The symposium in 2002 started a discussion about combining research activities from different fields. It also opened the discussion for alternative research strategies, e.g., coming from the action research tradition and related approaches. The discussion always reflected two important issues: The potential of these approaches for classroom development and the question of criteria of research quality in this kind of research.This aspect is the central issue of the symposium in 2004. Presentations had been selected offering different solutions for the question how to make research in science education practice-oriented and on a high level of quality. Presentations in the symposium of 2004 were about projects based on different forms of action research, 'Developmental Research', teachers' 'Learning Communities', or 'Content Focussed Coaching'.Looking at these different approaches asks for reflection about the underlying research methodology, data sampling and treatment. It asks for thinking about quality criteria as coming from classical empirical research, e. g., validity and reliability, whether they can be applied here. But, it also asks whether there are other criteria for another kind of research. These criteria may come from a critical paradigm of research, e.g., practice improvement and professional development, or from the discussion about quality criteria in interpretative research, e.g., value, authenticity, relevance, or importance. Thus, the question of quality is directly connected to how the results of research reach and influence practice, how findings from research affect the practitioners' field of activity.