« Previous | Contents | Next »
Listen
Tracking Homelessness: A Feasibility Study
ANNEX 3:DESIGN FOR A QUALITATIVE STUDY OF PEOPLE WITH EXPERIENCE OF HOMELESSNESS
A3.1 Given the sensitive nature of homelessness and the fact that the sample of interest may be scattered across Scotland, individual interviews would probably be required rather than focus groups. However, it might be that some focus group work targeted at finding out how people broke the cycle of homelessness would be useful and could be organised in Glasgow and Edinburgh where there are greater concentrations of individuals with experience of homelessness.
A3.2 Based on the figures from the SHS 2001, in any year there would be about 400 people with experience of homelessness in the past 5 years and 100 with experience of homelessness prior to this. Of these, about 70% would have consented to a follow up interview, giving a potential sample of over 250 people with experience of homelessness in the past 5 years and 70 people with experience of homelessness more than 5 years ago from each year of the SHS.
A3.3 Information to select the sample for the qualitative study could be obtained from the SHS directly. At present the questions in the SHS would allow one to identify whether a person has experience of sleeping rough, if they have ever applied to a local authority as homeless and whether they are currently on a council or housing association waiting list. In addition, there is a range of demographic information. To select a qualitative sample that covers a more detailed range of circumstances, it would probably be necessary to either add questions to the SHS, or to undertake a screening interview.
A3.4 A screening interview among those identified as ex-homeless (and who have consented to the recall question) could be based on brief telephone (if phone numbers are available) or doorstep interviews. The screening interview could collect additional information relevant to the selection of the purposive sample, such as how long people were homeless, whether they were ever in contact with homelessness agencies etc.
A3.5 Whichever approach was taken, an initial study should be undertaken with around 40 people, setting quotas relating to key variables of interest. With qualitative interviewing, a relatively small sample such as this could be drawn initially, and then supplemented if that later appeared necessary - i.e. should respondents continue to provide new ideas and information of interest as the interviews progressed.
A3.6 Interviews would likely be fairly lengthy including housing histories, causes of homelessness, the homelessness experience, routes out of homelessness, and the stability of current situation. Thus interviews of one or two hours would probably be required, and incentive payments should be offered.
« Previous | Contents | Next »