Over the last decade, the collaboration between researchers and educational organizations have continued to grow exponentially (Fullan, 2007; Sahlberg, 2011). Since the beginning of the 1980s, researchers in educational science, along with politicians, have promoted and emphasized the power of scientific, evidence-based achievement in education (Phillips, 1980) with the sole focus of informing policy makers about scientific evidence on how to improve educational results. (Nordin & Sundberg, 2014; Burns & Köster, 2016). This type of accountability focused approach and science-based discourse in education has been adopted by the Swedish government as well. Since early 2010, the Swedish education act clearly states that all educational policy must be based on research-based evidence as it is written into the Swedish Educational Act (1 chapter 5 §). Diana Stone (2013) notes that “governance emerges...from strategic interactions and partnerships of national and international bureaucracies with non-state actors in the market place and civil society” (p. 113 cited in Radin, 2014). As a result of the move toward more local governance in education, there is, moreover, a complex mix of levels of government, whereby the national government will apply pressure to the municipal level of government, in what Wilkoszewski & Sundby (2016) call softer steering strategies (p.447). Thus, the impact of implementing more scientific methods becomes a mixture of top down and bottom up, where dialogue and sensitivity to local context now comes into focus in a highlighted way.  One of the questions we ask in this research is:

How is the knowledge of school improvement created and negotiated by the different participans in collaborative school projects?

In this paper, we will explore those questions more deeply by looking through the lens of the micro level of local municipal educational policy and how it is enacted at the local school level.The authors of this paper are researcher from a university actively involved in supporting the implementation of research-based policy in a collaboration with the Swedish National Agency for Education (Skolverket) and municipal school boards. Significant for this kind of collaborations and the power-relation that It contains, is that it creates a situation where two institutions often have different roles. One inhabit the role of an agent of change (Pettigrew, 2003, Tajik, 2008),  and in this case it is the university, and to some degree Skolverket [Swedish National Agency for Education]. The role of the schools is that they have been designated as needing improvement. Skolinspektionen [the national school inspection agency] have deemed the schools as having unsatisfactory results or that the local resources to move towards improvment are not sufficient. To frame the process of collaboration, we draw on theories which derives from school development understood as a cycle. The cycle of change, or steady state, consists of six steps: (1) Initial disturbance (2) Feeling of need and decision to do "something" about the need, (3) Diagnosis of  need as a problem, (4) Search for solutions, (5) Application of a possible solution to the need, (6) Satisfaction or dissatisfaction resulting in a repeat of cycle (Havelock, 1973, 1976 in Dedering, Goecke, & Rauh, 2015, p 34). The content and the form of the collaboration is negotiated  in this ongoing cycel, where different forms of power (Reed, 2011) is one dimension, along with the possibility to re-fuse difference in sense making about how to continue. (Alexander, 2011).  We understand sense making  as  "an ongoing accomplishment that emerges from efforts to create order [   ] of what occurs" (Weick, 1993, p 635), with the notion that sense making itself is embedded in structures of meaning (Alexander & Smith, 2003) about school devolpment (Schuller, Jochems, Moos & Van Zanten, 2006; Zeichner, 1995).

The material presented consist of 4 cases. Each school serves as a case. The total amount of observational data from each case consist of 50 hours of observation. 20 hours of individual interviews with the headmaster. 10 hours of focus group interviews with the teachers involved. The qualitative method used can be described as having many of the same elements as participant-observations. Educators and researcher are part of the process as participants along with the school staff. Fieldnotes, documents and personal reflections ( such as impressions, interpretations and feelings) by the researchers are used as data for analysis. Reflections on the interdependability and subjectivity of the participant- researcher is examined and highlighted as part of the data analasis. The method used can also said to have elements of a implementation and evaluation type of reseach. (Håkansson och Sundberg 2012). The reseach is based on situated learning (Borko, 2004), which consider both the individuals involved and the situation within which it takes place, and is thus not easily contained or described in formal terms. In each case, the empirical material: the text (observational notes and transcribed interviews) is sorted and analyzed congruent with the six step of school change cycles (Havelock, 1973, 1976 in Dedering, Goecke, & Rauh, 2015). The process started by identifying critical incidents:"collecting observed incidents having special significance" (Flanagan & Wayne, 1954, p 327). We further included incidents that were highlighted as significant by the participants themselves.

