Social Contagion and Cascade Behaviors on Twitter

It has been found in a variety of face-to-face networks that diffusion of information, behaviors and sentiments extend up to two to four degrees of distance from the original source. This regularity has been popularized as the three degrees of influence phenomenon. Prior works have suggested a number of possible explanations to this pattern. In this paper, we study it in the context of an online network. We find similar results in this online setting to those already found offline. However, our approach suggests that two of the previously proposed explanations (increasing instability of connections at greater distances from the source and simple information decay) should not be central to explain the pattern.