Analysing volunteer engagement in humanitarian crowdmapping

• Main Author: Dittus, Martin Sebastian
• Other Authors: Capra, L.
• Published: University College London (University of London) 2017
• Subjects: 004
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.746912

Organisers of large crowdsourcing initiatives need to consider how to produce outcomes, but also how to build volunteer capacity. Central concerns include the impact of the first-time contributor experience, and the interplay of different modes of participation in larger organisations that host multiple strands of activity. How can volunteer capacity be built proactively, so that trained volunteers are available when needed? How important are opportunities for social encounter, either online or in person? We present four empirical studies of the Humanitarian OpenStreetMap Team (HOT), a novel setting where thousands of volunteers produce maps to support humanitarian aid. Its diversity of settings and activities provides an opportunity to observe the effects of different coordination practices within a single organisation. Participation is online and open to all, however volunteers need to learn specialist tools and workflows. To support newcomers, HOT organises offline events to learn the practice under expert guidance. Our research is motivated by a dual aim: first, to produce empirical evaluations of novel practices, informed by existing community concerns. Second, to revisit existing theories in social and behavioural science through the lens of this novel setting. We use statistical methods to observe the activity and retention of HOT volunteers. The full HOT contribution history is our primary source of empirical evidence, covering multiple years of activity. We can demonstrate that coordination practices have a marked impact on contributor retention. Complex task designs can be a deterrent, while social contribution settings and peer feedback are associated with a significant increase in newcomer retention. We further find that event-centric campaigns can be significant recruiting and reactivation events, however that this is not guaranteed. Our analytical methods provide a means of interpreting key differences in outcomes. We relate our findings to comparable settings, and close with a discussion of the theoretical and practical implications.