THE GOOD, THE BAD, AND THE NEUTRAL: TWITTER USERS’ OPINION ON THE ASUU STRIKE

Abstract

Aim/Purpose Nigeria’s university education goes through incessant strikes by the Academic Staff Union of Universities (ASUU). This strike has led to shared emotion on micro-blogging sites like Twitter. This study analyzed selected historical tweets from the “ASUU” to understand citizens’ opinions. Background The researchers conducted sentiment analysis and topic modelling to under-stand Twitter users’ opinions on the strike. Methodology The researchers used the Valence Aware Dictionary for Sentiment Reasoning (VADER) technique for sentiment analysis, and the Latent Dirichlet allocation (LDA) was used for topic modelling. A total of 10,000 tweets were first ex-tracted for the study. After data cleaning, 1323 tweets were left. Contribution To the researchers’ best knowledge, no published study has presented a senti-ment analysis on the topic of the ASUU strike using the Twitter dataset. This research will fill this gap by providing a sentiment analysis and drawing out sub-jects by exploring the tweets on the phrase “ASUU.” Findings The sentiment analysis result using VADER returned 567 tweets as ‘Negative,’ with the remaining 544 and 212 categorized as Positive and Neutral. The result of the LDA returned six topics, all comprising seven keywords. The topics were the solution to the strike, ASUU strike effect, strike Call-off, appeal to ASUU, student protest and student appeal. Recommendations for Practitioners Practitioners may also use the research to understand better the attitudes of their staff and students about the strikes to create actionable solutions before the suspension of the strike. Recommendations for Researchers Researchers can use this study’s findings to compare with other contexts of opinion mining. Impact on Society The findings of this study may be utilized by policymakers to better understand the viewpoints of citizens on the ASUU strike to propose feasible solutions. Future Research Future studies can collect information from other social networking and blog-ging sites.