Authors: Zina Baghi (NSW Department Of Education), Annette Waters (NSW Department Of Education) What happens when evaluation is asked to make sense of a program that simultaneously spans curriculum reform, evidence-based resource design, data analytics and school capability-building - across thousands of schools, amid widespread disruption? This paper presents the findings of a process evaluation of a large-scale government education program and uses that experience to interrogate what evaluation distinctively contributes when it operates at the boundaries of multiple disciplines, sectors and organisational roles. Operating within a complex policy space - where multiple related initiatives ran concurrently, each targeting overlapping aspects of school improvement - the program layered differentiated support from self-directed access to quality-assured, evidence-based resources, through to shoulder-to-shoulder guidance from educational leaders. Evaluating this required the team to engage fluently with education research, data analytics, evidence-based pedagogy, professional learning design and school improvement methodology - not as a methodological luxury, but as a necessity for understanding what was working, for whom, and why. The evaluation also operated across organisational boundaries. Two evaluation teams, one embedded within the program's delivery unit, the other in a central evaluation function, worked concurrently on different components, with findings integrated into a shared report. This arrangement surfaces rarely examined questions about co-production, methodological consistency and what happens to evaluation's integrity when the boundary between evaluator and implementer is not just navigated but structurally blurred. The findings reveal where bridges within the program held and where they fractured: between system policy intent and school-level practice, between co-designed improvement partnerships and variable local capacity, and between a program designed predominantly for primary schools and a secondary sector left largely overlooked. For the evaluation community, this paper argues that understanding these fractures - and building the cross-disciplinary bridges needed to address them - is precisely where evaluation's value is most needed, and most often undersold.
