The Shades and Shapes of the Pink Wave: Visual Perspectives of the Women's March

• In Event: Automated Image Analysis in Practice

Abstract

The information that media provides about political events and discussions related to gender issues lay the foundation of opinions and attitudes towards them. Therefore, it is crucial to understand the content and symbols that media outlets use to communicate messages in order to assess their impact and identify potential biases. In this project we focus on studying the ways in which newspapers portray the Women's March through the use of images. First, we study the prevalence of images and news related to the most recent Women's March on the covers of local newspapers in the U.S. We then use a novel technique from the computer vision field to conduct a visual structural topic model using the corpus of images from newspaper covers. We use this approach to achieve two goals: identify underlying frames in the images, and estimate the effect that newspapers' attributes, like ideology or region, have on the generation of such frames. Finally, we aim to compare these frames to those generated on other platforms like Twitter in an attempt to study whether there is any bias in the reporting of the Women's March with respect to what the movement aims to communicate. This project allows us to have a better understanding of the way in which women's issues are framed through the use of visuals in order to shed light on other mechanisms in which opinions about gender issues are generated.

Authors

• Jonathan Homola, Rice University

• Michelle Torres, Rice University