Developing knowledge for practice: Working through "trench--bench" partnerships

Abstract

Social work has long struggled with the challenge of ensuring that products of its research arm are useful to, and used by, its practice arm. Articles in this issue exemplify one of the most promising means to this important end: conducting research about and in partnership with practice. In many areas of biological and social sciences and in medicine, knowledge development has traditionally developed from to trench. The academic setting has been the where hypotheses and research questions are formulated, rooted (ideally) in theory and prior published research. Practice or policy has been the to which knowledge dissemination is directed through last, but all-too-often neglected, steps in the research process. The sufficiency of this approach for social work is challenged by persistent evidence that research findings have very little uptake by practice, and that new practices are rarely sustained in agency settings after an intervention trial or prevention program ends. If the production of useful research knowledge is our goal, as I think most readers would agree, we need to conduct our research differently. Bench to will not suffice; rather we need bench-trench partnerships, and may, in fact, need to more often work trench, to bench, and back to trench. Academic-practice partnerships need to be cultivated, respected, and nurtured from the outset of the research process. Both and trench give rise to research questions. We are familiar with the bench-generated research questions--those that arise from gaps in prior knowledge and represent the next logical step in an incremental progression of knowledge. With regard to knowledge for practice, this may be reflected in the progression of epidemiological research giving rise to clinical epidemiological studies, which then inform research (Proctor, 2003), which in turn informs the testing of interventions' generalizability to new settings and populations. Social work needs this kind of careful, incremental knowledge development, and we need a strong cadre of careful and committed researchers. Yet as articles in this issue demonstrate, practice-research, or bench-trench, partnerships can help generate research questions that have high priority to agencies. In the lead article, Harris and Franklin address the challenge of helping adolescent Hispanic mothers address the problems associated with their high pregnancy rates that jeopardize their educational success. Their study reflects the respective and mutual contributions of research and practice expertise to the conception, conduct, and completion of this important study. The study methodology reflects many strengths of a bench perspective, including use of a randomized experimental design, standardized measures, treatment manual, and follow-up assessment of effects. Yet, the benefits of a trench-bench partnership are equally apparent: The lead author brought 25 years of experience in family practice to implementing and monitoring the fidelity of the intervention; cultural values and norms informed the intervention and its delivery; and school administrators were involved in selecting the specific school sites for the study. Bench-trench partnership is also apparent in Baker and colleagues' study of recidivism at a shelter for run-away adolescents. In this intervention-informative research, the study team sought to gain information that could subsequently shape the development of interventions to reduce recidivism. Following Rothman's (1989) intervention-research model, researchers worked closely with agency staff in formulating the study questions and shaping the research design. Agency staff wondered if youths' histories of running away affected recidivism; thus practice observations helped to determine an independent variable for examination through the research. Existing administrative data, including that available from the agency's intake form, were used to measure some variables, thereby reducing some of the study's potential burden on agency staff. …