the formal
registration of new refugees, and the waiting time before refugees
could register ranged from months to years. This can result, as it
does in Lebanon, in what UNHCR terms ‘de-facto’ refugees—people
who are not legally recognized as refugees but who UNHCR serves
as though they were, providing the same services and support.
To overcome the long waiting times before registering, some
participants described how they used connections at the UN to
manipulate the waiting times for appointments. One told us,
“I would go to the UN, I made an appointment. Of course
they gave me one in nine months, and honestly, I pulled
some strings and I made the appointment earlier. I know
someone who works [at the UN], and he made it sooner
. . . It was in September, he rescheduled it to April. Five
months early.” (P182, Lebanon, Man)
In Uganda, participants described the diculties of waiting out-
side in long lines for days trying to register. A participant said,
“We would stay from morning to sunset waiting to be
registered, even the children suered from the sun. There
was no water to drink. People were suering ... some
COMPASS ’19, July 3–5, 2019, Accra, Ghana E. Schoemaker et al.
people were insulting us. They said that we Sudanese
were dirty.” (P45, Uganda, Woman)
As this quote suggests, many refugees reported that they were
often shouted at or humiliated by people who lived in the host
country. However, in these situations, they felt that they had no
choice but to endure the humiliation since they badly needed the
assistance that was being provided. One participant said,
“You just obey because you are from a foreign land.
You humble yourself because you need the food.” (P69,
Uganda, Male)
Even after refugees reached the front of the line and began the
registration process, obtaining an identity card often took multiple
days, in part because of the overwhelming number of people who
needed to be registered. A participant described,
“The process was very laborious and long. You could
not just walk into the interview room. The numbers of
people were overwhelming. It took a minimum of three
days. They interview you today, take your photo the
next day .. . And on the third day, you go and pick up
your card.” (P58, Uganda, Woman)
5.2 Renewing or changing data is dicult
Our participants also faced challenges changing or renewing their
information. In Lebanon and Jordan, the UNHCR requires all refugees
to renew their registration periodically, but many participants said
that it was dicult or impossible to do so. One participant said,
“For us, it is the renewal issue that is a nightmare ... and
it is stressing us all. We need to renew the UN papers
[every] one or two years, and in many cases we are not
able to do that.” P164, Lebanon, Woman
Once again, participants described how they were only able to
obtain an appointment with the UNHCR to renew their registration
months or years in the future. Being unable to renew registration
documents caused signicant problems for our participants, espe-
cially men, who may be stopped by security forces and found to
have expired papers. One participant described being taken away
and interrogated by police for several days. He told us,
“[The policeman] looked at me and said your card has
expired. I said I knew it was, but I was working on
renewing it. They told me to get in the car. I got in with
them and they took me for three days to the intelligence
agency.” (Focus group participant, Lebanon, Man)
Our participants also discussed how UN and other organization
workers were in a position of power over refugees, which made the
refugees afraid to follow-up too frequently on the status of their
case since, if they annoyed the UN workers, they could use their
power to discriminate against or make the refugees’ lives more
dicult. One participant told us,
“Maybe if I bother [the UN ocer] with many visits
and annoy him, he can just strike out my name.” (P124,
Jordan, Man)
In addition to diculties renewing UNHCR identity documents,
our participants described how changes in personal status, such as
getting married, created challenges because updating their regis-
tration is dicult and takes a long time. One said,
“Transferring something from one le to the next is a
hassle. Like marriage. I am in my parents le. I want to
marry a girl who is in her parents le. For us to have
our own le, it is very hard. Tomorrow . ..try calling the
UN. Ask them to give you an appointment and say you
want to review or edit my le ... they will say it would
take a year.” (P82, Lebanon, Man)
Similarly, trying to record the birth of a new child requires
refugees to navigate many dierent bureaucratic systems to register
the birth before being able to try and add the child’s information
to their UNHCR le. As one participant described,
“Newborns have to be added to the [family’s] UNHCR
le. A birth certicate is issued from hospital. This paper
is certied from mukhtar [town chief] who gives them
the new ocial birth certicate. Then they go to the mu-
nicipality to get it certied and registered in the system.
Then they take that to UNHCR to add children to their
le. It’s very dicult to communicate with UNHCR.”
(Focus group participant, Lebanon, Man)
5.3 Identity systems lack transparency
Another major challenge that our analysis surfaced is how, from
the refugees’ perspectives, existing identity systems suer from a
severe lack of transparency, including when their data is collected,
updated, used, and shared. Many participants explained that, when
asking detailed and intrusive personal questions during registration
processes, representatives from various organizations (including
the UNHCR) often did not explain why they needed all this informa-
tion or how it would be used, despite humanitarian organizations
commitment to ‘informed consent’ [
25
]. One reason for this lack
of explanation was due to workers at registration centers being
overwhelmed by the number of refugees. As one participant said,
“They did not explain [why they need this information].
They were also asking under pressure because the pop-
ulation was too big. That did not give us time to ask.”
(P46, Uganda, Man)
Despite not understanding why their personal data was being
collected, how it would be used, or how it would result in services
that help them, our participants explained that they really did not
have any choice but to provide the information that organizations
requested from them. One participant described,
“You as a refugee, any information .. .that is going to
take place, you cannot say that you don’t want to. You
just humble yourself and do it. If you don’t want to, that
means that you are bringing suering upon yourself.”
