Socioeconomic Differences and Spatial Recording Behaviour among Contributory Citizen Science Participants

Citizen science (CS) has developed as a form of public participation in conservation decision-making. Contributory CS that reports species occurrence records to online platforms has broadened the opportunities for a diverse audience freely to record and share species observations. However, given the common lack of spatial or temporal guidelines, this open and unstructured participation amplifies observer-driven biases in where and how biodiversity is recorded, shaped by who participates, their socioeconomic backgrounds, residential locations, and how far they travel to record. Understanding these socio-spatial dynamics of CS participants is critical for uncovering, and potentially correcting, the resultant implications for the design of representative and inclusive conservation actions.

While prior studies examined socio-demographic characteristics of CS participants, there remains a limited understanding of how these factors influence their spatial recording behaviour and data contribution patterns. This study addresses this gap by conducting an online cross-sectional survey of iNaturalistUK registered users, generating 2,500 responses. We collected socio-demographic and residence location data of the users, which we then linked to their observation records from 2008 to 2024. For each user, we first computed the maximum distance, mean distance and the overall spread of recordings from their residence location, then applied DBSCAN and k-means clustering to identify distinct recorder clusters based on the three spatial metrics. The resulting clusters were statistically modelled with socio-demographic variables to determine predictors of cluster membership, followed by analysis of recording volume and species diversity across clusters.

Our results revealed five distinct recorder clusters ranging from local recorders, who are likely to record only near their home or neighbourhood, to wide-range travellers, who are likely to explore large geographic areas and or travel long distances from their residence for recording. Socio-demographic variables such as age group, income, gender, marital status, dog ownership and physical or mental health condition significantly predicted cluster membership. Notably, wide-range travellers and regional roamers who record across regional landscapes tend to contribute more records and a greater variety of species. These findings underscore the critical role of human dimensions in shaping data biases in citizen science data.