Funding agencies and policy-makers have put increasing pressure on scientists to better clarify the usefulness of their research. It has been suggested that this may have led to an increased reflection on the societal relevance of research among the scientists themselves. However, this often is more an assumption than a carefully verified fact. This paper investigates whether reflection on societal relevance actually occurs and has a measurable effect on the choice of research and on dissemination activities performed by scientists. A survey was conducted among researchers in environmental science and technology at Swedish universities. Results show that researchers do frequently reflect upon the societal and environmental relevance of their work. We used path modelling techniques to assess how this influences their professional activities. Results show that reflection was important to explain both the choice of research and dissemination activities. Variables such as individual attitudes, experience and collaboration with external actors also affected these outcomes.