A Practical Guide for Social Media Research: Data Collection and Sentiment Analysis

Social media has emerged as a vital source of data for understanding people. However, inconsistencies regarding data collection and affective computing approaches may limit the utility of this data in communication research. Therefore, this research will act as a practical guide for communications scholars who wish to collect social media data to systematically analyze the emotions conveyed in social media conversations. While previous research has focused on the tools and technologies used, this study focuses on the data collection process and sentiment analysis as a common analytical method. The study compares data collection using APIs and third-party platforms, as well as sentiment analysis using state-of-the-art machine learning techniques. The results of the comparison will offer guidelines for social media researchers and practitioners and a public repository will be created to facilitate the use of these methods in computational social science research.