Fact check: What methods are used to estimate crowd sizes at protests and demonstrations?

1. Summary of the results

There are several established scientific methods for estimating crowd sizes at protests and demonstrations. The primary traditional approach is the Jacobs Method, which calculates crowd density by multiplying the total area by the number of people per square meter ^[1]. Modern technology has introduced more sophisticated approaches, including:

• Aerial photography analysis
• CCTV footage monitoring
• Digital counting systems with thermal cameras
• Public transport data analysis
• Mobile network data tracking
• AI-powered human shape recognition ^[1]

A particularly notable advancement comes from the University of São Paulo, where researchers developed a drone-and-AI combination method that significantly reduced the margin of error from 30% to 12% ^[2].

2. Missing context/alternative viewpoints

The accuracy of crowd size estimation is particularly crucial in political contexts, where numbers can be used to demonstrate political support and influence public opinion ^[2]. Traditional estimates often suffer from political bias, which has driven the need for more objective, technology-based solutions.

The challenge of accurate crowd counting is further complicated by the spread of disinformation and misleading information about protests on social media ^[3]. Real-world examples from Los Angeles demonstrations show how protest numbers are reported, though the methodology behind these counts isn't always transparent ^[4].

3. Potential misinformation/bias in the original statement

Several groups have vested interests in crowd size estimates:

• Political organizations benefit from inflated numbers that support their cause
• Law enforcement might prefer lower estimates for certain types of protests
• Technology companies developing counting solutions benefit from highlighting the unreliability of traditional methods
• Media organizations might use different counting methods depending on their political alignment

The development of more accurate counting methods, like the São Paulo drone system ^[2], represents an attempt to overcome these inherent biases and provide more objective measurements. However, it's important to note that even these improved methods still have a margin of error, albeit reduced.