For decades, research on participation in environmental governance has yielded ambiguous results on whether and how participation should benefit the environment. Many have expected participation to improve the quality of environmental decisions (e.g. through incorporation of local knowledge) and their implementation (e.g. through generating increased acceptance) (Beierle and Cayford 2002; Reed 2008). However, findings on the effectiveness of participatory environmental governance are highly contested, and there is only limited knowledge on ‘what works’ (Chess and Purcell 1999; Koontz and Thomas 2006; Newig and Fritsch 2009). This lack of consolidated knowledge stands in contrast to the huge number of case studies on participation in environmental governance that have been conducted over the course of several decades. These studies constitute an excellent and so far largely untapped resource for research. No systematic evaluation of this body of research is available to date; with the notable exception of the meta study by Beierle and Cayford (2002), which is restricted to the United States, and only partly covers environmental effectiveness of participation.

We report on results from a meta-analysis (case survey) of 250 published cases of public environmental governance from 22 different countries, conducted as part of the ERC Starting Grant project EDGE (‘Evaluating the Delivery of Participatory Environmental Governance using an Evidence-based Approach’). To our knowledge, this is the largest and most rigorous analysis of its kind in environmental policy and governance.

This survey of documented cases of (more or less) participatory environmental decision-making aims to systematically compare characteristics of the process, the context, and the outcomes of decision-making. The case survey method is used to transform qualitative data from rich case descriptions into (semi-)quantitative data, allowing for statistical analysis.s. To this end, a comprehensive coding scheme was developed, comprising 315 unique variables, each defined on the basis of theoretical insights from environmental governance and related disciplines (Newig et al. 2013). Each case was assessed independently by three coders based on their individual reading of the material, followed by a joint discussion to compare and discuss diverging codes. Overall inter-coder reliability is high.

Non-parametric correlation analysis indicates that the ‘intensity’ of participation (measured through a multitude of different variables) correlates highly with both environmental provisions of outputs and with social outcomes of decision-making processes. These findings prove particularly strong for the participatory dimensions of communication, deliberation and power delegation. Surprisingly, however, the mere inclusion of citizens and the lay public (as opposed to organised interest groups) correlates only weakly with environmental outputs and outcomes.

Employing advanced statistical techniques, such as ordinations and general linear modelling, we particularly focus on the interactions among independent variables (characterising participation) and dependent variables (environmental and social outcomes) to reach a detailed understanding of the mechanisms at play. Special attention is paid to contextual factors such as the specific (national) policy and societal background as well as conditions of temporality and the environmental resource at hand. With this strategy, we aim to move beyond structuring of the field to produce robust evidence on the conditions under which different forms of participation are likely to ‘work’ in environmental policy and governance.