Michael Reusens & Talita Greyling on "National Sentiment Statistics Through Social Media: Obstacles And Opportunities"

Michael Reusens holds a Ph.D in Business Economics and is currently coordinator for data & methods at Statistiek Vlaanderen. He is a highly qualified data scientist, did research in data mining and published in top journals. 

Talita Greyling specializes in well-being economics and quality of life studies and has a keen interest in fourth industrial revolution applications. She has developed the “Happiness Index” using Big Data and machine learning and is the director of the Gross National Happiness.today Project (GNH.today), which continuously research and develop the index. Since 2020 she has been actively involved in COVID-19 research projects. She is the associate editor for Pioneers in Quality of Life Theory and Research for the Journal of Applied Research in Quality of Life (impact factor 3.2) and a co-editor for economics in the Journal of Happiness Studies (impact factor 3.9). She is the Vice-President (Membership) of the International Society for Quality-of-Life Studies and is a Research Fellow of the Global Labor Organization (GLO). Talita is involved in several international collaborations related to well-being research.

Webinar content: 

Understanding the sentiment of a nation's population can provide valuable insights into societal well-being, political stability, economic trends and educational needs. Social media platforms offer a wealth of real-time data that can be analyzed to develop a national sentiment statistic. As such, it presents a lot of opportunities. However, there are also obstacles associated with using data science to automate national statistics based on social media.

In this presentation, we address these obstacles by discussing different topics related to the use of Twitter data to develop a national sentiment statistic for Flanders, Belgium, and the measure gross national happiness in different countries. We highlight the challenges associated with selection bias, annotator bias, and model performance. We then propose solutions to these challenges, including techniques for predicting annotation difficulty and correcting for selection bias.