_MG_7289
Crowdworkers at work.

project1-icon 

Many of today’s crowdsourcing marketplaces are characterized by cheap labor and simple tasks that are easy for people and hard for computers such as transcribing text and labeling images. Crowdsourcing complex creative tasks remains difficult because crowdworkers lack the motivation or skills to do the work. We explore how training, gamification, and priming improve the quality of complex creative work performed on crowdwork platforms. Training allows crowdworkers to gain useful skills through the execution of the task. Priming uses implicit mechanisms to induce observable changes in behavior. Athough priming has been used in the laboratory, its use hasn’t been explored extensively in software design.

This project explores how crowdsourcing platforms can motivate and provide training to crowdworkers in ways that benefit both the workers and the requesters of the work.  

People

Elizabeth Gerber, Northwestern U.
Mira Dontcheva, Adobe
Joel Brant, Adobe
Rob Morris, MIT Media Lab
Adam Finkelstein, Princeton U.

Papers

  • Dontcheva, M., Morris, R., Brandt, J., & Gerber, E.  Combining Crowdsourcing and Learning to Improve Engagement and Performance, Computer Human Interaction, 2014, ACM Press
  • Kittur, N., Bernstein, M., Nickerson, J., Gerber, E., Shaw, A., Lease, M., Zimmerman, J., Horton, J. The Future of Crowd Work, Computer Supported Cooperative Work 2013, ACM Press
  • Morris, R. Dontcheva, M., Finkelstein, A. Gerber, E. Affect and Creative Performance on Crowdsourcing Platforms,” Affective Computing and Intelligent Interaction Conference, 2013
  • Kittur, N., Bernstein, M., Nickerson, J., Gerber, E., Shaw, A., Lease, M., Zimmerman, J., Horton, J. The Future of Crowd Work, Computer Supported Cooperative Work 2013, ACM Press
  • Lewis, S., Dontcheva, M., & Gerber, E. Affective Computational Priming and Creativity, Proceedings of Computer Human Interaction 2011, ACM Press