Research shows that teachers as middle leaders are important for school improvement. However, we need to know more about the process of teachers becoming key persons for school-wide change.

In Sweden there is a shift going on from schools being led by principals to schools with distributed leadership between school leaders and teachers with professional development leader roles. These teachers are a new kind of middle leaders. In two evaluations (Olin et al, 2014; Nehez, et al, 2017) we have followed 60 such teachers, called process leaders, appointed by principals, in 13 comprehensive schools and 13 preschools in a Swedish municipality. The process leaders’ role was to lead and stimulate their colleagues in developmental work. To support the process leaders a nine half-day course was set up by two development leaders from the local school administration. The goals was to develop understanding of school improvement and strengthen the capacity to lead improvement processes. In this study a new analysis is done about how ideas was translated from course to practice and the role played by such translations in organisational change.

Schools/preschools that have kept the process leaders have developed from diffuse and confused organisations for both routine work and development to what we call “task and process governed organisations”. Teachers are assigned tasks within specific areas, in which they are resources for colleagues. Process leaders, more often than others, have coordinating tasks. They have widened their roles from, among others, administrative principal assistants to development leaders of their organisations. With the appointment of process leaders the principals have developed the distributed character of leadership including, and supporting, their own leadership.

Certain generative mechanisms (Pawson & Tilley, 1997) had important impact on the change of the organisations. Mechanisms are staffs’ positive reactions to being visible as professionals for each other, co-operation across borders, improved deliberative structure of meetings, better coupling between leadership and staff, getting staff ownership from many tasks and wide participation in activities, more systematic developmental processes, improvements visible for children and students, intake of new from other sites and research. In many of these processes process leaders used artifacts and templates learned during the course.

This paper examines the development of task and process governed organisations, the development of the process leaders to organisational developers, and the generative mechanisms the process leaders were part of, in relation to the content of the course. More exactly, we explore:

• What kind of ideas about leading development processes were introduced in the course for process leaders?
• How did the process leaders’ translate the ideas into practice?
• How do ideas and translations enable the change of the organisations?

The analysis is based on translation theory (Røvik, 2015), a Scandinavian new organisational theory. Translation theory focus how ideas travel between different contexts and are translated into local practices in the recipient contexts. Røvik claimes that there are certain rules that inform the translation processes and that rules are contextually dependent. He also emphasize that master ideas gets virus-like impact (Røvik, 2011). The leaders of the course did a translation already when constructing the course. The process leaders then did a further translation. Both these levels of translations are in focus in this study.

Insights to the research questions enhance understanding of how translation of ideas, into local practices in local sites, may be guided, interpreted or understood to change diffuse organisations into task and process driven organisations with the help of distributed leadership and teachers as middle leaders. This is of importance in contexts with governmental career reforms for excellent teachers, in Sweden so called first teachers, without relevant education and training.

Mixed method have been used in the evalutations: (1) a survey to all teachers, including the process leaders, at schools/preschools where process leaders have been implemented and (2) interviews with principals, process leaders and teachers at those schools/preschools. The data from these are re-analyzed with focus on the process leaders work tasks and translations of ideas from the education. In the re-analyze reported in this paper, we complement the methods used in the evaluations with document analyses on the syllabus as well as readings and assignments in the education. The translations made by both the course leaders and the process leaders are analyzed and related to the results from the evaluations to be able to answer the questions.

The preliminary results indicate that the development of task and process governed organisations can be derived from the education and training for process leaders. The tools and models for school improvement that were introduced and trained were chosen to develop the appointed teachers to process leaders in the role of change agents. For example, one of the assignments during the training was to analyze the own schools’ improvement history together with the principal with the help of an observation template as an analytical tool. By "copying" the observation template, the process leaders became aware of an aspect that made their assignments difficult, namely the amount of processes and ambiguities about who was responsible for what. They went from identifying many roles to focusing on a few development areas that were important to all teachers in order to support their students. Prioritizing organisational and professional learning processes lead to more formalized assignments for the process leaders. Moreover, the translation of ideas for structuring and clarifying developmental assignments and processes enabled other ideas to be translated in practices. One conclusion is the importance of clear aims and a relevant education and training for teachers in their middle leader role. A question that arouse from the study is how the training and the translations made based on it affected the principals and their willingness to have process leaders. In the evaluations made in 2014 and 2017 we identified that some of the principals who chose to abandon process leaders regarded the process leaders as threats to their own roles, describing the need to be in control of all the processes going on in their organisations. The principals in the task and process governed organisations, on the other hand, highlighted the possibility to better improve school development with the help of process leaders.

Olin, A., Lander, R., Blossing, U., Nehez, J. & Gyllander, L. (2014). Processledare för skolutveckling. Uppföljning av införandet av processledare i ett verksamhetsområde i Helsingborg. RIPS: Rapporter från Institutionen för pedagogik och specialpedagogik, Nr 5. Göteborg: Institutionen för pedagogik och specialpedagogik, Göteborgs universitet. Nehez, J., Gyllander Torkildsen, L., Lander, R., Olin, A & Blossing, U. (2017). På väg mot uppdrags- och processdrivna organisationer. Uppföljning av införandet av processledare i förskolor och skolor i Helsingborg. RIPS: Rapporter från Institutionen för pedagogik och special-pedagogik, Nr 5. Göteborg: Institutionen för pedagogik och specialpedagogik, Göteborgs universitet. Pawson, R. & Tilley, N. (1997). Realistic evaluation. London: Sage. Røvik, K.A. (2015). From fashion to virus: An alternative theory of organizations’ handling of management ideas. Organisation Studies, 32(5) 631-653. Røvik, K.A. (2015). Reformidéer og deres tornefulle vei inn i skolefeltet. In K.A. Røvik, T.V. Eilertsen & E. Moksnes Furu, Reformidéer i norsk skole. Spredning, oversettelse og implementering (p. 13-50). Oslo: Cappelen Damm Akademisk.