This study represents a systematic replication of a previous intervention which took place in schools in Mangere from 2003 to 2005. (McNaughton, MacDonald, Maituanai-Toloa, Lai, & Farry, 2006). The contexts and theoretical rationale are the same for the present study as those for the original Mangere study. The authors have repeated that historical and social context and theoretical framework in this report. In the first section, however, they briefly summarise the original study and also outline the form and the role of replication in the science represented in this report. This study adds to a growing body of research (e.g., Taylor et al., 2005; Timperley et al., 2003) which suggest the importance of promoting the critical analysis of evidence in schools. This study, however, reveals that there are some conditions to consider when encouraging such analysis in schools. Firstly the findings suggest, as others have found, that such analysis is likely to be dependent on external support in the form of collaborative research-practice-policy partnerships (e.g., Annan & Robinson, 2005; Lai et al., 2004). Secondly, the findings also caution against focusing on critically analysing evidence as a substitute for professional development that focuses on fine-tuning teachers' pedagogical and content knowledge. Thirdly, the complexities in the data around the first three phases of professional development highlight the need for more research to better understand the locus of change in student outcomes and the impact of schools and teachers on those changes. (Contains 29 tables, 29 figures and 4 footnotes.) [Funding for this research was provided by the Woolf Fisher Trust.]