The above situation makes the classroom learning environment very likely to be dominated by a teacher centred approach thus minimising the likelihood that students’ interest in learning science will be aroused (Hossain, 1994). In most cases, teachers are not cognisant of the need to take into account students’ prior knowledge in terms of influencing their practice. It is also very rare to find teachers sharing ideas with each other to improve their practice. This situation triggered me to decide to ask science teachers to use a constructivist teaching approach, prediction-observation-explanation (POE), in their practice. My understanding about the use of POE was that it guides teachers in their thinking about subject knowledge and pedagogy and creates new ways of sharing and discussing with their colleagues. Moreover, from the ideas from Gunstone (1995) regarding the importance of genuine collaboration (teacher with teacher, teacher with researchers) in the development and use of constructivist teaching approaches, I found myself becoming more focused and guided in ways that supported my thinking and my plans for action. Therefore, in order to support their learning about this strategy I also designed teacher collaborations within and across nearby schools (described in detail in method section).