Results of a Survey About the Perceived Task Similarities in Micro Task Crowdsourcing Systems

Recommender mechanisms can support the assignment of jobs in crowdsourcing platforms. The use of recommendations can improve the quality and outcome for both worker and requester. Workers expect to get tasks similar to previously finished ones as recommendations, as a preceding study shows. Such similarities between tasks have to be identified and analyzed in order to create task recommendation systems that fulfil the workers’ requirements. How workers characterize task similarity has been left open in the previous study. Therefore, this work provides an empirical study on how workers perceive the similarities between tasks. Different similarity aspects (e.g., the complexity, required action or the requester of the task) are evaluated towards their usefulness and the results are discussed. Worker characteristics, such as age, experience and country of origin are taken into account to determine how different worker groups judge similarity aspects of tasks.