Lessons Learned from Topic Modeling Analysis of COVID-19 News to Enrich Statistics Education in Korea

Abstract

This study aimed to investigate how mass media in Korea dealt with various issues arising from COVID-19 and the implications of this on statistics education in South Korea during the recent pandemic. We extracted news articles with the keywords “Corona” and “Statistics” from 18 February to 20 May 2020. We employed word frequency analysis, topic modeling, semantic network analysis, hierarchical clustering, and simple linear regression analysis. The main results of this study are as follows. First, the topic modeling analysis revealed four topics, namely “macroeconomy”, “domestic outbreak”, “international outbreak”, and “real estate and stocks”. Second, a simple linear regression analysis displayed two rising topics, “macroeconomy” and “real estate and stocks” and two falling topics, “domestic outbreak” and “international outbreak” regarding the statistics related to COVID-19 as time passed. Based on these findings, we suggest that the high school mathematics curriculum of Korea should be revised to use real-life context to enable integrated education, social justice for statistics education, and simple linear regression analysis.