Citizen science survey results published

Earlier this year, my first journal publication as faculty at UMD, “Surveying the citizen science landscape”, reported on results of a survey of citizen science projects that I conducted as a PhD student. The most important take-away is that citizen science is incredibly diverse. Just like in the early days of studying open source software, research and media often call attention to a few outliers that don’t really represent the full richness of the broader community of practice. In addition to a few such large-scale projects, our survey results primarily describe small-to-medium sized citizen science projects, mostly in North America, and largely focused on collecting ecological data. There are a few common strategies, characteristics, and feature sets that describe most of these projects for any given variable of interest, such as funding sources, data quality strategies, and the kinds of task-oriented and social activities available to volunteers. But these variables were all uniquely combined for each project, as the design of the entire enterprise has to work within the constraints imposed by resources, scientific standards, and project goals. There was no obvious cookie-cutter pattern of “right answers” to address common questions in project design. Like any other research, doing science with the involvement of volunteers requires designing within specific limitations in order to achieve specific outcomes, and the constraints each project faces are unique. There’s no magic formula to identify which combination of characteristics will create the right conditions for a particular project, of course. However, by describing the range of project characteristics and strategies for key operational considerations, we hope to help practitioners make important decisions about citizen science...

Citizen Science Data Usage

    To understand the potential value and applications of freely available, carefully curated open citizen science data, our research team is working with partners from the University of Michigan and the Cornell Lab of Ornithology on a small study of eBird data users’ practices and outcomes. Survey responses are currently under analysis; respondents documented a broad range of uses for eBird data across distinct contexts, including numerous conservation applications, academic research studies, educational uses at every level, and leisure-related uses, such as record-keeping in the birding community. The inclusion of effort information—documentation of the time and place that observations were made—was considered critical for most uses of the...

Human Computer Learning Network

To improve scalability and data quality, the eBird Human-Computer Learning Network (HCLN) blends emerging techniques that integrate the speed and scalability of mechanical computation, using advances in Artificial Intelligence, with the real intelligence of human computation to solve computational problems that are beyond the scope of existing algorithms. In addition to developing emergent filters and new models of contributor expertise, this work makes extensive use of the semantic links between observations and observers to mine additional information from the existing data in order to strategically address data quality issues. Our extended research team have found evidence of learning-through-doing and improvements in participant performance with accumulated project experience. Initial analyses also confirmed expectations about localization of observers’ knowledge of species, and showed that measures of performance effectively distinguish between observers who are and are not able to detect and identify “secretive” or ambiguous species. This project is a collaboration with the Cornell Lab of Ornithology, eBird, University of Michigan, and Oregon State...