This study focuses on real-world events and their reflections on the continuous stream of online discussions. Studying event diffusion on social media is important, as this will tell us how a significant event (such as a natural disaster) spreads and evolves interacting with other events, and who has helped spreading the event. Tracking an ever-changing list of often unanticipated events is difficult, and most prior work has focused on specific event derivatives such as quotes or user-generated tags. In this paper, we propose a method for identifying real-world events on social media, and present observations about event diffusion patterns across diverse media types such as news, blogs, and social networking sites. We first construct an event registry based on the Wikipedia portal of global news events, and we represent each real-world event with entities that embody the 5W1H (e.g., organization, person name, place) used in news coverage. We then label each web document with the list of identified events based on entity similarity between them. We analyze the ICWSM’11 Spinn3r dataset containing over 60 million English documents. We observe surprising connections among the 161 events it covers, and that over half (55%) of users only link to a small fraction of prolific users (4%), a notable departure from the balanced traditional bow-tie model of web content.