Abstract
Homeless people experience a unique set of challenges leading to pervasive health and social problems. An increasing number of researchers have harnessed photographic data to gain a unique perspective of marginalised groups. The aim of the study is to explore the feasibility of using photographs in research to understand the complex environment experienced by homeless people, with a special interest in mental health. Individuals who frequently attend homeless facilities in London were sensitively approached and asked if they would be interested in taking part in the ‘Views From the Street’ pilot study. Once agreement was confirmed through a formal consenting procedure, participants were asked to visually capture and upload their own digital photos, along with a brief description. The collection of data highlighted a number of barriers to engagement and acceptability, including issues around the level of familiarity with the recruiter, practicalities of participation, public perception of phone use, poor technical literacy, anonymity, and disassociation with the ‘homeless’ label. Recommendations are made for future research utilising photographic participatory designs with the homeless population.
Introduction
Homelessness
Homeless people experience a unique set of challenges leading to pervasive health and social problems. Homelessness is broadly defined by not having a home. While people typically associate this term with those who are sleeping rough, the definition of homelessness also encompasses those who are unable to access their accommodation, are staying at a night shelter, a women’s refuge or with a friend.1–3
According to Shelter.org, more than 300 000 people in the United Kingdom were homeless in 2017, a 4.3% increase from the previous year. 4 The environment experienced by homeless people is often complex and challenging. Not having access to stable accommodation can have a ripple effect on other areas of life, such as attaining work, sustaining social links, and maintaining safety. 5 The challenging conditions faced by this population can also have adverse effects on their health. An audit that was undertaken by Homeless Link in 2014 found that 73% of the homeless people surveyed had physical health problems and 80% reported concerns about mental health.4,6 The same audit revealed that mental health was nearly twice as high in the homeless population (45%) than the general population (25%). Similar trends were also observed across specific mental health diagnoses (eg, depression, posttraumatic stress disorder, bipolar, and schizophrenia). 6
Homelessness and social exclusion
Poor health, education, and personal development experienced by homeless people can result in social exclusion.4,5 A qualitative study exploring the perceptions of 10 shelter residents on belonging and overcoming social inclusion revealed that homelessness represented the absence of support and inclusion in one’s community. For this reason, some residents also expressed reluctance in reintegrating into the community that had ‘cast them off’. 7 Disconnection between the homeless population and the public can further compromise access to other support structures that can help alleviate some of the burdens associated with homelessness.
Homeless people also struggle with the stigma of not having a home. Their living conditions render them invisible, yet vulnerable to acts of prejudice, discrimination, and violence. Such stigma is also reflected in general society and legislation, through restrictions that can promote social and occupational exclusion in homeless people. 8
Concerns around social inclusion of the homeless population have further grown in light of the digital revolution. This is because digital media is becoming an increasingly important part of everyone’s lives. It is used for socialising, finding information, applying for jobs, and staying in contact with family and friends. The increasing dependence on digital media is believed to be leaving homeless people and other vulnerable groups behind. For instance, homeless people who struggle with digital literacy may experience an additional set of barriers in relation to finding work, transport, and housing, when resources are increasingly only available online.9,10
Photography as a research tool
Photographs have proved to be a useful communicative aid and facilitator for social inclusion in vulnerable, disempowered, and marginalised groups. 11 Visual data, particularly photography has the capacity to offer unique insights on a range of topics and communicate complex messages with visual simplicity. Photographs can confer meaning to the world, either in a metaphorical, literal, abstract, actual, or visual sense. 12
While traditional social research methods often present access barriers for perceived vulnerable and marginalised groups, visual research methods, in particular photography, have the potential to reveal information which is difficult to obtain from other approaches. 12
Participant-led data collection can help researchers break down the power hierarchies through allowing participants to prioritise issues that are important to them, as opposed to predetermined topics set by researchers. 12 The process of participation can also lead to self-exploration and construction of positive identities of the self. 12 The shift in power dynamics between the researcher and the participant can create a basis of co-creation, leading to richer data. Moreover, the use of photo elicitation can facilitate engagement between the researcher and the participant, creating an alliance which can be invaluable to both parties. 12
A study carried out by Jan-Paul Brekke 13 which attempted to understand the daily lives of asylum-seekers in Sweden found that photographs created a positive effect on the relationship between him and the refugees. It was reported that asylum-seekers looked forward to seeing how their images turned out and displayed a sense of ownership of what they had produced. Photographs were used to build an alliance with a population group that were not native to the country and may therefore be difficult to build rapport with.
