Description
Evaluation research should be influential, helping to guide behavior, shape policy, and inform decision-making activities. For evaluation research to wield its influence, findings must be communicated in a way that enhances the knowledge of relevant stakeholders. However, previous research on dissemination has only begun to shed light on the effects of various dissemination techniques across multiple stakeholder groups. This study extends this knowledge on dissemination by describing the path dissemination takes, assessing the influence of, and preference for, different dissemination techniques, and how reported influence and preferences differ between stakeholder groups. This study also examines individuals' beliefs regarding the evaluation and following the dissemination of findings. Representatives from San Diego County mental health treatment programs (n = 77), Behavioral Health Services division of the Health and Human Services Agency (CMH; n = 4), and the Health Services Research Center (HSRC; n = 4) were interviewed to assess the influence of varying dissemination techniques used to communicate evaluation research findings. The dissemination techniques included an 88-page comprehensive report, a 36-page shortened report, two single paged summaries of findings, and oral presentation of findings during regularly held meetings. Multiple regression analyses for nested (clustered) data were used to assess the overall influence of, and preferences for, each dissemination technique. Additionally, these analyses were used to assess potential differences in influences and preferences corresponding to stakeholder group membership, and stakeholder involvement in the evaluation. Results suggest representatives from CMH report significantly more involvement in the evaluation than program representatives (?^ = 1.77, z = 8.87, p < .0005). Additionally, the influence of meetings was significantly greater than the influence of the documents, t(16) = -3.93, p = .001, which did not differ significantly by stakeholder group, t(16) = 0.95, p = .358. Findings also suggest the reported appeal of the shortened comprehensive report was significantly greater than the appeal of the full-length comprehensive report. When asked about preferences for dissemination techniques, participants rated the 36-page shortened report significantly more appealing than the 88-page comprehensive report, t(7) = -8.52, p < .0005), which did not differ significantly by stakeholder group using family-wise alpha protection, t(7) = -3.95, p = .008. Findings from this study suggest that, regardless of participant type, meetings supplemented by shortened reports are the best method of dissemination for maximizing influence of findings. These results are consistent with research on passive diffusion, which has been shown to be ineffective and unlikely to result in influence. Should influence of evaluation research be the primary objective, an active dissemination approach is needed.
