Workflow Transparency in a Microtask Marketplace

dc.contributor.authorKinnaird, Peter
dc.contributor.authorDabbish, Laura
dc.contributor.authorKiesler, Sara
dc.date.accessioned2023-06-08T11:45:09Z
dc.date.available2023-06-08T11:45:09Z
dc.date.issued2012
dc.description.abstractInterdependent tasks in Mechanical Turk (MTurk) can be managed efficiently with a workflow, a sequence of tasks through which work passes to its completion. We ask if workers should be informed about the workflow, which we call workflow transparency. Transparency could motivate workers or induce social loafing. We describe three experiments to determine the effects of workflow transparency in MTurk. We compared a text description of the workflow, a visualization of the workflow, and the combination of text and visualization with a control condition giving no workflow information. Workflow transparency marginally increased volunteerism on a charity identification task (experiment 1) and significantly increased volunteerism and quality on a business identification task (experiment 2). Results were weaker with a less experienced worker sample (experiment 3). We suggest further research on the design of workflow information to increase workers' motivation.en
dc.identifier.doi10.1145/2389176.2389219
dc.identifier.urihttps://dl.eusset.eu/handle/20.500.12015/4949
dc.language.isoen
dc.publisherAssociation for Computing Machinery
dc.relation.ispartofProceedings of the 2012 ACM International Conference on Supporting Group Work
dc.subjectworkflow
dc.subjectvisualization
dc.subjectcscw
dc.subjectcrowdsourcing
dc.subjecttask motivation
dc.subjectproductivity
dc.subjectentitativity
dc.titleWorkflow Transparency in a Microtask Marketplaceen
gi.citation.publisherPlaceNew York, NY, USA
gi.citation.startPage281–284
gi.conference.locationSanibel Island, Florida, USA

Files

Collections