The rejoicing third in social networks

When two people quarrel, the third rejoices! In this old quote the mentioned third one does not appear in a good light. One wishes that the people in social networks interact and connect open-mindedly and friendly – simply in a social way. However, within network theory the so called “tertius gaudens” plays an important role, particularly due to his “antisocial” behavior. How is it possible that a player, who apparently separates other players by causing a quarrel and on top of everything, is even happy about it can be that important – even desirable – for a healthy network?The rejoicing third can be regarded as the antagonist of the “tertius iungens”, the “third who joins” or the “contact person”. If he has information, which is not available to others, he shares it to deepen and strengthen the relation to his environment. Consequently, close networks with strong relations develop around him.

Thus the rejoicing third kind of lives for the information advantage and tries to utilize it profitably. He is looking for news and gets those through a large and continuously growing circle of acquaintances to whom he only has a superficial relation. To ensure his information advantage he prevents his contacts from getting to know each other in order to exchange information. For the network this has a meaning in two respects: In contrast to the networks which only consist of “contact persons” and therefore only exchange their own ideas without any new input, this network grows and has a continuous access to new information. In other words: the contact to the outside world would be missed. This means again, that the network is highly dependent on the rejoicing third as an information source, because there are no other alternative channels to receive this information. As a result, the rejoicing third is boon and bane at the same time, as his “egoistic” approach causes a continuous development of the network.

A closer look at the whole networks reveals only very rare homogeneities. They rather consist of different loosely interconnected clusters. Within these clusters there are close relations between players. They usually speak the same language, share the same interests and often are geographically interconnected. Now the Tertius  Gaudens comes into play. He bridges the so-called structural gaps and connects the clusters with each other. To find these players COMPASS scans big social networks. In order to receive information or – from a marketing perspective even more important – to spread information, these players who are able overcome structural gaps, need to be activated. For this purpose knowledge about the network is needed, meaning information of every person who interacts on a special topic with others. In a next step the key persons of this network have to be identified.

With more than 300 million users worldwide, strong communication within the network and a continuously growing user bases Twitter is the optimal data source. With the support of a Social Media Analysis Tool, developed by COMPASS, a large dataset can be collected through the Twitter API and stored in a database. Starting point is always a query which consists of one or more terms. The first interim result is a data file with information on users that commented to the searched terms. Apart from the metadata, such as the amount of followers of the users, also the users which interact via retweets or answers are recorded. Therefore, a network consisting of knots (user) and edges (connection between the users) develops.


For getting an idea on how such a network can look like, the image above reflects the network of Twitter-users who are engaged in the term “e-bike”. The color represents the language of the users and the size of the circle displays its centrality. Apart from the centrality other important measured values exist within the network theory. How those are calculated and what they mean will be mentioned in the next blogpost on the subject network analysis.