This paper explores the interplay between positive and negative relationships in social media networks, a dimension that has been largely overlooked in previous research. The authors analyze signed networks from three widely-used websites: Epinions, Slashdot, and Wikipedia. They compare two theories of signed networks—structural balance and status theory—to understand how these networks are structured and how links are used. Structural balance theory, rooted in social psychology, predicts that balanced triangles (three mutual friends) are more common than unbalanced triangles (two friends with a common enemy). Status theory, on the other hand, posits that positive links indicate higher status and negative links indicate lower status, which can be propagated through multi-step paths. The authors find that while structural balance theory aligns well with undirected networks, status theory better explains the directed nature of online networks. Status theory captures more of the observed patterns, particularly in the context of directed networks. They also find that positive links are more likely to occur in well-embedded groups of nodes, while negative links act as bridges between these groups. Additionally, they observe that reciprocation of positive links follows balance theory, while reciprocation of negative links shows a hybrid behavior, influenced by both balance and status theories. The study provides insights into how people use linking mechanisms in social computing applications and highlights the importance of considering both undirected and directed interpretations of network data. The findings have implications for understanding individual variation in linking behavior and the role of status in shaping network structures.This paper explores the interplay between positive and negative relationships in social media networks, a dimension that has been largely overlooked in previous research. The authors analyze signed networks from three widely-used websites: Epinions, Slashdot, and Wikipedia. They compare two theories of signed networks—structural balance and status theory—to understand how these networks are structured and how links are used. Structural balance theory, rooted in social psychology, predicts that balanced triangles (three mutual friends) are more common than unbalanced triangles (two friends with a common enemy). Status theory, on the other hand, posits that positive links indicate higher status and negative links indicate lower status, which can be propagated through multi-step paths. The authors find that while structural balance theory aligns well with undirected networks, status theory better explains the directed nature of online networks. Status theory captures more of the observed patterns, particularly in the context of directed networks. They also find that positive links are more likely to occur in well-embedded groups of nodes, while negative links act as bridges between these groups. Additionally, they observe that reciprocation of positive links follows balance theory, while reciprocation of negative links shows a hybrid behavior, influenced by both balance and status theories. The study provides insights into how people use linking mechanisms in social computing applications and highlights the importance of considering both undirected and directed interpretations of network data. The findings have implications for understanding individual variation in linking behavior and the role of status in shaping network structures.