Research on homelessness in Comparative Perspective

The CUHP network held its concluding conference in Brussels on November 3-4 2005. The conference presented the most important conclusions of this 3-year thematic network and dealt with the issues of defining homelessness, ethics and theoretical approaches, quantitative and qualitative methods. An introductory paper reported on European values and housing and homeless populations and the perspective of NGOs working with the homeless. Each paper was discussed by experts from outside the network, either from other European countries and from the USA. Simultaneous translation was provided in French, English and Spanish.

Session 1: Housing and Homelessness in a European context

The morning of the first day was chaired by Fintan Farrell (European Anti-Poverty Network - EAPN) with an opening address by Francis Wurtz, MEP.

The first paper was a review of the work of CUHP over the previous three years written on behalf of the INED team by Efi Markou (Administrator) and Maryse Marpsat (Principal Scientific Officer of CUHP).

The second paper was in two parts. The first part, presented by the Hungarian team, discussed housing issues and housing policies in the context of European values, and the relation between housing policies and homelessness. The second part, from the UK team, analysed the contributions of NGOs that participated in CUHP workshops in relation to their perspective on homelessness, new homeless populations, developments in services, and research agendas.

- ’European values and homeless policies: Examining the broader and narrower policy lines’ by Eszter Somogyi and Ivan Tosics (Metropolitan Research Institute, HU) and Joan Smith (Cities Institute, London Metropolitan University, UK)

The discussant for this session was Ingrid Sahlins, University of Gotenburg.

- Comment: Ingrid Sahlin (Gotenburg University, SE)

Session 2: Defining homelessness, theoretical approaches and ethical questions

The afternoon of the first day was chaired by Freek Spinnewijn (FEANTSA)

Definitions

Defininitions of homelessness was a major question for the thematic network and the synthesis paper compared three definitions fully explored during the network: the statutory definition of the UK, the social exclusion definition of Italy, and the operational definition developed in France for use in social surveys. The conclusion was that divergence of definition is based on three factors: whether legislation on homelessness exists or not, the linkage between social housing allocations and the definition of homeless people, and whether homelessness is considered multi-dimensional or housing focussed. The paper also discussed the possible European wide definitions including that proposed within FEANTSA which excludes some categories of people (the majority of whom are women) defined as homeless under the UK legislation. The paper also discusses uses that should be addressed by the LAEKEN indicator.

- ’The problem of definitions: Points of similarity and differences’ by Maryse Marpsat (Institut National d’Etudes Démographiques, FR)

Roser Cusso, commentating on this paper, argued that one way to negotiate between national definitions is to take a Rawls perspective and ask the question – Do these people deserve aid and what can we do to serve them ? Martha Burt, Urban Institute – US, argued for taking a common minimum definition (’without housing last night’) and then layering further definitions on top. This would achieve common homelessness figures for the narrowest definitions up to the broadest. However, European researchers argued that in countries without adequate shelters the numbers would differ from reality. An alternative is to compare the development of policies across countries..

- Comment : Roser Cusso (Université Libre de Bruxelles, BE)

Theories of homelessness and research ethics

Antonio Tosi presented the paper on theoretical approaches to homelessness. He presented the possibility of different typologies. First the ‘literal’ homeless facing extreme housing exclusion and extreme poverty including social marginalisation in all its forms. What proceeds and what surrounds this extreme version of homelessness ? But there is also temporary homelessness as there is temporary poverty. He argued for the application of paradigms of poverty to the study of homelessness, rather than housing exclusion approaches.

- ’Homelessness as a process: Theoretical approaches and social construction of the question’ by Rossana Torri and Antonio Tosi (Dipartimento di Architettura e Pianificazione, Politecnico di Milano, IT)

Jean Marie Firdion described the scientific survey approach undertaking by INED to establish the number and social origin of homeless people in France, and the ethical assumptions made for that study. The study used the theoretical approach of Pierre Bourdieu and the work of US sociologists on early life events. The team were concerned to link individual experience with structural effects. The central approach was to use the same methods and theories to understand the homeless population as used for the non-homeless population

- ’Ethics, concepts, and methodology’ by Jean-Marie Firdion (Institut National d’Etudes Démographiques, FR)

