Gender Differences in Language About Feminism: Results From Sentiment Analysis and Use of Emojis on Twitter

Abstract

Social networks, such as Twitter with its around 192 million active users per day, are increasingly changing the way how people access information, communicate with each other, express opinions and discuss a wide range of topics. An example of a rather controversial topic is feminism. This study tries to shed light on the used language and emojis when discussing feminism on Twitter. Emojis are graphic symbols, representing inter alia facial expressions, but also objects, food or drinks animals, or emotions and feelings. For the analysis, 195.843 evaluable tweets were collected between the end of February until the beginning of March 2021, covering the International Women’s Day and part of Women's History Month. Then, a quantitative approach is employed to evaluate the sentiment value of tweets on a lexical level. Sentiment analysis enables the investigation of public emotions about events, opinions, persons etc. Together with the sentiment value of the emojis, it provides the basis to analyse the identified words and topics of the discussions on Twitter. Additionally, as Twitter does not provide the gender of a user, the gender is tried to be derived from unstructured data such as the screen or username as well as the description. Results indicate that female users send in average tweets with a more positive tone than male users, while negative tweets are not significantly different between genders. Emojis are only used in a part of all tweets. The emojis used is correlated to the sentiment value of the tweet.



Author Information
Peter Bußwolder, Independent Scholar, Germany

Paper Information
Conference: ECAH2021
Stream: Sexuality

This paper is part of the ECAH2021 Conference Proceedings (View)
Full Paper
View / Download the full paper in a new tab/window


Comments & Feedback

Place a comment using your LinkedIn profile

Comments

Share on activity feed

Powered by WP LinkPress


Share this Research

Posted by amp21