#Metoo: People’s concerns, emotions, and shared information on Twitter

Tahamtan, Iman

Permalink

https://hdl.handle.net/2142/105381 Copy

Description

Title

#Metoo: People’s concerns, emotions, and shared information on Twitter

Author(s)

Tahamtan, Iman

Issue Date

2019-09-24

Keyword(s)

• Sexual harassment
• Sexual assault
• #metoo
• Social movement
• Text mining
• Social networks
• Social media

Abstract

This study uses text mining to explore the tweets posted and shared by people regarding the #MeToo movement. We used the Twitter’s Application Programming Interface (API) to collect English language tweets that contained #MeToo or MeToo keywords. Using RStudio, 17956 tweets (re-tweets excluded) were retrieved and analyzed. The data was cleaned, tweets with more than 5 hashtags and screen names with multiple tweets were excluded, resulting in 10952 tweets. The most frequently shared words were #metoo (n=10701), women (n=2728), movement (n=1879), sexual (n=1330), harassment (n=818), rape (n=724), accused (n=717), don’t (n=678), people (635), and stand (n=510). The most frequent negative sentiments were harassment (n=638), rape (n=575), assault (n=280), abuse (n=239), afraid (n=212), uncomfortable (n=172), allegations (n=165), bad (n=136), hurts (n=136) and wrong (n=133). The top positive sentiments were support (n=223), love (n=119), powerful (n=86), golden (n=73), survivor (n=68), respect (64), safe (57), free (53), protect (53) and supporting (n=50). The network analysis of keywords with a correlation of higher than 0.6 demonstrated 5 clusters of keywords: {study, hurts, mentor, afraid, @nypost}, {#metoo, forces, legal, test}, {represent, caught, middle, accusers, unions}, {#fightfor15, standing, UK, workers, fighting}, and {teaching, consent, debate, kids}. Results demonstrated the major topics shared by people on Twitter regarding sexual harassment and the MeToo movement. For example, one cluster pointed to a recent study which indicated managers are afraid of mentoring women after the #MeToo movement.

Series/Report Name or Number

• Machine learning
• Social media
• Information use
• Data visualization
• Big data

Type of Resource

text

Permalink

http://hdl.handle.net/2142/105381

Owning Collections