This approach has also been used with the homeless population. Café Art, a social enterprise run an annual photo-contest that aims to empower homeless people to tell their stories through photography to promote greater social connection with the general public.14,15 The use of photographs can also be harnessed to instigate change beyond the participator level. The use of photographs in research can be used as a springboard for debate to enhance the understanding of issues that affect homeless people on a daily basis. It can also be used to encourage political activism to propel issues important to marginalised groups to the forefront.9,12
Given the multiple benefits of using photographs in research with marginalised groups, this pilot project aims to explore the feasibility of using photographic data collection to understand the complex environment experienced by homeless people, with a special interest in mental health.
Methods
Participants and procedure
Homeless people who attended 1 of 2 homeless/community centres in London between March and July 2017 were invited to be involved in the project. Authorisation to recruit from the centres was approved by management staff. The recruitment lead (BI) volunteered in both centres 4 months ahead of recruitment to build trust and rapport with the homeless people using the facilities. She attended each centre every other Tuesday in the evenings from 5:00 to 7:00 p.m., spending a total of 16 h at each centre. Active data collection was carried out once a week between 5:00 and 7:00 p.m. over 3 weeks in July (accounting for a total of 6 h of recruitment time at each centre). Homeless people who attended the centres were sensitively approached and asked if they were interested in taking part in the ‘Views From the Street’ project. It was explained that participants would be required to upload photos that capture their day-to-day life living on the streets. If a potential participant expressed interest in taking part, an information sheet was provided, highlighting their right to withdraw at any point. It was anticipated that the homeless population may not readily have access to photographic equipment. 9 The project therefore provided participants with Samsung Galaxy Phone Ace 4 devices with a 1.3-megapixel camera. Participants were instructed to upload their photographs and accompanying descriptions (including thoughts and feelings) to a designated website. Information about the photo uploading process and the web address were listed and stuck on the back of the phone to ensure easy access to instructions. Wi-Fi access was organised with a local Starbucks coffee shop as communal Wi-Fi was not available in the homeless / community centres. Free refreshments were also arranged for participants, to create a welcoming and ambient environment.
Data collection method
The photo elicitation method was used to collect data from participants. While photo elicitation traditionally involves inserting a photograph (taken by the participant or produced by the research team) into a research interview, the study’s main data source consisted of a photograph that was supplemented with a brief text-based description. It should be noted that photo elicitation does not assume an underlying objectivity but acknowledges the power of social constructs and the individuals’ unique elicitations and personal narratives. 12 The photos submitted by the participants were used to elicit information on topics that are of interest to them. The project team encouraged participants to submit any photo they desired and did not specify a criteria or theme. This was to enable participants to set the agenda through the topics they choose to address in their photography. However, the project was especially interested in evaluating how mental health was presented in the data. The proposed data analysis for the study was a thematic analysis which incorporates both photographs and accompanying descriptions.12,16,17
Results
Homeless people were actively recruited over 3 weeks. The pilot study aimed to collect data for 12 participants. Only 3 people agreed to take part in the project out of the 30 people who had interacted with the study recruiter, and (of those) 10 people who were formally invited to participate. The piloting phase of the study was halted due to a range of challenges experienced in relation to the feasibility and acceptability of the project. As a result, this prompted a shift in the article to understand why uptake to the project was poor. Discussions with; staff at the recruitment sites, homeless people using the centres, and reflections by the recruitment lead were used to identify factors that may have contributed to a lack of engagement to the project. The remainder of this section will outline the barriers to engagement and recruitment identified during the recruitment phase.
