Scottish Government

Appendix 3: Sample Recruitment for the Outcome Evaluation

As a preliminary to the interviews, members of the research team visited each of the agencies in order to agree a set of recruitment procedures. At these meetings the agencies were invited to express their preferences regarding the process of recruitment, taking into account their own convenience and workloads and the likelihood of securing clients' agreement to participate in the research. In every case, the project's preferences were adopted as the recruitment procedure for that agency. Reflecting the agencies' varying preferences and circumstances, a different procedure was agreed for each of them. However, in every case, the recruitment procedure that was agreed involved the participation of the agency; a) in contacting parents for permission to approach their child; b) in introducing the research to clients; or c) both of these. All of the documentation required for the recruitment of clients was prepared by staff at the Centre for Drug Misuse Research and delivered to the agencies. This included; a detailed description of the recruitment procedure and the agency's role in it; information sheets for clients and parents describing the research and what their participation in it would involve; letters to clients and parents introducing the research and requesting their consent; and consent forms. The agreed procedures are set out in Appendix 5.

At the agency visits, the number of each project's current cases and their rate of recruitment of new cases were established and recruitment targets agreed. We agreed with the agencies that we would seek to interview all of their current cases together with all new cases over what we anticipated would be a 6 week recruitment period. The rate at which the projects recruited new cases dictated that the samples would have a clear bias towards current cases. The numbers agreed for each of the projects was as follows.

The combined number of potential interviews for the four agencies was 111. While we did not anticipate being able to recruit all of those cases, the agencies were confident that, on the basis of the agreed procedures, a high response rate could be achieved. In the event, the recruitment of clients for interview proved to be frustratingly difficult. Instead of being accomplished within a period of 6 weeks, the recruitment process took nearly 5 months.

The delay was caused by the severe difficulties which the agencies reported experiencing in making contact with clients or their parents and in securing the various consents. This was partly, they said, a product of competing priorities and disruptions to staffing. In addition, though, recruitment difficulties were undoubtedly compounded by the vulnerable nature of the client groups. For example, the Aberlour project workers had to choose their moment carefully in deciding when to introduce the research to their clients since they could frequently be under the influence of psychoactive substances. Similarly, an initial lack of response to letters on the part of parents/carers meant that the East Ayrshire and Reiver project staff had to follow them up with telephone calls. In the case of the East Ayrshire parents, their misuse of drugs or alcohol made securing a response especially difficult. Perth Connect experienced a particular problem with their clients many of whom were profoundly disaffected as a result of what they saw as the unjustified departure of an especially popular worker. In consequence, many of the young people at this project expressed the view that there was no point in them participating in the research since nobody appeared to value their views anyway.

The original design of the outcome evaluation was for three interviews to be conducted with each client at 6 monthly intervals. However, as a result of the above delays, the research team's advisory group agreed; 1) to extend the period of data collection and 2) contingent upon this, to reduce the number of follow up interviews from 2 to 1 with the second round of interviewing taking place 8 months after the first one. Table 2 shows the final response rate for the first round of interviews together with the number of refusals and non-responses.

* The contrast between the potential client numbers and the number contacted by the agencies is explained by the fact that the number of potential clients fell over the course of the recruitment period. According to the agencies, they contacted all of the clients who were available for inclusion in the study.

** When we visited the project initially we were informed that they would have 29 potential clients. However, this figure was subsequently revised downwards when we were later informed by the Project that they had only 15 active cases.

*** We assume that all 15 potential cases were contacted although we were unable to obtain confirmation of this from the agency.

Of the 87 potential interviewees actually contacted by the agencies, interviews were conducted with 42 clients, a response rate of 48 %. This might appear to be a disappointing return, but it needs to be viewed in the context of the nature of the clientele whose participation was being solicited and the particular difficulties associated with Perth Connect. It should also be noted that a major reason for the low rate of recruitment was the non-response of parents at Reiver and East Ayrshire.

On a positive note, a high proportion of the clients (37) participated in the second round of interviews. This included 12 clients of Aberlour Outreach, 11 clients of the Reiver Project, 10 clients of the East Ayrshire Young Carers Project and 4 clients of Perth Connect. The 5 clients who did not participate in the second round of interviews included 2 who had moved and could not be traced and 1 for whom it proved impossible to arrange an interview despite repeated attempts to do so. One of the Aberlour clients declined to be interviewed on the second occasion while another had serious mental health problems and it was deemed inappropriate to attempt to interview her.