Combined use of qualitative and quantitative approaches to program evaluation is a strategy which is attracting some interest in evaluation circles. However, its applications are few and its rules, related to the choice of devices and data analysis, are still poorly defined. In this article, the authors present an application of a triangulation strategy evaluating the effects of a child abuse and neglect prevention program. Using objectivation and generalisation techniques, the qualitative data then become comparable to the quantitative data. This comparison makes it possible to better interpret the results, raise new hypotheses and identify some unanticipated results.