Availability sampling is a nonprobability sampling technique that does not use random selection for choosing its population of interest; instead it selects its participants because they are available and convenient (Rubin & Babbie, 2013). Participants will be referred by Passageways Intake Center, a Union Station Homeless Services program in the City of Pasadena. Passageways, is the entry-point to the homeless services network in Pasadena. They provide intake, assessment, medical, mental health services, as well as referrals to shelters and rehabilitation programs to homeless individuals in Service Planning Area (SPA) 3. Passageways agency staff will refer eligible clients (chronically homeless adults with co-occurring substance abuse disorders and mental illness diagnoses) that are interested in participating in this study and becoming stably housed. The participants will be placed into housing in January and will be provide with shelter during the coldest part of the year. Homeless shelters in the winter months typically have an increase of clients due to the weather. Having shelter during the winter months may be an incentive to engage willing participants for this study. Also participants will be given $50 gift cards every four months their posttests (interviews and …show more content…
If done properly this design can help protect against most internal validity threats, by having a group that will receive the HF integrated service model intervention and another group that will not. This allows the study to control for the possibility that other factors not related to the intervention are responsible for any differences between the pretest, posttest and follow up test results. This research design uses random assignment with both the control and treatment conditions. By using random assignment, both groups will be equivalent so that it can be presumed that the differences seen may be largely due to the treatment. Due to having a control group, this design can eliminate history, instrumentation, maturation, and testing conditions as an internal validity threat. Due to having random assignment, selection, regression of the mean, and interactions can be ruled out as an internal validity