Democratic forecast: Small groups predict the future better than individuals and crowds.

Area of Science:

• Social Psychology
• Collective Intelligence
• Forecasting Science

Background:

• Expert predictions are often inaccurate and costly.
• Existing collective intelligence methods like prediction markets are selective and expensive.
• Social deliberation is known to enhance general knowledge problem-solving.

Purpose of the Study:

• To investigate if social deliberation can enhance prediction accuracy in a cost-effective and inclusive manner.
• To determine if small, organized groups can outperform individual predictions and traditional "wisdom of crowds" methods.
• To challenge the notion that discussion hinders collective intelligence in forecasting.

Main Methods:

• Small groups of lay individuals engaged in organized social deliberation.
• Compared predictions from deliberating groups to individual predictions and large non-deliberating groups.
• Utilized consensus-based aggregation of predictions.

Main Results:

• Deliberation and consensus significantly improved individual prediction accuracy.
• Aggregating just two consensual predictions outperformed the "wisdom of crowds" aggregation of 100 individual predictions.
• Organized small groups produced better predictions than individuals forecasting alone.

Conclusions:

• Social deliberation and consensus-seeking enhance the accuracy of predictions about unknown facts.
• Small, organized groups offer an efficient, inclusive, and inexpensive solution for group forecasting.
• Collective intelligence can be effectively leveraged through discussion and consensus for improved forecasting.