« Previous | Contents | Next »
Listen
Tracking Homelessness: A Feasibility Study
CHAPTER SIX: RECOMMENDATIONS
6.1 As was outlined in Chapter 2, there are two main reasons why longitudinal research on homelessness is required from a policy perspective: to aid our general understanding of routes in and out of homelessness; and to develop more robust evaluations of particular projects and initiatives aimed at tackling homelessness. Our key recommendations to the HMG are presented below under these two main headings. In addition, some technical recommendations are made in relation to enhancing the feasibility and/or effectiveness of a range of methods for undertaking longitudinal research with homeless people.
GENERAL UNDERSTANDING OF HOMELESSNESS
6.2 We strongly recommend that the HMG consider pursuing a 'full retrospective' study of ex-homeless people identified in the SHS. Such work could potentially have great policy value in identifying effective 'routes out' of homelessness, and would be relatively inexpensive to undertake. Two approaches could be taken here:
- additional questions related to homelessness could be included in the SHS. These would generate valuable quantitative information on routes out of homelessness;
- a sample of the ex-homeless people in the SHS could be followed-up for a qualitative study to investigate their routes out of homelessness in more detail.
6.3 We also recommend that the HMG press for appropriate questions to be added to the SHS to enable the extent and nature of 'hidden homelessness' to be investigated. This data is likely to be extremely valuable from a policy perspective; both in highlighting the situation of an often overlooked homeless group, and in enabling the potential additional demand for accommodation resulting from the 2001 Housing (Scotland) Act and Homelessness Bill to be estimated more accurately.
6.4 We also recommend that the HMG encourage the relevant authorities to undertake 'full prospective' longitudinal studies of at risk groups, with the highest priority given to young people leaving care. Such 'tracking' research would be expensive, but is likely to have significant benefits not only for the development of homelessness prevention policies, but also in facilitating 'holistic' responses to the multitude of problems these vulnerable groups face. If these studies are not undertaken, attempts should be made to track at risk people through administrative data for later qualitative interviews.
6.5 'Semi-retrospective' tracking studies of a range of homeless groups could be undertaken using administrative data from the HL1 returns, the RSI monitoring database and the Shelter database to obtain the relevant samples. However, such an approach would be expensive, and the policy applications are less obvious than those of some of the other approaches recommended here. Specific research questions would have to be identified which this method would be cost-effective in answering before it would be sensible to proceed. Also, mechanisms to obtain the required consent to satisfy the Data Protection Act 1998 would have to be developed in consultation with the relevant holders of the dataset.
MORE ROBUST EVALUATIONS TO ASSESS LONG-TERM IMPACTS OF INTERVENTIONS
6.6 We strongly recommend that the HMG encourage the development of longitudinal evaluations of specific programmes and projects to enable their long-term benefits to service users to be assessed. It is particularly recommended that some components of these evaluations should be standardised to allow comparison across programmes and projects.
6.7 We would recommend that a substantial longitudinal evaluation of the Glasgow Hostel Closure programme should be pursued along the lines suggested in this report. This is not only because of the intrinsic value of the data produced for the rolling programme of re-provisioning currently underway in Glasgow, but also because the relatively large numbers involved would enable robust quantitative comparisons of outcomes to be undertaken.
6.8 We strongly recommend that local authorities be asked to report in their homelessness strategies how they will use/encourage the use of longitudinal methods to assess the impacts of projects and policies they have pursued in their area to tackle homelessness.
TECHNICAL RECOMMENDATIONS ON IMPROVING FEASIBILITY / EFFECTIVENESS OF LONGITUDINAL RESEARCH
6.9 The lessons learned in previous longitudinal studies in both the US and UK with regard to effective tracking of homeless people should be borne in mind when developing any future research projects with a prospective element.
6.10 In order to use administrative data to assist in maintaining contact with any sample of homeless people, Data Protection issues should be resolved with any of the relevant parties. Appropriate 'permission to locate' and/or consent forms should be agreed in advance.
6.11 The possibility of increasing the proportion of people that give permission in the SHS to be re-contacted should be investigated as a matter of priority. Other potential wordings of the consent question should be tried in the pilots for the survey.
« Previous | Contents | Next »