Sequential collaboration is a process to elicit dependent individual judgments inspired by large-scale online collaborative projects such as Wikipedia and OpenStreetMap. In this process, entries are created by initial contributors and are adjusted or maintained by subsequent contributors forming a sequential chain of adjusted or maintained entries. Sequential collaboration was adapted to numerical judgment formation showing that judgments are adjusted less over a sequential chain and become increasingly more accurate comparable to aggregating independent judgments (wisdom of crowds). The accuracy of the previous judgment as well as contributors’ expertise have been found to be key determinants of whether a previous judgment is adjusted and how accurate this adjustment is. As contributors in sequential collaboration are only informed about the latest judgment in the sequential chain but do not receive any additional information, we examined how information about the previous contributor may affect adjustments in sequential collaboration. In five studies examining previous contributors’ expertise, gender and group membership, we did not find any substantial influences over and above contributors’ own expertise and the accuracy of the previous judgment. This demonstrates that sequential collaboration seems to be a process that minimizes social biases in group judgments. These results shed light into why online collaborative projects yield highly accurate information by using a bias preventing collaboration method.