Pascale Pichon was the discussant for both papers and raised three issues: To what extent does the dynamic understanding of poverty also enable us to understand homelessness as a non-linear process?. If homelessness is non-linear then it is important to focus on the experience of homeless people, their reduction in choices, and their ‘career’. Theoretical choices are the foundation of the researcher’s scientific position.
Marti Burt, Urban Institute – US, led the discussion. She argued that whether homelessness was a process or a spiral most people escape and it was important to focus resources (as in the US) on those homeless more than two weeks on the street. The French team reported, however, an increasing precariousness amongst the homeless – more youth, more women, more unemployed and even some working poor – although they have not enough data on this. Ingrid Sahlin argued that through ‘life-studies’ it is possible to cover a great deal of time. Lia Van Doorn reported on her study of 64 long-term homeless – half are doing well but half are not and at risk of falling back into homelessness. Luidi Leonori argued that homelessness and exclusion are two different things – the people who lack a home are not the same as those who are excluded. Pascale Pichon argued that exclusion is not homelessness but we need to understand the specificities of homelessness and have common conceptual tools through which to approach homelessness.

- Comment: Pascale Pichon (Université Jean Monnet, Saint-Etienne, FR)

Session 3: Quantitative and Qualitative Methods

The Friday morning session was chaired by Nicolas Razafindratsima (Institut National d’Etudes Démographiques, FR)

Quantitative methods

The joint paper from the Danish and Spanish teams looked at two methods used for undertaking cross-sectional studies and longitudinal studies. The Spanish team had used survey and interview-based studies for both cross-sectional studies and longitudinal studies. The Spanish team stressed the importance of sampling strategies and registering who is missing from subsequent sweeps, and the use of comparative data (from Los Angeles and Washington D.C., US, Paris and London). The Danish Social Research Institute had used a citizenship registration system to undertake both cross-sectional and longitudinal studies using register data. Only those using specific types of shelter are registered but registers are useful for longitudinal research making it possible to reconstruct a life-course of movements in and out of homelessness as well as identify background causes and the effects of intervention.

- ’Quantitative methods in Homelessness Studies: A critical guide and recommendations’ by Lars Benjaminsen (The Danish National Institute of Social Research, DK), Manuel Munoz, Sonia Panadero and Carmelo Vázquez (Universidad Complutense de Madrid, ES)

The discussion of this paper was led by Luigi Leonori (Sante Mentale Exclusion Sociale -Europe, BE) from who reported on a 10 European capital survey of users of homeless services, of whom 30% had mental health issues. He argued that a new and important issue was homelessness among immigrants and migrants, especially illegal immigrants with no access to services. Conference delegates reported problems of accessing information on illegal immigrants because NGOs were wary of offering this data. This was confirmed by the Danish team in relation to their register data.
A US delegate reported over 70% response rate in their longitudinal research and a Spanish study had tried the same methods as US and achieved over 50%. One delegate raised the question of the cost of quantitative methods for the results obtained.

Qualitative Methods

The joint paper from the Dutch and UK teams reported on five qualitative studies. Lia Van Doorn (NL) and Megan Ravenhill (UK) used different qualitative methods to understand the biographies of homeless people and their situation on the street, and on leaving the street, and their relationships with the institutional environment. Petra Van Leeven (NL) worked with agencies and clients to create a holistic action-planning method for Dutch NGOS. Svetlana Stephenson (UK) used both qualitative and quantitative interview methods to study the street lives of children in Moscow, as did Joan Smith (UK) to research the family background of young homeless people in the UK. The results of the latter study were used in the design of an early intervention action-research programme in the UK. The paper demonstrated that qualitative studies undertaken independently in different countries produced parallel findings in relation to similar homeless populations. It also showed that the findings of qualitative studies could be used by NGOs in their work.

- ’A European perspective on the use of qualitative methods in the study of homeless-ness: methods, findings, theoretical understandings and policy interventions’ by Megan Ravenhill, Joan Smith (Cities Institute, London Metropolitan University) and Svetlana Stevenson (DASS, London Metropolitan University), London Metropolitan University, UK and Lia Van Doorn and Petra Van Leeuwen (NIZW, NL)

Martin Gruber was the commentator for this paper and discussed how his research in Hamburg paralleled the findings of Megan Ravenhill (UK) and Lia Van Doorn (NL). All the studies relied on theoretic sampling, not representative sampling, but their findings are coherent across countries. Scientific research is also driven by it’s assumptions and the important issue is to be explicit about the epistemological framework that has been adopted. The remarkable similarity is that they have focussed on individual factors in relation to homelessness and all are policy orientated.

- Comment: Martin Gruber (Hamburg University, GE)

The complementarity of quantitative and qualitative research

Martine Quaglia from INED reported on her ten years of field work with INED including both quantitative research and qualitative studies. She discussed the different roles of large representative samples that allow for modelling populations and qualitative studies of agencies and the lives of individuals and the relationship between the whole and the parts of the whole. She also raised the question of how different research was funded in different countries – some countries did not have the financial resources to undertake qualitative research.