(P31, Uganda, Man)
In addition to being required to provide all information requested
from them without understanding how it would be used, our par-
ticipants discussed how the ow of information occurs in only one
direction: they provide information to organizations but are unable
to get information back from organizations. Many participants said
their attempts to contact organizations after registering with them
were unsuccessful. For example, a participant in Lebanon who was
having diculty registering her daughter in school due to lack
of documentation said she was told to call the UNHCR “hotline”,
Identity at the Margins COMPASS ’19, July 3–5, 2019, Accra, Ghana
but this line was never answered. Other participants faced similar
challenges trying to renew their UNHCR registration. One said,
“I went to renew [my registration] and they gave me a
number to call. I called the number, and they said your
card is not eligible.” (P47, Lebanon, Man)
In the absence of ocial explanations for why organizations need
certain pieces of information, refugees often came up with their own
explanations. For example, although a refugee’s biometric data is
primarily collected to verify their identity when receiving services,
one participant perceived this data would be used to determine her
ethnicity and therefore her eligibility for refugee status. She said,
“They did [the biometric scans] because they wanted to
conrm if I was a Ugandan or Sudanese. I was scared
about the whole process.” (P121, Uganda, Woman)
Participants were also acutely aware that the information they
provided was somehow linked to vulnerability assessments that
organizations perform to determine who is eligible for assistance.
The lack of transparency into how these vulnerability assessments
are calculated meant that our participants usually did not under-
stand an organization’s criteria for distributing aid and, in many
cases, believed that the process was unfair. One participant said,
“We see [other] families with same aged children. Some
of them even have cars and still they receive food aid.
My brother has four young children, he receives nothing
from the UN.” (P185, Lebanon, Woman)
Several participants also perceived that only those who had
“wasta” or connections seemed to receive certain benets or favors.
For example, one participant described how it was impossible to
nd employment unless you knew the right people, telling us,
“In short, it is only the people that have wasta . . . they
get work and that is that.” (P120, Jordan, Man)
Without ocial explanations of how vulnerability assessments
are calculated, refugees formed their own perceptions of factors
inuencing their eligibility for assistance and then took steps to
try and maximize their chances of receiving services. For example,
many participants perceived that if a man registered as part of the
household, the family would be denied aid. One participant said,
“During this time, the UN has been denying a lot of
people the assistance they are providing. There are some
people who see it that if there is a man, that they won’t
get assistance.” (P182, Lebanon, Man)
As a result of this perception, we heard numerous stories of how
men would choose to not register with UNHCR. Instead, the woman
and children would register, with the woman listed as the head of
the household. As one NGO employee said,
“We met a lot of men whose wife and children are regis-
tered, and [he] did not register . . . the men with family,
they don’t want to register, but they register their fam-
ily.” (NGO employee, Lebanon, Woman)
Registering the woman as the head of the household gave her
control over the assistance that the family received, such as food,
shelter, and, in Uganda, ownership of the plot of land granted to
each refugee household. However, this new dynamic of female-
headed households also led to a range of unforeseen consequences,
particularly since all our participants are from relatively patriarchal
societies in which men are typically the decision-makers and con-
trol the family life. For example, one participant in Uganda, who
also worked with an NGO that provided support to families and
survivors of domestic violence, described how the women being
the head of household issue is a big cause of domestic violence.
Men want their position and control back, and may abuse women
to get it. Further, we observed additional gender dynamics that
arose when organizations, including UNHCR, assigned community
leadership roles to legally-registered women. One man said,
“The UN . . . assigned a woman as the head of this neigh-
borhood. So how are we expected to talk and deal with
her . . . it bothers me. It’s hard for me accept issues re-
lated to my life decided by not only a woman, but also
a strange woman to me.” (P124, Jordan, Man)
5.4 Data privacy and information sharing
We turn now to our participants feelings towards the privacy of
their personal data and how it might be shared across organizations.
Personal data privacy is particularly important for refugees, who
have often ed persecution and fear being targeted if identied. In
addition, the increasingly digital nature of identication systems
means it is easier for organizations to share data than ever before.
As discussed in Section 4.2, many participants gave their per-
sonal information to any organization that might provide assistance.
However, our analysis shows that the data sharing practices of or-
ganizations is opaque and most participants had no idea if or how
organizations might share their personal information with other
organizations. As one participant in Jordan put it, “we never asked
and we really do not know.”.
When asked how much they cared if organizations shared their
data with others, we received a range of answers. Some participants
viewed data sharing among organizations as a positive and said
that they did not mind if UNHCR shared their information since it
might lead to additional opportunities for employment or assistance.
Other participants said that they did not care about organizations
sharing men’s data, but did not think it was appropriate to share
women’s personal information. As one participant said,
“I would kill them if they share the women’s photos and
information. This demeans the women more than the
men.” (P124, Jordan, Man)
Some participants were so concerned about the potential con-
sequences of data sharing that they avoided registering altogether.