Level of familiarity with recruiter
The person responsible for recruiting participants volunteered at the service 4 months ahead of recruitment in an attempt to develop an alliance with people who regularly use the centers. While attempts to engage people to the project were made, the 22 h of time spent in each recruitment site appears not to have allowed enough exposure to the recruiter. The recruiter reported that the people using the centres changed on a weekly basis making it difficult to become familiar with those who attend. On the other hand, 2 out of the 3 people who agreed to take part were highly familiar with the recruiter. This may suggest that familiarity with the person recruiting for the project was important for recruitment.
Practicality of participation
The level of uncertainty faced by homeless people could have had a negative impact on their ability to commit to the project. Homeless people are often faced with a unique set of challenges, such as finding food, shelter, and maintaining safety. Finding the time or space to engage in project activities can increase the level of burden experienced by homeless people. In addition, some of the homeless people approached to take part in the project expressed discomfort in borrowing a mobile phone device, in worry that it would become damaged. Living conditions on the streets can make the devices more amenable to theft, breakage, or loss. One study found that homeless people experienced a 56% turnover of phones, in a period of 3 months. Moreover, what may be a simple task of gaining access to a power supply, can be more difficult for the homeless. 17
Public perception of phone use by homeless people
Another issue that may deter homeless people from using technology is the perceived public perception of their use of digital devices. The public’s response to seeing a homeless person with a camera or mobile phone might be judgmental and lead them to question, ‘how the homeless person can afford luxuries, but not afford to pay rent’? These prejudices are based on multiple other misconceptions that homeless people are only ever on the street because they do not have and/or cannot afford a home, that homeless people waste their money, and that many homeless people are actually ‘fakes’. One study found that some people who beg for money on the street were concerned that carrying a seemingly expensive device would negatively affect the amount of donations they received. 18
Poor technical literacy
A study carried out in the US utilising data from 421 homeless adults found that the vast majority (94%) currently owned a mobile phone and more than half currently owned a smartphone. The majority (85%) said that they used a mobile phone daily, 76% used text messaging, and 51% accessed the Internet on their mobile phone. 19 In contrast to these findings, the 10 people formally invited to take part in the study did not have a mobile phone and reported having poor technical literacy. The inability to sufficiently use a smartphone (eg, connecting to the Internet, uploading photos using a web link, and resolving technical issues), may have deterred people from participating.
Anonymity
The study’s data collection procedure, in essence, involved asking participants to share their realities as homeless people. However, it was discovered that the homeless people approached were not always comfortable sharing their day to day experiences. Homeless people may experience harsh and challenging circumstances that they might not be interested or comfortable in sharing. One of the people approached to take part in the study reported fears that family and friends would identify him in the photos and learn about his situation (which they were unaware of), highlighting concerns around anonymity. It should be noted that staff working with homeless people at the respective centres did not perceive anonymity to be an issue prior to the recruitment phase. This suggests that homeless people may feel differently about disclosing aspects of their life or identity.
Dissociation with homeless label
While the people approached to take part in the study were considered to be homeless in line with the definition provided by Parliament in the United Kingdom, 1 this label did not resonate with everyone. One of the people who agreed to participate in the project did not identify with the ‘homeless label’, suggesting that people living on the streets might not always classify themselves in the same way. While the current project did not further explore this, some literature around homelessness suggests that the term ‘homeless’ is loaded with stigma and could lead some to choose to dissociate themselves from this classification.8,20
Discussion
The aim of the pilot study was to explore the feasibility of using photographic data to better understand the day-to-day experiences of homeless people. During the data collection phase, a number of barriers to recruitment emerged as a result of the project’s data collection approach. Recommendations addressing the barriers outlined in the paper are summarized in Table 1.
Barriers and solutions to recruiting homeless people to research using photographs.
