Citizen science (CS) projects typically have citizen scientists with different levels of expertise and agency contributing data or knowledge. Every contribution leaves traces of their involvement, including metadata such as locations or emails. Through four case studies this paper explores the generation, use, and publication practices of CS projects’ metadata. We use a mixed-method approach combining document reviews, interviews, and an online survey, to generate insights into current metadata practices and perceptions of project contributors and organisers. We identify several weaknesses in CS projects’ data collection practices: Participants have only limited awareness of the metadata they contribute, and the privacy implications it can have. Matching expectations between project contributors and organisers regarding acknowledgement is crucial - and metadata play a key role. Projects need data processes and documentation aligned with open science principles, and clear communication to contributors about the data they collect and use. Finally, projects need to consider the mental models of contributors in relation to personal data and associated risks. We derive key considerations that data-intensive CS projects should make in their initial design phase, to generate consistent metadata in line with their participants’ expectations, which in turn increases transparency and thus can increase data reuse.
