Following the results/so what/details format…
With age, the network of male mobile subscribers start to become predominantly composed of females. At age 25, the network of a male subscriber is 56% male; at 60 it drops to 50%. The same cannot be said for females: the percentage of their networks that are male is consistent at around 44%.
4 distinct age groups were identified based on the structure of their social networks : 18-23, 24-27, 28-46, 47-60. The youngest group had cliques; the oldest keep a very densely connected small network. Note that these age brackets seem to correspond to life stages : student, young adults, couple with children, late professional.
So what
Since different age groups behave differently and each culture also behaves differently, each telco performing social network analysis cannot rely on “fixed rules” for analysis. Most vendors that have a black box that works for one country may not work for another. In some cases, it may be necessary to create and analyze multiple social networks for each “segment”. Thus when you select a tool for SNA, ensure that it can accomodate different definitions of social networks and also allow you to define your network appropriately.
Additionally, when using SNA for segmentation and viral marketing purposes, you can now reasonably expect the diffusion and word-of-mouth patterns to be different for different age groups. As such influencers will also be different in different age groups. Make sure your tool is able to take these segment differences into account.
Details
Dataset included a 6 month dataset from a Belgian operator covering 3.3M users and 6BN call/SMSs. This dataset was compared to national distributions and found not to be biased. Network graph was constructed with reciprocal edges resulting in 3M nodes and 7M edges.
Alina Stoica, Zbigniew Smoreda, Christophe Prieur, Jean-Loup Guillaume. Age, Gender and Communication Networks. NetMob 2010.
Tags: age, gender, mobile social networks