For example, a Syrian refugee living with his family in a one-room
apartment in Lebanon told us,
“Everybody was registering with the UN, but we did
not. We were suspicious and scared. We don’t know if
the UN shares information with anyone, so that is why
I did not share many things with them.” (Focus group
participant, Lebanon, Man)
Many participants felt that some of the information they pro-
vided was safe for organizations to share (e.g., their names or family
size) but were concerned that sharing other kinds of information
that they had been required to provide might lead to future political
persecution or immigration problems. One participant told us,
COMPASS ’19, July 3–5, 2019, Accra, Ghana E. Schoemaker et al.
“The most important question [UNHCR] asked is: what
area in Syria are you from, what is your political ideol-
ogy, have you been a part of protests? .. .the problem is
when they share the questions with the embassies for
resettlement, they go back to the initial le. For example,
if someone has participated in organizing protests or
revolts, this aects his [immigration] case with some
countries.” (P82, Lebanon, Man)
Our participants’ fear of political persecution extended beyond
their concerns regarding information sharing among organizations
and impacted their own sharing practices on social media. But, in
contrast to the lack of knowledge or ability to manage and control
the personal data held by humanitarian organizations, participants
described how they actively manage personal privacy on social
media. Some discussed how they avoided revealing political infor-
mation, for example, by refraining from engaging in conversations
online that touched on the topic of politics. A participant said,
“Mostly when I am chatting [on Facebook], I ask [my
friends] what is happening in their life, when can we
meet again, future plans . . . For me the topic I don’t talk
about is politics. Because it is what caused us problems
that made us refugees. I have fear of it. Sometimes I feel
like if I engage in such talk, someone may just come
and attack me.” (P71, Uganda, Man)
Other participants described how they, and other refugees they
knew, went even further by using fake names on any email accounts
or social media proles in an eort to hide their identities from the
authorities. One participant told us,
“My email is a pseudonym. My prole is a pseudonym.
There is no real name ...Because they will apprehend
me, catch or arrest me immediately for anything. Most
people [use fake names].” (P147, Lebanon, Man)
Yet another participant believed that it did not matter if people
hid their names or not, the government would be able to know
about anything that the participant posted online, telling us,
“With Facebook there is no privacy. When you have
your photos, whether you are hiding or not for security
reasons, already they know.” (P71, Uganda, Man)
In addition to concerns about online surveillance and tracking
of social media activity by authorities, participants worried about
physical inspection of their devices by security forces and described
how they tried to manage these risks. One participant in Lebanon
explained how he had been stopped, interrogated, and beaten at
security checkpoints. During these interrogations, he was glad he
had left the smartphone he used behind, a conscious decision to
prevent the police going through his phone, either identifying his
contacts or using the content to further detain or harass him.
6 DISCUSSION
Our analysis shows how the current state of humanitarian iden-
tity systems presents many challenges for refugees. We now use
Bardzell’s feminist HCI framework [
8
] of six design qualities (plu-
ralism, ecology, embodiment, self-disclosure, advocacy, and partic-
ipation) to make sense of these challenges and suggest a starting
point for the design of improved identity management systems.
6.1 Pluralism
The quality of pluralism refers to “design artifacts that resist any
single, totalizing, or universal point of view” [
8
]. Our analysis sug-
gests that the digital systems used by humanitarian organizations
to manage beneciary data have been designed for a typical (i.e.,
“universal”) refugee rather than to support diverse or partial per-
spectives of dierent refugees in dierent contexts [
24
]. Indeed,
although our research took place with diverse communities in mul-
tiple countries, many of the digital systems in use were the same
UNHCR systems despite contextual and cultural dierences.
UNHCR’s ProGres is designed to be a universal registration
system, intended for use in the diverse contexts in which the or-
ganisation operates. There is evidence that the same universal,
totalising design leads to common experiences and outcomes for
refugees beyond the three countries our research focused on [
28
].
Latonero et al.’s 2019 study of refugee experiences in Italy, includ-
ing interactions with UNHCR’s ProGres, found "The systems that
use identity data to classify individuals along political and economic
lines can have lasting eects on their rights and freedoms." They also
found common experiences in relation to informed consent and
agency, noting that "Migrants exchange identity data for resources
without meaningful consent" [
28
]. More broadly, there are eorts
to standardize the digital management of refugee support, with
organizations such as the UN’s World Food Programme seeking to
integrate all its support through their beneciary management plat-
form SCOPE [
14
], and aid providers such as the UK’s Department
for International Development funding research into universal stan-
dards for refugee registration [
17
]. The commonality of Latonero et
al.’s ndings and the trend towards system standardization points
towards possible generalizability of our broader ndings, although
further research is needed to conrm this conclusion.
However, although this universal approach to rendering refugees
legible to organizations has many benets for the organizations
(such as simplied reporting, training, and system maintenance),
we see that it leads to challenges for individual refugees who are
constrained within the institutional bounds of these systems. For
example, many refugees have no choice as to what identities or reg-
istration cards they are eligible for, or the categories and labels that
organizations assign to them. Further, whilst these categories, such
as refugee status or specic vulnerability criteria that determine
access to specic services, are based on assessments of refugee
needs, individual refugees are unable to negotiate exibility in the
nature of the identity or in their identication and eligibility.
