Project background:

Participation in voluntary associations is vital for a healthy democracy. With most research focusing on the conventional associations, e.g. labor unions or sports groups, the recent popularization of online social groups, e.g. Facebook and Meetup groups, calls for more investigation of the newer forms of associations. Online social groups ease and transform means of participation and can add valuable insights on the associational life of individuals in the digital age. As part of my dissertation, this project will examine how the network characteristics of online voluntary associations create and are reinforced by the engagement and mobilization of the current and prospectus members in major political events, such as the presidential elections and midterm elections. I plan to use the data I collected from Meetup.com to answer my research questions. Python and R will be my main tools in analyzing the data.

Proposed project objectives:

1. Design and develop visualizations demonstrating how the social network structure of Meetup groups affects members’ future political participation and mobilization.
2. Deepen my skills in using Python and R for analysis and visualizations. 

There definitely have been ups and downs during my first few weeks of the fellowship.

Ups:

Initially, I was planning to analyze the data collected from New York City because that was all I had. Specifically, the dataset was a longitudinal dataset (2002-2017) that contained participation records of attendees in 8,581 groups, including attended event(s) and its corresponding time; attendees’ group membership(s); and users’ basic information, participated group, and event location and descriptions. During the first couple of weeks of the fellowship, I was able to expand my data by collecting information from more cities on – San Francisco, Chicago, Austin, Seattle, etc.. This potentially gives me more power to compare how cities differ in their online groups’ network structures and how the characteristics of the networks might impact the organization of the social groups differently.

As a beginner in Python, I also spent a lot of time googling, trying out codes, and troubleshooting. Although not an expert yet, I have learned a lot on coding in Python and am getting comfortable with it. I was able to create some descriptive visualizations, such as change in the volume of memberships over time and a co-participation network with weighted edges.

Downs:

I’m currently also narrowing down the research questions for my dissertation so that I can come up with visualizations of networks that actually speak to my research questions. However, I’m at the stage where I’m not sure what questions I’m actually asking therefore the types of visualization I should be making. In order to move myself forward, I’m reading more literature to understand the field of social networks better.

This has been my experience at the Studio so far. In the remainder of the fellowship, I plan to finalize my research questions and come up with preliminary visualizations that are theoretically useful and visually entertaining.

-Qianyi Shi