Storm Crowds: Evidence from Zooniverse on Crowd Contribution Design
53 Pages Posted: 20 Sep 2017 Last revised: 14 Apr 2019
Date Written: April 13, 2019
The proliferation of platforms with distributed content production, including open source software development, crowdsourcing, and citizen science, has led to scholarly interest in understanding this phenomenon. A parallel and growing stream of research explores the impact of modularity whereby contributions are moulded by whether they can be performed independently and then aggregate to a broader and deeper contribution. This paper seeks to contribute to this literature by empirically disentangling the impact of modularity on distributed contributions. To the best of our knowledge, it is the first paper to provide evidence of the causality of this relationship. Our setting is Zooniverse, the largest citizen science platform in the world, where volunteers (editors) contribute to scientific research by performing data analysis tasks. Like many other citizen science projects, most Zooniverse projects involve tasks that have low complexity, low interdependency, and a well-defined structure. Our choice of setting is motivated by a format change that occurred in the Zooniverse project Cyclone Center, which allows us to observe editor contribution levels in formats with varying levels of modularity. Our difference-in-differences analysis shows that when modularity decreases, editors respond by contributing fewer total edits weekly, and more extended edits than would be predicted in the absence of a change. We believe our findings are relevant to a growing number of citizen science projects, and to other distributed content production initiatives with simple, well-defined, and independent tasks.
Keywords: crowdsourcing, contribution design, contribution divisibility, zooniverse
Suggested Citation: Suggested Citation