In contrast to the systems used to register and manage service
delivery, we show how the social media platforms that refugees
use to share information do a better job of supporting pluralism
and dierent points of view. Although social media platforms such
as Facebook and Twitter have well-described limitations and con-
straints around privacy and exibility over data sharing and identity
construction (e.g., [
18
,
19
]), they oer far greater means of individ-
ual end-user agency and control over personal data. Many refugees
exercised individual control in the use of these systems, for example
using fake names, being careful about revealing information online
(not least political aliation or views), or using dierent phones in
specic contexts (i.e., not taking their phone from Lebanon to Syria).
This shows that individual refugees have the capacity and desire to
Identity at the Margins COMPASS ’19, July 3–5, 2019, Accra, Ghana
exercise control and manage digital platforms, which would apply
to institutional systems as much as it does to social media platforms.
In other words, when individuals are able to express dierent points
of view and exercise control over personal data, they do.
6.2 Ecology
The quality of ecology integrates an awareness of design artifacts’
eects in their broadest contexts of use, as well as an awareness
of how design artifacts aect all stakeholders [
8
]. It invites us to
attend to the ways that design artifacts reexively design us. This
quality manifests in our data in several ways.
For example, UNHCR’s registration system includes the designa-
tion of a “head of household”, the person that is issued food rations
and, in the case of Uganda, granted ownership of the government-
issued plot of land. This designation is part of UNHCR’s registration
system, which works the same in Lebanon, Jordan, and Uganda as it
would anywhere else. The head of household is typically the eldest
family member. The universalism of this sociotechnical system has
diverse eects. It structures the family relationship so that it is
legible to a system that then confers rights and privileges—either
conrming the man as the head of the household or, in some cases,
restructuring the family unit. This restructuring of family life by reg-
istration systems can take dierent forms. For example in Uganda,
because women often ed before male relatives, they were regis-
tered as heads of households, making them the ocial recipients
of UNHCR rations and Ugandan government land. Many women
described how this granted them a greater degree of agency and
control over their family life. It also meant they took the place for-
merly occupied by their husbands or male family members, which
sometimes caused problems when the men subsequently arrived
and sought to reclaim their position of power and authority.
Another example of ecology can be found in the power asymme-
tries between refugees and those responsible for providing services
to them. Specically, we show how refugees lack of power and
desperation for services make them vulnerable to exploitation. For
example, in Uganda, refugees were forced to hand over a portion
of their food ration to the people who delivered the food to their
homes. These examples illustrate the diverse relationships between
sociotechnical systems and their broader contexts of use, as well
as how social life and relationships interact with technology, with
both positive and negative outcomes.
6.3 Embodiment
The quality of embodiment urges “focalizing the agency of inter-
action not on the interface or its designer, but the bodies, motivat-
ing drives, and primordial urges of users” [
8
]. In our research, we
see many ways in which the identity management systems that
determine refugees’ legal status and eligibility for services have far-
reaching consequences for refugees’ “bodies, motivating drives, and
primordial urges.” For example, refugees’ legal recognition through
registration with UNHCR was often instrumental in gaining access
to basic, tangible services that were essential for survival, including
shelter, food, and health services. In addition, challenges renewing
refugee status sometimes led to mobility problems for refugees,
especially men. We heard stories from participants of how they
would be stopped and interrogated by security forces, and detained
or beaten up if they did not possess valid identication.
Given the high stakes for refugees when it comes to how the data
they provide to organizations does or does not result in tangible
benets necessary for survival, it is unsurprising that they strive to
actively negotiate the identities available to them (despite limited
opportunities to do so), consciously weighing the benets and con-
straints associated with dierent statuses in order to access services,
ensure eligibility for employment, and preserve spatial mobility.
For example, many refugees have partial, or perceived, knowledge
of organizations’ registration processes, which leads them to adopt
dierent strategies in negotiating these processes—including avoid-
ing registration completely in some cases or selectively registering
members of households because they perceive that female-headed
households receive more benets than male-headed households. In
addition, many refugees we spoke with voiced concerns about reg-
istration interview questions, such as place of origin, and anxiety
about the consequences of this data being shared (e.g., on future
immigration opportunities or safety from political persecution).
6.4 Self-disclosure
The quality of self-disclosure refers to the extent to which software
renders visible the ways it eects users as subjects [
8
]. It calls users’
awareness to what the software is trying to make of them and cre-
ates opportunities for users to (re)dene themselves to become the
kind of user the software is for, or set aside the parts of themselves
that are less relevant to the software. In other words, the software
gives users an identity they are pressured into accepting.
The quality of self-disclosure on the part of humanitarian reg-
istration systems is particularly important in our context. Indeed,
the very process of registration and the establishment of eligibility
to specic services rests on institutionally dened categories into
which individual refugees must conform. At the same time, refugees
and other displaced persons are by denition extremely vulnerable.
They are desperate for assistance and, as our data shows, many
will strive to dene themselves in ways that make them legible to
these systems if there is even a chance that it will lead to additional
services. In these contexts, the organizations collecting the data
hold all of the power, which also highlights challenges obtaining
true consent to collect and share people’s personal data [25].
