Linguistics alumnus analyzes social data in work for the Gender Gap Tracker
Junette Gonzales is a recent linguistics and psychology alumnus of Simon Fraser University (SFU). Her fascination with the study of language motivated her to become of a research assistant in Professor Maite Taboada’s Discourse Processing Lab. The research conducted in this lab examines discourse structure through linguistic and computational perspectives. Looking back at her time in the lab, Gonzales is glad that she had the opportunity to be a part of using big data to promote equality.
Initially intending to declare a major in cognitive science, Gonzales took Linguistics 220 – Introduction to Linguistics as part of the cognitive science curriculum. In this class, she became fascinated by the different subfields in the study of language. From the patterning of speech sounds in different languages to the perception and production of speech, Gonzales knew that she wanted to spend her time at SFU focused on linguistics.
In upper division, Gonzales took Linguistics 482W – Discourse Analysis which was taught by Professor Maite Taboada. The material that was covered in the course prompted Gonzales to read up on the work being done in Taboada’s Discourse Processing Lab. She was particularly interested in the SFU Opinions and Comments Corpus, a compilation of more than 630,000 comments from The Globe and Mail that has been used to identify constructive and toxic comments. Knowing that the work being done in discourse analysis could be used to tackle realworld problems using real-world data, Gonzales applied to become a research assistant in the lab.
When she was given an opportunity to work on a new project in the lab, The Gender Gap Tracker, Gonzales knew that she would be able to continue working to tackle those real-world problems that had caught her attention when she first encountered Taboada’s research. The Gender Gap Tracker was developed in collaboration with Informed Opinions and monitors the proportion of women and men quoted in news stories in mainstream Canadian media. The tracker identifies who is mentioned and quoted from news articles with the goal of promoting gender parity in the quoted sources.
“I wanted to help bring social change by showing that there is a huge difference in how often women are used as sources compared to men,” says Gonzales, “we need to close this gap to get balanced perspectives in news media.”
Gonzales worked as a manual annotator on the Gender Gap Tracker. She would highlight the relevant information such as the people mentioned, their gender, and the content of the quotes from news articles. By doing this, the team working on the tracker could ensure that the it was accurately recognizing and assigning names from the articles to the correct gender of the person cited. “If I annotate for a person who is quoted in a news article, we wanted to see the same person being identified by the tracker,” says Gonzales.
It is often the case that the world of big data is viewed as a field for computing scientists, but looking at big data through the lens of social science brings a new perspective to the collection, analysis, and reporting of human-centred and human-created data. Gonzales was drawn to the Discourse Processing Lab because she knew that thinking critically about the linguistic data that is available in so much of our news media could reveal how we behave as a society. Through analyses of this data, linguists can present information that can help us improve as a society while maintaining ethical standards that prove to the broader community that data can and should be trusted.
“I consider my biggest success at SFU as being part of the Gender Gap Tracker,” says Gonzales, “even though I have completed my work on it, I know that it will continue to encourage change regarding representation in news media.”