YouTube Comments on Gene-Edited Babies: What Factors Affect Diverse Opinions in Comments?

Abstract

This study explored the factors that influence video popularity and diverse opinions in the comments of YouTube videos about gene-edited babies. 107 most viewed videos and corresponding 56,912 direct comments about gene-edited babies were collected from YouTube. We examined how the uploader characteristics, delivery styles, video tones, and video frames affect the diverse opinions measured by sentiment polarity, sentiment divergence, and the number of topics in the commentary. The sentiments and topics of comments were analyzed using IBM Watson Natural Language Understanding. We found that more effective videos are relatively longer videos from user-created channels with a large number of subscribers and presented in a neutral tone, which is more likely to provide unbiased and comprehensive knowledge about gene-edited babies. Based on our findings, suggestions were made for viewers about how to pick high-quality content, and insights were provided for content creators about how to create compelling videos.