The lack of self-disclosure of these systems is also apparent in
individuals’ lack of understanding of how organizations make de-
cisions about their eligibility for services, in ways that echo other
types of systems where black-box processes, such as algorithms,
determine important outcomes. The experience of people who have
little to no understanding of how organizations assess their vul-
nerability or needs is similar to the increasing number of ways
in which unknown algorithms determine the choices and options
available to individuals all over the world, from credit scores and
search results to online prices and news information. Transparency
of these vital decision-making processes could enable broader un-
derstanding of how organizations distribute scarce resources, yet
such self-disclosure might also invite eorts to game the system.
At the heart of this debate is of course the nature of the relation-
ship between institutions and individuals, and who gets to decide
whether trust and transparency can co-exist.
COMPASS ’19, July 3–5, 2019, Accra, Ghana E. Schoemaker et al.
6.5 Advocacy
The quality of advocacy focuses on ensuring technology is pro-
gressive and attempts to bring about political emancipation [
8
].
Although the UNHCR and other organizations in our research os-
tensibly exist to advocate for refugees and improve their quality of
life, these organizations are also focused on and working within the
status quo, and must conform to the political, social, and cultural
environments of the countries in which they operate. For example,
we saw how, when host governments say that no more refugees
are allowed to be registered, the UNHCR must follow their deci-
sion, even if it negatively impacts refugees. As such, organizations
need to pay attention to the ways in which they may perpetuate
regressive and harmful practices and structures [8].
Another way that the quality of advocacy manifests in our work
is in how refugees who were desperate for services chose to share
sensitive personal information with organizations despite skepti-
cism that those organizations would deliver the services promised.
In these examples, participants described how they worried that the
very organizations supposed to advocate for them instead used their
data in ways that only beneted the organization (i.e., to obtain
funding) but did not provide benets to refugees.
We also see many examples of how, when organizations’ prac-
tices and policies clash with refugees needs and opinions, the orga-
nizations have all the power, with refugees often forced to conform
to organizations’ requirements. For example, in most cases, refugees
have no choice but to share their data if they want access to services.
They are also required to provide all the information requested, with
no choices for holding some information back or asking that certain
pieces of information not be shared with other organizations.
Many systems also currently prevent refugees from being able
to advocate for themselves. For example, the descriptions we heard
of ways in which individuals were unable to inuence or change
the systems that recorded their information illustrate how those
systems, and the processes they are part of, do not allow space for
individuals to control the data systems hold about them. Further-
more, as digital systems enable easy sharing of data, refugees have
limited ability to control where their data goes. We heard many
people describe their surprise and concern about how organizations
had shared their personal data with others.
We also acknowledge the bias towards refugees’ perspectives
in our study. The goal of our work was to examine refugees’ ex-
periences with, and perspectives on, the digital identity systems
that humanitarian organizations use to collect and manage their
personal data. As such, we did not talk with organizations to un-
derstand their side of the story, and acknowledge that many of the
challenges our data surfaced engage with complex, nuanced, and
non-trivial issues that organizations work to try and overcome.
6.6 Participation
Finally, the quality of participation refers to valuing participatory
processes that lead to the creation of interactive systems [
8
]. In our
research, the design of digital humanitarian systems generally fol-
lowed a top-down approach, focusing on organizations as the users
of such systems, catering to the organizations’ needs, priorities, and
perspectives, and making refugees legible in ways determined by
those organizations. But our study shows that refugees themselves
are also an important category of “users” of these systems, with
dierent needs and priorities than organizations. Unfortunately,
the design of current digital identity systems has not prioritized
the participation of refugees, leading to challenges for refugees as
they seek to establish and manage identities via these systems.
We hypothesize that participatory methods, in which refugees
are treated as active contributors to the design of humanitarian
systems, might result in many benets for both organizations and
refugees. For example, the increased condence in systems that
better account for refugees’ concerns and priorities might lead to
higher rates of registration and stronger protections for vulnerable
beneciaries. In addition, there might be increased eciency due to
more accurate data as refugees work to maintain up-to-date records;
or there might be resource savings that result from a transfer of the
eort required to perform data management from organizational
sta to individuals. Of course, beyond the benets to organizations
is the greater control and dignity for refugees themselves that would
result from systems designed around their needs and interests.
Encouragingly, and based in part on our work, the UNHCR has
recognized the need to be able to consult better with refugees and
other displaced populations on issues of digital identity. Over the
past few months, our team has engaged with the UNHCR in a
project to design a participatory toolkit that will provide refugees
with opportunities to participate in the design or customization
of UNHCR systems deployed in their specic country or context.
The toolkit will facilitate consultations with refugees on key is-
sues, including their views on registration and identity manage-
ment, recommendations for improvements, identifying dierent
experiences, including dierential access or discrimination (e.g.
for refugee women and girls), privacy and information sharing
concerns, perceptions of biometric technologies, and more.
Our early prototypes draw inspiration from prior work on par-
ticipatory toolkits, such as the city of Helsinki’s Participation Game
[
15
] that enables city employees, residents, and other stakehold-
ers to provide input to improve the city’s operations and services.
The goal is for the new UNHCR participatory toolkit to eventually
become a part of PRIMES (Population Registration and Identity
Management Ecosystem) [
48
], the UNCHR’s ecosystem of inter-
operable tools, which will be the digital platform for beneciaries,
UNHCR, and other partners to interact in the digital sphere.