- ’Qualitative vs quantitative. Why are these two approaches appropriate to understand homelessness?’ by Martine Quaglia (Institut National d’Etudes Démographiques, FR)

Paul Koegel of Rand Health, US commentated on all papers reporting that his own work on homelessness in Los Angeles covered both qualitative studies and quantitative (both survey and register, cross-sectional and longitudinal). The only work not found in Europe is the rigorous randomised trials used for evaluating programmes in the US. He argued taht quantitative and qualitative methods are complementary, together they give a more complete story and this is also true of randomised clinical trials. But there is no simple relationship between theoretical and methodological models.
Antonio Tosi (It) argued that both quantitative and qualitative research are needed to establish the critical links between the macro and micro in understanding homelessness. Moreover both types of methods have to be applied not just to understanding homelessness but to understanding households that are precariously poised in the housing and labour markets. There was a discussion on what was meant by the term ‘culture’ of homeless people? Was it just an adaptation to living on the streets? Not just adaptation but a way of life (Ravenhill, UK). However, it is important not to call this a ‘culture’ because many do not join in, they are lone wolves (Sahlins, Sweden). This is a similar problem to the term ‘culture of poverty’ – it is important to have semantic that does not lock people into homelessness (Koegel, US).

- Comment: Paul Koegel (Rand Health, USA)

Session 4: Experience from other countries

This session was chaired by Hugues Feltesse (DG Employment, Social Affairs and Equal Opportunity)

Martha Burt, Urban Institute, US, opened this session by reporting on US approaches to ending long term homelessness, especially for people with disabilities by 2012. She reported the US definition of ‘chronic homelessness’ (more than 1 year or 4 spells in 3 years) and that half of long term homeless had 3 or more disabilities including mental illness, addictions or physical ill-health, and these used a lot of resources (hospitals, police, prisons). In the US they have established this is a finite population through point-in-time surveys and annual data. Evaluation research in New York and Philadelphia has established that Permanent Supported Housing works if people have low demands placed on them and there is support. Preventive measures include not allowing people to leave jails without housing. Evaluation studies have shown these measures work.

- ’US approaches to ending long-term homelessness for people with disabilities’ by Martha Burt (Urban Institute, USA)

Iskra Dandolova reported that the no policy on homelessness in Bulgaria might be a policy. As with other Eastern European countries homelessness was not recognised in Bulgaria as it was assumed that the state produced housing and employment for all. There were homeless people before 1989 but no term for them or restricted to those sleeping rough, and this understanding has continued. The term has not been extended to include the houseless (forced living in shared housing), insecure housing, and overcrowded and poor physical conditions. There are no state institutions or specialised services and no strategy for prevention or protection and no legal recognition of homelessness. In a typology of homeless conditions (15) it is important to include the potentially homeless (temporary shelter), the concealed (enforced sharing) and the obvious (roofless or rough sleeping). Young people are without hope of escaping from the situation. Very limited services through new NGOs.

- ’The case of Bulgaria’ by Iskra Dandolova (Institut of Sociolgy, Bulgarian Academy of Sciences)

Peter Gyorgi, Metropolitan Research Institute, Hungary, was the discussant for these papers. Some respondents in Hungary report they have been homeless for two years but they lost their apartment 10 years before. Therefore homelessness is a different thing for policy makers, service providers and homeless people. There is a difference between describing a social phenomenon and measuring it, and a problem of bring together the views of different actors. The real question is: Who are the people who can’t make ends meet for themselves and their families from their own means and need services and institutions. This is also different from describing and measuring the problem.
In the discussion it was argued that measurement is in relation to policy objectives, or to a lack of policy objectives. Public spending raises the question of who deserves to have money spent on them? Are immigrants less deserving than others? In London where greater options have been created for the homeless there are more sanctions (including repressive sanctions) as well. Another question was - How is it that US authorities are listening to their researchers? Koegel argued that it is comfortable for US politicians to give support to street homeless, not all homeless, because they have continued to cut housing subsidies. It was reported some people are dealing with their own poverty and being on the edge of homelessness by migrating from Britain and Ireland to Bulgaria where housing is cheap.

The Round Table

A round table was coordinated by Maryse Marpsat (CUHP) and included Martha Burt, Iskra Dandolova, Martin Gruber, Paul Koegel, Ingrid Sahlin
Maryse Marpsat posed three questions for the final round table of external experts – what has appeared important, what has been missing, any messages for homeless research. The replies from the six experts are given in the brief report below.

- Round Table Discussion