Abstract by Ellie Van De Graaff

Personal Infomation

Presenter's Name

Ellie Van De Graaff



Degree Level




Abstract Infomation


Computer Science

Faculty Advisor

Christophe Giraud-Carrier


Quantitatively Evaluating Social Capital on Twitter


Much of communication today has changed from being face-to-face to being conducted through a screen, whether it be via texting, emails, or social media. In our study, we seek to determine a method to quantitatively measure social capital on Twitter. We do this by evaluating how often people tag, retweet, and direct message the people they consider their close friends on Twitter, as well as how often they feel it is reciprocated. In this way, we are looking to be able to discern who a person’s close friends are just by their activity on Twitter. We collect our data with a series of surveys on Amazon’s Mechanical Turk asking people about their Twitter usage and evaluate this data analyzing for frequency of answers between users. A significant contribution of this study is that once able to identify who a person’s close friends are, we can also determine which people would be best to contact if that person requires assistance in any manner.