7 CONCLUSION
This paper examines refugees’ experiences with and perspectives
on the digital identity systems used by humanitarian organizations
to collect, manage, and share their personal data. We show how
refugees face many challenges negotiating the digital systems that
render them legible to service providers, discuss strategies refugees
employ to exercise agency and control within these systems, and
demonstrate the impact of these systems on their social relation-
ships. We use Bardzell’s lens of feminist HCI [
8
] to make sense of
these challenges and suggest a starting point for engaging refugees
in the design of improved identity management systems.
ACKNOWLEDGMENTS
We sincerely thank all our participants. This work was funded by
DFID under contract number PO40107257.
Identity at the Margins COMPASS ’19, July 3–5, 2019, Accra, Ghana
REFERENCES
[1]
Konstantin Aal, Marios Mouratidis, Anne Weibert, and Volker Wulf. 2016. Chal-
lenges of CI initiatives in a political unstable situation-case study of a computer
club in a refugee camp. In Proceedings of the 19th International Conference on
Supporting Group Work. ACM, 409–412.
[2]
Konstantin Aal, Anne Weibert, Reem Talhouk, Vasilis Vlachokyriakos, Karen
Fisher, and Volker Wulf. 2018. Refugees & Technology: Determining the Role of
HCI Research. In Proceedings of the 2018 ACM Conference on Supporting Groupwork.
ACM, 362–364.
[3]
Arthur Allison, James Currall, Michael Moss, and Susan Stuart. 2005. Digital
identity matters. Journal of the American Society for Information Science and
Technology 56, 4 (2005), 364–372.
[4]
Asam Almohamed and Dhaval Vyas. 2016. Designing for the Marginalized: A step
towards understanding the lives of refugees and asylum seekers. In Proceedings of
the 2016 ACM Conference Companion Publication on Designing Interactive Systems.
ACM, 165–168.
[5]
Asam Almohamed and Dhaval Vyas. 2016. Vulnerability of displacement: chal-
lenges for integrating refugees and asylum seekers in host communities. In
Proceedings of the 28th Australian Conference on Computer-Human Interaction.
ACM, 125–134.
[6]
Asam Almohamed, Dhaval Vyas, and Jinglan Zhang. 2017. Rebuilding social
capital: engaging newly arrived refugees in participatory design. In Proceedings
of the 29th Australian Conference on Computer-Human Interaction. ACM, 59–67.
[7]
Jennifer Barano, R Israel Gonzales, Jay Liu, Heidi Yang, and Jimin Zheng. 2015.
Lantern: Empowering refugees through community-generated guidance using
near eld communication. In Proceedings of the 33rd Annual ACM Conference
Extended Abstracts on Human Factors in Computing Systems. ACM, 7–12.
[8]
Shaowen Bardzell. 2010. Feminist HCI: taking stock and outlining an agenda for
design. In Proceedings of the SIGCHI conference on human factors in computing
systems. ACM, 1301–1310.
[9]
Roy F Baumeister and Mark Muraven. 1996. Identity as adaptation to social,
cultural, and historical context. Journal of adolescence 19, 5 (1996), 405–416.
[10]
Virginia Braun and Victoria Clarke. 2006. Using thematic analysis in psychology.
Qualitative research in psychology 3, 2 (2006), 77–101.
[11]
Deana Brown and Rebecca E Grinter. 2016. Designing for transient use: A
human-in-the-loop translation platform for refugees. In Proceedings of the 2016
CHI Conference on Human Factors in Computing Systems. ACM, 321–330.
[12]
Rogers Brubaker and Frederick Cooper. 2000. Beyond âĂIJidentityâĂİ. Theory
and society 29, 1 (2000), 1–47.
[13]
Ana Maria Bustamante Duarte, Auriol Degbelo, and Christian Kray. 2018. Explor-
ing Forced Migrants (Re) settlement & the Role of Digital Services. In Proceedings
of 16th European Conference on Computer-Supported Cooperative Work-Exploratory
Papers. European Society for Socially Embedded Technologies (EUSSET).
[14]
Catherine Cheney. 2019. How digital identity can address
both protection and inclusion. https://www.devex.com/news/
how-digital- identity-can-address-both-protection-and-inclusion- 94449.
(2019).
[15]
City of Helsinki. 2018. Participation Game. https://www.hel./helsinki/en/
administration/participate/channels/participation-model/participation- game/.
(2018).
[16]
Lizzie Coles-Kemp, Rikke Bjerg Jensen, and Reem Talhouk. 2018. In a New Land:
Mobile Phones, Amplied Pressures and Reduced Capabilities. In Proceedings of
the 2018 CHI Conference on Human Factors in Computing Systems. ACM, 584.
[17]
Paul Currion, Bryan Pon, and Emrys Schoemaker. 2018. Identity at the Margins:
refugee identity and data management. Caribou Digital Publishing (2018).
[18]
Emiliano De Cristofaro, Claudio Soriente, Gene Tsudik, and Andrew Williams.
2012. Hummingbird: Privacy at the time of twitter. In Security and Privacy (SP),
2012 IEEE Symposium on. Citeseer, 285–299.
[19]
Bernhard Debatin, Jennette P Lovejoy,Ann-Kathrin Horn, and Brittany N Hughes.
2009. Facebook and online privacy: Attitudes, behaviors, and unintended conse-
quences. Journal of computer-mediated communication 15, 1 (2009), 83–108.
