Abstract

We started our journey six years ago with a review of literature on, and models of, successful school leadership for improved student outcomes. When the findings of this review were combined with the results from five case studies of successful school principalship, it resulted in a preliminary model of successful school principalship. This model hypothesises that successful school principalship is an interactive, reciprocal and evolving process involving many players, which is influenced by and, in turn, influences, the context in which it occurs. We argued that more needed to be done to test the preliminary model. Taking our own advice we examined a range of areas using further analysis of the qualitative case study data, detailed analysis of the subsequent quantitative surveys of principals and teachers and actual school literacy and numeracy results, including those results taking school socio-economic status into account. These areas, which have been reported in a number of published works, include the ‘what’ and ‘who’ of successful school principalship, leadership tensions and dilemmas, instructional leadership, evaluation and accountability, decision making, schools in high poverty communities, small schools, and principals in late career.

Our results suggested that some variables have a much stronger relationships with student outcomes than other variables. These variables were grouped around areas such as school capacity building, evaluation and accountability and socio-economic status. Other variables, such as principal characteristics, were only found to be weakly and indirectly related to student outcomes.

Learning from our results, the need to use broader outcomes of schooling to measure success and the need to gather data from multiple sources, especially from principals and teachers, we moved to the final part of our research. This presentation focuses on our results and the revised models and a re-conceptualisation of successful school principalship for improved student outcomes based on model building and powerful multi-level statistical analyses.