Minimizing social bias with sequential collaboration: The role of contributor features in dependent judgments

Collaborative online projects rely on sequential collaboration; a process in which entries are adjusted consecutively by contributors. Sequential collaboration has recently been examined for numerical judgment aggregation for which it produces highly accurate information. However, the impact of additional information about previous contributors on subsequent judgments remains unclear. Thus, Experiments 1A (N=85), 1B (N=186), and 1C (N=302) examined the effect of presented expertise, gender, and group membership of the previous participant, respectively, on participants' behavior in sequential collaboration. We did not find a significant effect of any of the presented features on change probability or change magnitude. In Experiments 2 (N=538) and 3 (N=878), we focused on previous participants' expertise as a highly relevant presented feature. We did not find an effect of previous participants' expertise in Experiment 2. Experiment 3 revealed significant but small effects below the smallest effect size of interest. Overall, these findings suggest that sequential collaboration shows some robustness against simple social influences highlighting its potential as a method for judgment aggregation. This study adds to the understanding of how collaborative processes can be optimized for accuracy and reliability. For collaborative projects these results further emphasize that sequential collaboration is a successful and bias-preventing way of collaboration.