Title
Analyzing the Overturning of Roe vs Wade on Twitter using Natural Language Processing Techniques
Date of Award
Spring 5-2023
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computational and Data Sciences
First Advisor
Erik Linstead
Second Advisor
Elizabeth Stevens
Third Advisor
Elia Eiroa Lledo
Abstract
In 1973, the historic U.S. Supreme Court (SCOTUS) case of Roe vs. Wade provided the constitutional right
to abortion. However, on May 2, 2022, Politico magazine leaked the draft opinion on the Dobbs v. Jackson Women’s Health Organization. The leak generated a surge of users to post their opinion on the case that would eliminate abortion as a constitutional right. Then, on June 24, 2022, SCOTUS overturned Roe vs. Wade. In this thesis, we aim to investigate the public opinion and reaction towards the overturning of Roe vs. Wade. We collected 20,640,166 tweets using Twitter API for Academic Research and an open-sourced dataset published during two periods. The first period was a week before Politico magazine leaked the
SCOTUS decision and the week after. The second period was a week before and over a week after the
overturning of Roe vs. Wade. Using natural language processing techniques, including sentiment analysis,
emotion recognition, topic modeling, and bi-grams, we could develop insight into public opinion based on
the posted tweets. Our research investigates if there is a change in sentiment over time, a change in the
emotion expressed within the text over time, and which topics are most common within the collection of
tweets. The results demonstrate a significant increase on the day of the Politico leak, which showed that
most of the tweets published on that day expressed a positive sentiment. However, in the weeks before and
after the overturning of Roe v. Wade, we witness a decrease from the beginning of the period up to the day
of the overturn. Regarding emotion recognition, there is a significant decrease in the proportion of tweets
classified as expressing optimism. There’s also an increase in the proportion of tweets expressing anger when comparing the day of the Politico leak and the day of the overturn. The topic model we applied to the
tweets published on the day of the Politico leak revealed that states’ rights and children were discussed. Using bigram of the most negative tweets, we witnessed gun control and healthcare as words that frequently occurred within the collection.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
G. Pinto, "Analyzing the overturning of Roe vs Wade on Twitter using natural language processing techniques," M. S. thesis, Chapman University, Orange, CA, 2023. https://doi.org/10.36837/chapman.000485