[20]
James D Fearon. 1999. What is identity (as we now use the word). Unpublished
manuscript, Stanford University, Stanford, Calif (1999).
[21]
Karen E Fisher, Reem Talhouk, Katya Yemova, Dalya Al-Shahrabi, Eiad Ya, Sam
Ewald, and Rob Comber. 2017. Za’atari Refugee Cookbook: Relevance, Challenges
and Design Considerations. In Proceedings of the 2017 CHI Conference Extended
Abstracts on Human Factors in Computing Systems. ACM, 2576–2583.
[22]
Karen E Fisher, Katya Yemova, and Eiad Ya. 2016. Future’s Butteries: Co-
Designing ICT Wayfaring Technology with Refugee Syrian Youth. In Proceedings
of the The 15th International Conference on Interaction Design and Children. ACM,
25–36.
[23]
Leo A Goodman. 1961. Snowball sampling. The annals of mathematical statistics
(1961), 148–170.
[24]
Donna Haraway. 1988. Situated knowledges: The science question in feminism
and the privilege of partial perspective. Feminist studies 14, 3 (1988), 575–599.
[25]
Richard Hugman, Eileen Pittaway, and Linda Bartolomei. 2011. When âĂŸdo no
harmâĂŹis not enough: The ethics of research with refugees and other vulnerable
groups. The British Journal of Social Work 41, 7 (2011), 1271–1287.
[26]
Azalea Irani, Kriti Nelavelli, Kristin Hare, Paula Bondal, and Neha Kumar. 2018.
Refuge Tech: An Assets-Based Approach to Refugee Resettlement. In Extended
Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems.
ACM, LBW554.
[27]
Konstantinos Kazakos, Siddhartha Asthana, Madeline Balaam, Mona Duggal,
Amey Holden, Limalemla Jamir, Nanda Kishore Kannuri, Saurabh Kumar,
Amarendar Reddy Manindla, Subhashini Arcot Manikam, et al
.
2016. A real-time
ivr platform for community radio. In Proceedings of the 2016 CHI Conference on
Human Factors in Computing Systems. ACM, 343–354.
[28]
Mark Latonero, Keith Hiatt, Antonella Napolitano, Giulia Clericetti, and Melanie
Penagos. 2019. Digital Identity in the Migration and Refugee Context: Italy Case
Study. Data and Society (2019), 1–45.
[29]
Mark R Leary and Robin M Kowalski. 1990. Impression management: A literature
review and two-component model. Psychological bulletin 107, 1 (1990), 34.
[30]
Alvaro Martin, Ana Isabel Segovia, et al
.
2018. Digital Identity: the current state
of aairs. Technical Report.
[31]
Wim Meeus, Jurjen Iedema, Marianne Helsen, and Wilma Vollebergh. 1999.
Patterns of adolescent identity development: Review of literature and longitudinal
analysis. Developmental review 19, 4 (1999), 419–461.
[32]
Sara Nabil, Reem Talhouk, Julie Trueman, David S Kirk, Simon Bowen, and
Peter Wright. 2018. Decorating Public and Private Spaces: Identity and Pride
in a Refugee Camp. In Extended Abstracts of the 2018 CHI Conference on Human
Factors in Computing Systems. ACM, LBW552.
[33]
Hamid Nach and Albert Lejeune. 2009. The impact of information technology
on identity: Framing the research agenda. (2009).
[34]
Jean S Phinney. 1990. Ethnic identity in adolescents and adults: review of research.
Psychological bulletin 108, 3 (1990), 499.
[35]
Samar Sabie, Jay Chen, Azza Abouzied, Fatma Hashim, Harleen Kahlon, and
Steve Easterbrook. 2017. Shelter Dynamics in Refugee and IDP Camps: Cus-
tomization, Permanency, and Opportunities. In Proceedings of the 2017 Workshop
on Computing Within Limits. ACM, 11–20.
[36]
Paul Schmitt, Daniel Iland, Elizabeth Belding, Brian Tomaszewski, Ying Xu, and
Carleen Maitland. 2016. Community-level access divides: A refugee camp case
study. In Proceedings of the Eighth International Conference on Information and
Communication Technologies and Development. ACM, 25.
[37]
Seteney Shami. 1996. Transnationalism and refugee studies: Rethinking forced
migration and identity in the Middle East. Journal of Refugee Studies 9, 1 (1996),
3–26.
[38]
Anouk Smeekes, Maykel Verkuyten, Elif Çelebi, Ceren Acartürk, and Samed
Onkun. 2017. Social identity continuity and mental health among Syrian refugees
in Turkey. Social psychiatry and psychiatric epidemiology 52, 10 (2017), 1317–1324.
[39]
O Stickel, D Hornung, K Aal, M Rohde, and V Wulf. 2015. 3D Printing with
Marginalized Children-An Exploration in a Palestinian Refugee Camp, ECSCW
2015: Proceedings of the 14th European Conference on Computer Supported
Cooperative Work, 19-23 September 2015, Oslo, Norway, Nina Boulus-Rødje,
Gunnar Ellingsen, Tone Bratteteig, Margunn Aanestad, Pernille Bjørn. Springer,
URL: doi 10 (2015), 978–3.
