RAMESES II reporting standards for realist evaluations

Joanne Greenhalgh,Justin Jagosh,Gill Westhorp,Trish Greenhalgh,Geoff Wong,Ana Manzano

doi:10.1186/s12916-016-0643-1

Joanne Greenhalgh, Justin Jagosh + Show 4 more

Open Access

https://doi.org/10.1186/s12916-016-0643-1

Copy DOI

Abstract

BackgroundRealist evaluation is increasingly used in health services and other fields of research and evaluation. No previous standards exist for reporting realist evaluations. This standard was developed as part of the RAMESES II project. The project’s aim is to produce initial reporting standards for realist evaluations.MethodsWe purposively recruited a maximum variety sample of an international group of experts in realist evaluation to our online Delphi panel. Panel members came from a variety of disciplines, sectors and policy fields. We prepared the briefing materials for our Delphi panel by summarising the most recent literature on realist evaluations to identify how and why rigour had been demonstrated and where gaps in expertise and rigour were evident. We also drew on our collective experience as realist evaluators, in training and supporting realist evaluations, and on the RAMESES email list to help us develop the briefing materials.Through discussion within the project team, we developed a list of issues related to quality that needed to be addressed when carrying out realist evaluations. These were then shared with the panel members and their feedback was sought. Once the panel members had provided their feedback on our briefing materials, we constructed a set of items for potential inclusion in the reporting standards and circulated these online to panel members. Panel members were asked to rank each potential item twice on a 7-point Likert scale, once for relevance and once for validity. They were also encouraged to provide free text comments.ResultsWe recruited 35 panel members from 27 organisations across six countries from nine different disciplines. Within three rounds our Delphi panel was able to reach consensus on 20 items that should be included in the reporting standards for realist evaluations. The overall response rates for all items for rounds 1, 2 and 3 were 94 %, 76 % and 80 %, respectively.ConclusionThese reporting standards for realist evaluations have been developed by drawing on a range of sources. We hope that these standards will lead to greater consistency and rigour of reporting and make realist evaluation reports more accessible, usable and helpful to different stakeholders.

Highlights

3 Rationale for evaluation 4 Programme theory 5 Evaluation questions, objectives and focus 6 Ethical approval
Through discussion within the project team, we considered the findings of our review of recent examples of realist evaluations and developed a list of issues related to quality that needed to be addressed when carrying out realist evaluations
Once the panel members had provided their feedback on our briefing materials we constructed a set of items for potential inclusion in the reporting standards and circulated these online to panel members

Summary

Introduction

Explanation Realist evaluations are used for many types of programmes, projects, policies and initiatives (all referred to here as ‘programmes’ or ‘evaluands’ – ‘that which is evaluated’ – for ease of reference) They are conducted for multiple purposes (e.g. to develop or improve design and planning; understand whether and how new programmes work; improve effectiveness overall or for particular populations; improve efficiency; inform decisions about scaling programmes out to other contexts; or to understand what is causing variations in implementation or outcomes). The purpose has significant implications for the focus of the work, the nature of questions, the design, and the choice of methods and analysis [30] These issues should be clearly described in the ‘background’ section of a journal article or the introduction to an evaluation report. Theory-based understandings about ‘what works, for whom, in what contexts, and how’ are, transferable; One of the tasks of evaluation is to learn more about ‘what works for whom’, ‘in which contexts particular programmes do and don’t work’ and ‘what mechanisms are triggered by what programmes in what contexts’

Objectives

Methods

Results

Discussion

Conclusion