Analyzing Facebook - A visual analysis of social structure and usage incentives of 100 facebook users

Analyzing facebook A visual analysis of social structure and incentives of 100 Facebook users usage Created by Simon Stolz using R, Adobe Photoshop and Gephi. Number of Friends On average participants had476 Facebook friends. The histogramm shows that some networks contained more than 1000 friends. 2. 200 400 600 800 1000 1200 However, being asked how many actual friends they have, participants reported an average of only 72.45 This means around 15% among all Facebook friends are "true friends". 40 50 The scatterplot shows how the correlation between Facebook friends and actual friends is low. The correlation coefficient is only O.17 (o = no correlation, 1 = perfect correlation) 200 400 600 800 1000 1200 Facebook friends Having more Facebook friends does not mean having more actual friends. Mutual Friends On average Facebook friends of a participant had 27.08 mutual connections. A randomly selected Facebook friend, in a network of 476 friends, knows around % of all the other friends. Usage Motives I asked participants what they are using Facebook for. Each score was assessed with three questions, similar to what is quoted in the barplot. "Sharing things with others that I would find interesting to see myself" Altruism "To express my feelings and to express and show myself" Self-expression "To recall memorable events and to Ego-centric keep track of my life" 2 4 6. 10 Determinants of interaction frequency Using OLS-Regression I tested which factors have the strongest impact on the frequency of information exchange in the form of wall posts, comments and private messages on Facebook. Four variables turn out to have pronounced associations: Dispersion of the network Education Altruism Social Capital Information Exchange Small-world network features Ego-centric Number of Self-expression Facebook friends Age Gender As intuitively understandable individuals who report motives of self-expression frequently write wall posts and are more likely to comment. However, individuals with higher education are significantly less likely to write wall posts, but write as many comments and messages. Networks that were particularly large and included many Facebook friends are associated with significantly more information exchange of all types. A dispersed network, as it may reflect an open attitude and befriending with strangers, is associated only with more messages. To read the full thesis, please visit www.simonstolz.de Although users report to use Facebook for mostly altruistic reasons, it is primarily the degree of self-expressional motives that actually make people post. Simon Stolz (.de) Infographic “Analyzing Facebook", 2014 Actual friends Frequency 400 500 12 00L 007 008