icon bookmark-bicon bookmarkicon cameraicon checkicon chevron downicon chevron lefticon chevron righticon chevron upicon closeicon v-compressicon downloadicon editicon v-expandicon fbicon fileicon filtericon flag ruicon full chevron downicon full chevron lefticon full chevron righticon full chevron upicon gpicon insicon mailicon moveicon-musicicon mutedicon nomutedicon okicon v-pauseicon v-playicon searchicon shareicon sign inicon sign upicon stepbackicon stepforicon swipe downicon tagicon tagsicon tgicon trashicon twicon vkicon yticon wticon fm
15 Apr, 2021 12:15

Tweets more likely to go viral if they’re negative, study finds

Tweets more likely to go viral if they’re negative, study finds

Using advanced language analysis techniques, researchers in Spain have found that Twitter posts with negative sentiments are more likely to go viral than their positive counterparts, once all external factors are controlled for.

Researchers from the SINAI Department of Computer Science, CEATIC and Universidad de Jaen conducted an analysis of Twitter posts during the highly divisive Catalan independence referendum in Spain in 2017. 

Virality has become a powerful and much sought-after phenomenon online, with many businesses and online personalities seeking to boost their influence or fortune by ‘going viral’ on social media. Virality is a double-edged sword, however, and going viral for the wrong reasons can be the downfall of a business, individual, or movement. 

To understand what drives virality on Twitter, “one of the main channels for the dissemination of political messages” according to the study authors, they examined 46,962 Tweets from 25,847 different user accounts over the span of one week in October 2017. 

Once they controlled for factors such as the number of followers the poster has, the specific content, or features within the post itself including things like hashtags, as well as the particular topic being discussed, the researchers found that tweets with negative sentiment on the topic reached a wider audience than positive sentiment tweets.

This is an initial foray into sentiment analysis with regards virality, and the results may depend on specific circumstances or audiences, the researchers concede, though their conclusion remained that “negativity in a tweet increases the probability of retweeting.”

Text mining, natural language processing and sentiment analysis are emerging fields when it comes to online discourse, and researchers are gradually developing automatic identification and analysis of the opinions and emotions expressed in online texts, which may significantly impact the future of social media, online discourse, and wider political discussion.

Think your friends would be interested? Share this story!