aes24 International Evaluation Conference

Authors: Andrew McLachlan (NSW Department of Education)

At first glance, realist methodology seems ideally suited to investigating implementation problems (Dalkin et al., 2021). It is versatile in that it draws on theories from diverse fields of social inquiry. It is pragmatic in that the theories it adopts are deemed useful only in so far as they offer explanatory insight. And it is transferable; realist methodology is less concerned with generalising findings than in understanding how programs work under different conditions and circumstances.

As for implementation science, its founding aim is purpose built for realist work; it seeks to improve the uptake of evidence-based practices by investigating the barriers and facilitators to implementation. Yet despite the affinity between realist methodology and implementation science, so far there have been few attempts to formalise the relationship (Sarkies et al., 2022).

This paper offers insights into how evaluators can harness realist methodology to better understand challenges of program implementation. It demonstrates how implementation concepts like fidelity (the degree to which a program is delivered as intended), adaptation (the process of modifying a program to achieve better fit), and translation (the ability to transfer knowledge across organisational borders) can be combined with realist concepts to develop a more active understanding of context.

In showing how to construct program theories that are responsive to changing conditions, the paper promises to equip evaluators with tools that can help them navigate the complexities of program implementation in their own work.