Designing and Conducting Single System Research on Social Work Practice

Single system research designs (SSRDs) have been used within the field of social work since the middle 1960s. Many hundreds of examples of this evaluation methodology have appeared in a wide variety of social work journals and reflect empirical evaluation efforts based on diverse practice perspectives. The presenter has published a large number of SSRDs in social work and will use this background to illustrate the practical value this methodology has in helping practitioners empirically evaluate the outcomes of their own practice. The workshop will provide an overview of the historical origins of SSRDs outside of the field of social work and of their introduction to our field over 50 years ago. The basic assumptions that this research method is based upon will be examined, and the relatively simple prerequisites needed to undertake a SSRD in one's own practice will be described. These include the need to select one or more reliable and valid outcome measures that can be repeatedly assessed over time, to collect such data, to portray the data on a line graph, and to visually draw any inferences that may be derived from this graphic display of the data. Simple SSRDs such as the B and AB designs may be used to answer simple questions, such as Did my client improve over the course of treatment? Or Did any improvements last over time? More complex questions involving causal inference may be answered by using SSRDs of greater internal validity such as the ABA, ABAB, multiple-baseline, and alternating treatments designs. In some practice settings, SSRDs can be used to compare the effects of one treatment versus another, within a given client. These designs can also be used with larger groups of client systems, such as therapy groups, couples, organizations, and communities. At the macro-level, the logic of SSRDs combined with inferential statistics such as time-series analysis is often used to evaluate the outcomes of public health and social welfare policies. The presenter will describe the fundamentals of SSRDs, and illustrate their actual use in clinical and macro practice, drawing upon his own published work and that of others, focusing on the social work literature. Real life examples will be used, initially reflecting simple designs, and then moving on to more sophisticated studies permitting legitimate causal inferences. Suggestions on how to successfully write up SSRDs for publication will be provided, as will handouts describing internet resources useful in learning how to create the graphs needed to display SSRD data. Numerous examples will be provided on how SSRDs have been used by social work doctoral students to complete dissertation research requirements. The limitations of this approach to practice evaluation will be discussed, and how some of these limitations may be overcome. The use of SSRDs for the purposes of practice evaluation will be linked to relevant standards within the NASW Code of Ethics and the CSWE EPAS. Attention will be given to the ethical application of SSRDs.