Journal Article

Integrating Social Media Sentiment Analysis and Traditional Polling

Alsharhan, Eiman Tawfeeq, Al-Nashmi, Eisa Nashmi

2025

Overview

This research explores the ongoing debate on whether social media sentiment analysis can replace traditional polling and advocates for the combined use of both methods to capture nuanced public perceptions expanding the geographic and linguistic scope of public opinion literature, this study focuses on Kuwait and the Arabic language. It examines public reactions to the Ministry of Higher Education's decision on June 9, 2023, to suspend scholarships for medical students at universities in Jordan and Egypt. The goal was to assess how data from both social media and traditional surveys can aid the government in understanding public opinion towards various policies. A significant contribution of this study is the development of the first multi-label emotion-classification tool specifically designed for Kuwaiti Arabic. This tool's unique capability to interpret and analyze the intricacies of the Kuwaiti dialect provided invaluable insights into local contexts and sentiments, setting a new benchmark for cultural and linguistic specificity in emotion-classification tools. The findings indicate that, while social media alone may not fully convey actual public opinion on political issues, adding survey data yields a richer, multi-dimensional perspective. This approach emphasizes the importance of integrating both methodologies to inform policymaking. The study further highlights the critical role of public opinion in shaping government policies and suggests that using both methods could lead to more informed decisions, enhance public trust, and reduce the likelihood of electoral sanctions.

• Publisher: جامعة الكويت - مجلس النشر العلمي
• Subject: السياسات الحكومية / الكويت / تصنيف المشاعر / وسائل التواصل الاجتماعي