Quantifying Social Media Sentiment in Predicting the Malaysian 13th General Elections

Abstract

The 13th General Elections (GE13) were regarded as the most competitive, contentious and divisive ever in the history of Malaysian General Elections. It was aptly described as the ‘mother of all elections’, while Prime Minister Najib Razak termed it as the ‘social media election’. Communications scholars and analysts have long studied the increasing role of social media in political and democratic activism. However, to what extent is the online discourse and sentiment in the social media reflective of the actual voting pattern and behaviour? This paper hypothesised that social media can predict future electoral outcomes. Methodologically, this research employed a web content analysis across major social media platforms throughout the GE13 campaign period. The data and findings of this research showed a significant 95 % prediction accuracy in comparison to the actual GE13 results.