On October 13, 2023, over 50 people gathered in Berkeley, CA for the Forecasting in the Social Sciences Workshop, which brought together leaders from across academic disciplines to present new research findings, share knowledge, and continue charting a path forward for prediction in the social sciences.
The event featured Anna Dreber (Stockholm School of Economics) as the key note speaker, who gave a talk on predicting replication outcomes and the questions we need to consider: Which results can we trust? What happens when you give the same data to many researchers? How do their analyses differ, and are they predictable?
Throughout the rest of the day, we heard from a fantastic lineup of speakers: Don Moore (UC Berkeley), Nicholas Otis (UC Berkeley), Linnea Gandhi (University of Pennsylvania), Dean Yang (University of Michigan), SSPP Principal Investigator Eva Vivalt (University of Toronto), Edward Miguel (UC Berkeley), David Bernard (Paris School of Economics), and Anya Samek (UC San Diego). Their research spanned topics like forecasting in Behavioral Experiments, Development Economics, Organizations, and forecasting long-run outcomes.
We closed the day with a panel moderated by SSPP Principal Investigator Stefano DellaVigna (UC Berkeley) and featuring Don Green (Columbia University), Yan Chen (University of Michigan), and Seth Blumberg (Google), which discussed the challenges and directions forward for forecasting.
You can find the presentations slides here.
Reminder: The SSPP has launched our new Superforecaster Panel, which is comprised of 90+ academic researchers who will regularly take surveys posted on the platform. There is no limit to the number of surveys you can post and no criteria about how long you must wait between each survey.
We highly encourage you to take this opportunity to collect predictions for your project and consider publishing a survey on the SSPP! If you need any help getting started, our Social Science Prediction Platform Survey Guide contains an abundance of helpful information from what can be forecast to how to design your Qualtrics survey. As always, our team is eager and available to help out during this process if you need.