Mobile Social Network (MSN) is a social network where one or more elements share interests or commonalities by connecting with one another using mobile devices. However, contrary to traditional online social networks, efficient communication and information exchanged depends on mobile users' context. In order to increase the scalability of the information exchange, network nodes need to be associated and grouped through an efficient set of rules and criterions, to minimize the time and amount of data messages flowing in the network. This article shows how social-based metrics can increase the efficiency of layer 2 associations in mobile networks, where nodes are able to join the network through their context information and probable interactions, optimizing the information exchange at MAC layer. Through a case study of mobile communications with different mobility patterns and dissemination approaches, we show that the social-based metrics increase overall network efficiency: maximize the percentage of information delivery, decrease the time to disseminate data traffic and reduce network usage.