Uncovering social support among users affected by violence on social media: An analysis of violence-related posts and reactions on Reddit (Preprint)

Author:

Li LifangORCID,Wang 2nd Tao,McManus 3rd Sally,Stewart 4th Robert,Roberts 2nd Angus

Abstract

BACKGROUND

Domestic and other forms of violence are widespread and can cause victims’ distress. Reddit, a popular social media platform, has become a space where victims can share their experiences of violence and seek support from others. However, research is lacking on the type of information shared by victims about their experiences and on how other Reddit users respond to their posts.

OBJECTIVE

To fill the research gap, we aim to use data from seven violence-related subreddits to characterize how violence victims were supported by other users.

METHODS

We (1) identified themes in victims’ posts about violence using BERT models; (2) analyzed the sentiments of posts and their corresponding comments, testing for variations by theme; and (3) examined the types of support users provided in all comments and in negative comments responding to the three most common themes by using Biterm Topic Modelling.

RESULTS

The results show that Reddit provides a platform for victims to share their experiences of violence, and that these can be grouped into eight themes. The most common themes were “Emotional Pain and Communication Struggles in Abusive Relationships”, “Emotional Impact and Coping Strategies in Abusive Relationships” and “Physical Violence and Confrontations in Domestic Settings”. Responses (comments) tended to carry positive or neutral sentiment, and were more likely to provide emotional or informational support than instrumental or appraisal support.

CONCLUSIONS

The results indicate that victims of violence relying on social media platforms like Reddit for support may need additional support from services to meet appraisal needs (information which useful for self-evaluation).

Publisher

JMIR Publications Inc.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3