Tech Science Seminar

Date: 

Tuesday, February 12, 2019, 3:00pm to 4:30pm

Location: 

CGIS Knafel K354

Assessing Gender Representation in Podcasts Via Transcript Analysis By the Example of NPR’s Planet Money

NPR is at the forefront of podcasting, the most quickly growing digital medium in the United States, with programs that attract tens of millions of monthly listeners. At the same time, podcasts are known to suffer from a lack of diverse perspectives. Its creators tend to be disproportionally men, which has been shown to lead to blind spots in terms of the subjects that are covered. 

This paper analyzes 528 episodes of the popular NPR show Planet Money, revealing a marked gender imbalance in on-air time favoring male speakers. The majority of this difference is explained by male guests being featured three times as frequently as female guests. I further find that an increase in the number of female hosts is associated with a small increase in the number of female guests, consistent with existing literature on newsroom diversity. Overall, my findings suggest that changes in staff diversity are necessary but not sufficient in achieving increased diversity on air.

Podcast_Gender_Study

 

Speaker: Jan Geffert is a senior in Currier House concentrating in Computer Science. He is interested in how society and technology shape each other and in particular how digital technologies can serve as tools of liberation. Before coming to Harvard, Jan was Campaign Coordinator at Schüler Helfen Leben, Germany's largest youth-led humanitarian organization, building the digital infrastructure supporting more than 80,000 student activists.

A Multi-Agent Simulation for Bike-Sharing to Identify a Feasible Cap

As urban populations grow, the importance of last mile connectivity and sustainable transportation grows, and bike-sharing offers a sustainable, healthy, and congestion-limiting option for commuters. However, planning systems has been difficult, as it can be difficult to identify locales that would benefit from cycling infrastructure. I will present a multi-agent simulation that simulates the City of Boston and its bike-sharing program, Bluebikes, hoping to find the required system capacity to meet bike-sharing demand in Boston.

Bike_Share_Simulation

Speaker: My name is Dhruv Gupta and I’m a junior at the College studying Computer Science and Government. I’ve been particularly excited about urban mobility ever since I played SimCity 4 back in the day. That led me to explore the intersections between local government and technology, hoping to find a way to build a more sustainable future. Outside of classes and research, I'm a part of Harvard CBE, a sustainability consulting group, as well as Harvard Raftaar, Harvard's Bollywood dance group. Fun fact about me: I’ve got 2.5 fake teeth!

See also: Seminars