Can Experience be Trusted? Investigating the Effect of Experience on Decision Biases in Crowdworking Platforms

Abstract

Companies increasingly involve the crowd for collective decision making and, to aggregate the decisions, they commonly average the scores. By ignoring crowdworkers’ different levels of experience and decision biases, this method may not favor the best outcome. Alternatively, decisions can be weighted in favor of the more experienced judges in the crowd. However, previous research is inconclusive as to whether more experienced individuals are any better at avoiding decision biases. To answer this question, we conduct online crowd-based experiments with a range of treatments, comparing the anchoring effect of individuals with different levels of experience. Results indicate that not only does greater experience not protect crowdworkers from the anchoring effect but it increases their confidence in their decision, compared to less experienced individuals, even if they are wrong. Our findings provide valuable insights for both researchers and practitioners interested in improving the effectiveness of crowdworking decision-making.