This thesis describes a study of the implementation of Indonesia's national policy on active learning. Active learning was first introduced in Indonesia in the 1970s. Since then four different national curricula have rearticulated the policy, and numerous donor and government funded projects have attempted to support implementation. Notwithstanding all of this effort at reform, the gap between policy and practice remains wide. Most schools and classrooms remain little changed. A wide range of teaching practices is employed by Indonesia's more than three million teachers and across its 260,000 schools. However, with a few exceptions, a casual look in any one of these schools will reveal poor conditions, few books or teaching aids, and traditional chalk-and-talk' teaching methods. The study addresses two questions: (1) How do teachers translate active learning methodology in the classroom? (2) What factors impede the implementation of active learning? A mixed-method, qualitative, case study approach was adopted to answer these questions. The case selected was a group of teachers, schools and school clusters in three districts in the remote eastern Indonesian province of North Maluku. The data collection phase of the study took place in 2007. Data analysis was conducted in the following years. Four main data gathering methods were employed: (1) document analysis, (2) survey questionnaire, (3) semi-structured interview and informal discussions, (4) field visits to schools and class observation. The study is underpinned by a conceptual framework, which developes the theory of House (1979), and subsequently House and McQuillan (1998), who proposed three perspectives for the analysis of education reform. This study argues that each of three perspectives is necessary to an understanding of educational reform: the technical, the political and the cultural. The study found that implementation of the policy of active learning has essentially failed in this case, due to a combination of technical, political and cultural factors and the interaction between these three. Is is argued that the dominance of the technical perspective in previous studies and in the policy process is, in part, a result of policy borrowing, whereby technical innovations are borrowed from the West and, with the support of international donors, are implemented in Indonesia ‚Äö- without the cultural context necessary for successful implementation. This thesis thus further develops the theory by taking House and McQuillan's (1998) three perspectives, expanding the meaning to include higher level political perspectives and broader cultural and technical perspectives, and applying this conceptual framework in a new setting, that of Indonesia, a developing nation. This study concludes with a number of recommendations for policy makers and researchers. The key recommendation is that Indonesian policy on classroom practice should be made in a way that is more cognizant of the technical, political and cultural realities. What is required is a review of current policy and implementation expectations, and an adjustment of the policy to fit with Indonesian realities. These realities, it should be stressed, are not uniform and vary according to the local context. This recommendation is significant, not only for Indonesia, which spends substantial amounts of money every year on reform programs which fail repeatedly due to lack of awareness of the need to address cultural, technical and political factors. It also offers a warning to other developing nations and for the international donors which support them.