While the results suggested that recruitment efforts were largely unfruitful, it was noted that a high level of familiarity with the person responsible for recruitment may facilitate engagement with the research project. For recruitment procedures that are carried out by the research team, we recommend for recruitment staff to show a higher level of presence (more than 22 h) in the recruitment sites. This will enable researchers to become familiar with potential participants and gradually build an alliance. While it might be possible for recruiters to increase their presence before and during the recruitment phase, the amount of time required to develop an alliance with an ever-changing service-user group may be time and resource intensive. It may therefore be more practical to ask staff who frequently work at the centres to recruit for the study, to benefit from long-standing relationships forged by staff prior to the research project.
The findings also suggested that the participation process may have been too complicated and impractical for the homeless population. Given that many homeless people did not have access to a smartphone and/or know how to use one, it may be necessary to provide technical training on the photographic device being used, to enable participants to effectively take and upload their photographs. Furthermore, centralising project processes to 1 site may hold a number of benefits for those taking part. Reducing the number of people and sites that participants have contact with (eg, study team, collaborating services, and Starbucks cafe) will help streamline research processes by reducing the burden of travelling to different locations, on participants.
Issues concerning anonymity can be addressed using the ‘planning identity’ consent procedure. This approach allows participants to select the level of anonymity they would like to have, ranging from non-disclosure to full-disclosure of participant identity and/or identifiable information. Careful planning is therefore required ahead of recruitment, to identify the best approaches to conceal or reveal identities. 12
Future research should aim to work with homeless people and professionals working in homeless services, to understand how best to address issues around anonymity, identity, and public perception of homeless people using technology, to develop more acceptable research procedures. 21
The use of photographic data collection with homeless people may yield a number of advantages. First, photography has widely been reported as an important tool for breaking down communication barriers. The idea that photographs can convey complex messages that are easily understood 12 is famously depicted by a commonly used idiom ‘one picture is worth a thousand words’ coined by an advertising executive Fred Barnard in 1921. 22 Photographs can be a powerful mode of communication which can promote a higher level of understanding of homelessness. Second, photographs can help forge strong alliances between the research team and participants by allowing participants to set the agenda through the photographs that they decide to put forward. This is contrary to the marginalisation that is typically experienced by homeless people in society. 8 Third, the homeless population are at a heightened risk of (physical and mental) health problems, as well as social and occupational exclusion. The use of photographs can help build bridges between the wider community and the homeless population. Photographs can be utilised to tell visual stories to influence public perception, policy makers, and other methods that can lead to actionable programmes, on issues that are important to homeless people.9,12
A number of limitations are noted in relation to the study’s proposed methodology. Experiences of homelessness may vary across gender, geographic locations, type of homelessness, and level of accessibility to support structures. 17 The findings may therefore only be relevant to the small sample of people approached in the project and therefore cannot be generalised to all homeless people. The data collected in the project might be affected by participant reporting biases. Participants may only put forward topics that they are comfortable in sharing. While this might fall in line with the approach of allowing participants to set the agenda in research, the data generated might only reflect what the participant is willing to share and not the issues that truly matter to them. 9
Finally, the barriers to engagement proposed in this article are based on limited interaction with homeless people and staff working in the 2 homeless/community centres. While we have attempted to increase reliability by triangulate our findings with the existing body of literature on homelessness and/or photography data collection, it is possible that other important contributing factors to recruitment were not picked up on.
Conclusion
Photographic data collection offers an inclusive approach to working with marginalised groups such as homeless people. It also offers opportunities to change public perceptions and policies on topics affecting marginalised groups. To effectively utilise this approach, co-production between the research team and homeless people should be adopted to develop procedures that address the barriers highlighted in the article (level of familiarity with the recruiter, practicalities of participation, public perception of phone use, poor technical literacy, anonymity, and disassociation with the homeless label). Furthermore, future research using photographs should aim to develop a clear plan for public and policy engagement (eg, knowing what, who, and when to influence) to enhance the impact of the research findings.23,24
Footnotes
Funding:
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests:
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Author Contributions
AD, XW, BK, FS and BI Conception or design of the work; BI data collection; BI, AD, XW data analysis and interpretation; AD, XW drafting the article; AD, BI, BK critical revision of the article; AD, XW, BK, FS and BI final approval of the version to be published.