Collaboration between schools, municipalities and researchers from universities can take many different paths, and it is therefore challenging to anticipate how long each part of the six steps in the process may take. The time allotted for this project is two years, but the since the process is an iterate cycle, the path may vary from school to school. New problems and questions arise and may drive the research into a different path, dependent on the needs of the schools. Our interpretation of differences and similarities between the different school may make it challenging to draw conclusions in a broader sense. It thus, becomes vital to identify the various dimensions of power that educators, researchers, teachers and principals use themselves. Equally important to consider is the local knowledge that the teachers and principals inhabit as well as how this knowledge can contribute to the interpretation of the process. Our work will add to the knowledge of the impact of educational policy, and at the same time hopefully contribute to the understanding of how power is used and how school development are made meaningful and negotiated by school development actors involved. It will potentially also play a role in the actualization of research of school development.

Alexander, J. (2004). Cultural Pragmatics: Social Performance between Ritual and Strategy *. Sociological Theory, 22(4), 527-573. Alexander J. & Smith P (2003). The Strong Program in Cultural Sociology: Elements of a Structural Hermeneutics in Alexander, J. (2006). The meanings of social life : A cultural sociology. Oxford ; New York: Oxford University Press. Biesta, G. (2007). Why "what works" won't work: Evidence-based practice and the democratic deficit in educational research. Educational Theory, 57(1), 1-22. https://doi.org/10.1111/j.1741-5446.2006.00241.x Borko, H. (2004). Professional Development and Teacher Learning. Teachers and Teaching, 8(November), 3-15. https://doi.org/10.1080/135406002100000512 Burns T:& Köster.F.( 2016) Modern governance in education. Educational Research and Education. OECD http://dx.doi.org/10.1787/9789264255364-3-en Cochran-Smith, M., & Donnell, K. (2006). Practitioner inquiry: Blurring the boundaries of research and practice. In J. Green, G. Camilli, & P. B. Elmore (Eds.), Handbook of complementary methods in education research (pp. 503-518). Mahwah, NJ: Lawrence Erlbaum. Dedering, K., Goecke, M., & Rauh, M. (2015). Professional background and working practices of consultants in school development: Initial empirical findings from Germany. Journal of Educational Change, 16(1), 27-52. Dewey, J. (1938). Experience and Education. Education, 50(3), 96. https://doi.org/10.1017/CBO9781107415324.004 Emirbayer, M., & Mische, A. (1998). What Is Agency? 1. American Journal of Sociology, 103(4), 962-1023. Flanagan, J., & Dennis, Wayne. (1954). The critical incident technique. Psychological Bulletin, 51(4), 327-358. Freire, P. (2005). Pedagogy of the oppressed- 30th anniversary edition. New York: The Continuum International Publishing Group. Fullan, M. (2007). Achieving Large-Scale Reform. In R. McLean (Ed.), Learning and teaching for the twenty-first century: festschrift for Professor Phillip Hughes (pp. 137-144). New York: Springer US. Hargreaves, A. (1996). Transforming Knowledge: Blurring the Boundaries Between Research, Policy, and Practice 1. Educational Evaluation and Policy Analysis, 18(2), 105-122. OECD, TALIS 2013 Results: An International Perspective on Teaching and Learning. (2014). https://doi.org/10.1787/9789264196261-en Pettigrew, P. J. (2003). Power, Conflicts, and Resolutions: A Change Agent's Perspective on Conducting Action Research within a Multiorganizational Partnership. Systemic Practice and Action Research, 16(6), 375-391. Phillips, 1980 Radin, 2014 Reed, I. (2013). Power. Sociological Theory, 31(3), 193-21 Sahlberg, P. (2011). Finnish lessons : What can the world learn from educational change in Finland. New York, NY: Teachers College Press.