The Wisdom of Virtual Crowds


Project status granted
Project time 01.05.2024- 30.04.2027
Website https://gepris.dfg.de/gepris/projekt/529885974
Funding source DFG - Individual Grants Programme
Project number HE 3437/12-1; LA 1316/5-1; GO 2635/2-1
Keywords Wisdom of Crowds; Virtual Reality; Metaverse; Decision Making Behavior

Most managerial decisions are decisions under uncertainty. Hence, corporations need to anticipate future developments and generate estimates for key input variables entering their planning process. The decision quality and hence the returns for firms and their stakeholders crucially depend on the accuracy of the employed estimates. To improve estimation accuracy, corporations often let a group of individuals deliberate about the estimation problem and thereby exploit the ‘wisdom of the crowd’. For instance, averaging over multiple judgments typically increases accuracy because individual estimation errors can cancel out. Alternatively, behavioral aggregation based on group discussions does not only exploit the diversity of individual beliefs but also seeks to obtain more accurate estimates by facilitating social interaction between group members (exchange arguments, obtain feedback, question others’ point of view). In response to the recent pandemic, however, policy makers across the globe imposed strict rules on how social interactions could take place. Face-to-face interactions were mostly prohibited, forcing almost all kinds of meetings to be conducted virtually (e.g., via video conferences or in the virtual reality). These interaction types are highly distinct. In video conferences, group members are confronted with real-life faces on a computer screen, but they lack an immerse environment in which the social interaction takes place. Virtual reality provides users with an immerse environment, but people typically interact using virtual characters (avatars), limiting the range of facial expressions, gestures, and body language. It is - from a scientific standpoint - unclear how such a regime shift from conducting offline deliberations to working in virtual teams affects the accuracy of group estimates, i.e., the wisdom of crowds. This research project aims at filling this gap by experimentally exploring the estimation performance of groups who either interact face-to-face, via video conference, or in the virtual reality. In addition, it seeks to identify the channels that explain potential differences in estimation accuracy across the different types of social interaction. Understanding the potential of virtual crowds with respect to their estimation performance is not only vital for firms to best respond to pandemic-related restrictions on social interactions. Making greater use of virtual crowds could also contribute towards the fight against climate change as it reduces the need to business-related travelling and hence CO2-emissions.