[40]
Reem Talhouk, Tom Bartindale, Kyle Montague, Sandra Mesmar, Chaza Akik,
A Ghassani, M Najem, H Ghattas, Patrick Olivier, and Madeline Balaam. 2017.
Implications of synchronous IVR radio on Syrian refugee health and community
dynamics. In Proceedings of the 8th International Conference on Communities and
Technologies. ACM, 193–202.
[41]
Reem Talhouk, Ana Bustamante, Konstantin Aal, Anne Weibert, Koula
Charitonos, and Vasilis Vlachokyriakos. 2018. HCI and refugees: experiences
and reections. Interactions 25, 4 (2018), 46–51.
[42]
Reem Talhouk, Sandra Mesmar, Anja Thieme, Madeline Balaam, Patrick Olivier,
Chaza Akik, and Hala Ghattas. 2016. Syrian refugees and digital health in
Lebanon: Opportunities for improving antenatal health. In Proceedings of the 2016
CHI Conference on Human Factors in Computing Systems. ACM, 331–342.
[43]
Reem Talhouk, Vasillis Vlachokyriakos, Anne Weibert, Konstantin Aal, Syed Ish-
tiaque Ahmed, Karen Fisher, and Volker Wulf. 2017. Refugees & HCI Workshop:
The Role of HCI in Responding to the Refugee Crisis. In Proceedings of the 2017
CHI Conference Extended Abstracts on Human Factors in Computing Systems. ACM,
558–565.
[44]
Uganda Oce of the Prime Minister. 2018. Refugees Management. https://opm.
go.ug/refugees-management/. (2018).
[45]
UNHCR. 2018. Biometric Identity Management System. http://www.unhcr.
org/en-us/protection/basic/550c304c9/biometric- identity-management- system.
html. (2018).
[46]
UNHCR. 2018. Jordan Factsheet. https://reliefweb.int/report/jordan/
unhcr-jordan- factsheet-june- 2018. (2018).
[47] UNHCR. 2018. Lebanon. http://www.unhcr.org/lb/shelter. (2018).
[48]
UNHCR. 2018. PRIMES: Population Registration and Identity Management
EcoSystem. https://www.unhcr.org/primes.html. (2018).
[49]
UNHCR. 2018. Registration via ProGres. http://www.unhcr.org/en- us/
registration.html. (2018).
COMPASS ’19, July 3–5, 2019, Accra, Ghana E. Schoemaker et al.
[50]
UNHCR. 2018. Syrian Regional Refugee Response. https://data2.unhcr.org/en/
situations/syria/location/36. (2018).
[51] UNHCR. 2018. Uganda. https://data2.unhcr.org/en/country/uga. (2018).
[52]
Sara Vannini, Ricardo Gomez, Megan Carney, and Katharyne Mitchell. 2018.
Interdisciplinary Approaches to Refugee and Migration Studies: Lessons from
Collaborative Research on Sanctuary in the Changing Times of Trump. Migration
and Society 1, 1 (2018), 164–174.
[53]
Alan S Waterman. 1982. Identity development from adolescence to adulthood:
An extension of theory and a review of research. Developmental psychology 18, 3
(1982), 341.
[54]
Edgar A Whitley, Uri Gal, and Annemette Kjaergaard. 2014. Who do you think
you are? A review of the complex interplay between information systems, iden-
tication and identity. (2014).
[55]
Ying Xu, Adrian Holzer, Carleen Maitland, and Denis Gillet. 2017. Community
building with co-located social media: A eld experiment with syrian refugees.
In Proceedings of the Ninth International Conference on Information and Commu-
nication Technologies and Development. ACM, 16.
[56]
Ying Xu and Carleen Maitland. 2016. Communication Behaviors When Displaced:
A Case Study of Za’atari Syrian Refugee Camp. In Proceedings of the Eighth
International Conference on Information and Communication Technologies and
Development. ACM, 58.
[57]
Ying Xu and Carleen Maitland. 2017. Mobilizing Assets: Data-Driven Community
Development with Refugees. In Proceedings of the Ninth International Conference
on Information and Communication Technologies and Development. ACM, 10.
[58]
Ying Xu, Carleen Maitland, and Brian Tomaszewski. 2015. Promoting partic-
ipatory community building in refugee camps with mapping technology. In
Proceedings of the Seventh International Conference on Information and Communi-
cation Technologies and Development. ACM, 67.
[59]
Dr Eiad Ya and Karen E Fisher. 2018. Syrian Youth in ZaâĂŹatari Refugee Camp
as ICT Wayfarers: An Exploratory Study Using LEGO and Storytelling. (2018).
[60]
Eiad Ya, Katya Yemova, and Karen E Fisher. 2018. Young Hackers: Hacking
Technology at Za’atari Syrian Refugee Camp. In Extended Abstracts of the 2018
CHI Conference on Human Factors in Computing Systems. ACM, CS21.
[61]
George Yerousis, Konstantin Aal, Thomas von Rekowski, David W Randall,
Markus Rohde, and Volker Wulf. 2015. Computer-enabled project spaces: Con-
necting with Palestinian refugees across camp boundaries. In Proceedings of the
33rd Annual ACM Conference on Human Factors in Computing Systems. ACM,
3749–3758.
