Introduction:Methods to assess evidence and to use that evidence to inform practices and policies are under developed in the area of social services. Although health professions have developed robust methods in recent decades to collect, analyze and synthesize scientific evidence and to inform clinical recommendations, the use of these methods often remains difficult in social services. A taskforce was implemented to address this and to propose a method that may be more appropriate for the social sciences.Methods:The project was comprised of four steps: (i) performing a qualitative review of discussions between experts, (ii) designing a cognitive map of the data, (iii) conducting a systematic literature search, and (iv) comparing the data from the meetings with experts with the scientific literature. These steps were completed using the grounded theory approach. In order to test the method developed, focus groups were then conducted and four case studies were used to assess the evidence and provide recommendations for youths with mental health problems and for elderly care.Results:Although robust scientific data remain crucial when developing recommendations for practice, results showed that these data are incomplete if considered alone, and that contextual (circumstances in which the intervention is delivered) and experiential data (how the intervention is perceived by stakeholders) must also be taken into consideration. A method to triangulate these three types of data is proposed. Using this technique, the value of the data is established by means of various measurements that converge towards the same result or that provide a consistent overall picture or some important nuances that need to be considered, as illustrated by the four case studies.Conclusions:The proposed method can be used to address the limitations that are inherent to the use of techniques and procedures drawn from the medical field when assessing evidence and developing recommendations for the social sciences. The case studies that the proposed method is not only a viable option to methods drawn from medicine, but also adds to the quality of the recommendations that are made and is more congruent with the epistemology of social sciences.
How to get at Causality in the Social Sciences: Multiple Regressions Versus Case Studies
