Democratic innovations and publication bias – what literature on public participation in environmental governance tells us

Demands for open science, i.e. for transparency at all stages of the research process and unbiased dissemination of research findings, are becoming ever more widespread, including the dynamic and diverse field of democratic innovations. While these calls are well justified and warranted, the actual direction and degree of biases in democratic innovations and public participation research are difficult to trace and, thus, remain rather diffuse. Hence, this paper sets out to map the field and shed some light on those aspects and topics dominating the debate. Ultimately, the aim is to trace directions of publication bias and its symptoms in public participation research. To approach these questions, I rely on a data set of 305 cases of public participation in environmental decision-making, originally elicited through a comprehensive case survey meta-analysis. This method traces published, qualitative case study accounts and transfers them into quantitative data by means of a standardized coding procedure relying on a comprehensive coding scheme. While its original purpose was to study the impact of public participation on environmental governance outcomes, the data set offers a unique opportunity to gather insights on the academic field of democratic innovations and public participation (in the environmental realm) and to explore publication bias within this realm. In a first step, I describe the field and map patterns of high or low attention paid to various aspects of public participation, including contextual, procedural and outcome aspects. Previous analyses focusing on the outcome dimension of participation processes, for example, (see Koontz et al. 2020) have shown that studies in this field focus foremost on social and procedural outcomes (e.g. conflict resolution, learning) and pay less attention to environmental or further institutional implications. In a second step, patterns of publication bias are addressed more specifically by comparing attention to specific aspects and themes in the literature across different publication types (e.g. peer-reviewed journal papers vs. grey literature). This analysis again includes contextual, procedural and outcome aspects, but also more complex patterns and research themes involving multiples aspects, for example links between power-sharing and outcome acceptance or between deliberation and trust building. For these steps, methods of descriptive and inferential statistics are used. Results show that publication bias can indeed be traced – for instance, failures of public participation are underrepresented vis-à-vis success stories – but its actual degree depends on the factors and questions at hand. The paper concludes with a discussion of potential sources of bias and implications of these findings for the field.