Linguistic factors in digital entertainment success: How review readability affects movie outcomes on Chinese online platforms
Xi Fang,Aihua Han,Yifang Luo,WonHo Choi,Fan Zhang
DOI: https://doi.org/10.1016/j.entcom.2024.100911
IF: 2.072
2024-12-11
Entertainment Computing
Abstract:Online movie reviews play a crucial role in shaping the digital entertainment landscape by influencing consumer decision-making. However, little research has investigated how the readability of these reviews impacts film performance. This paper examines the relationship between the readability of reviews on online platforms and movie market outcomes, including box office revenue, audience ratings, and popularity. We collected a dataset of 116,771 Chinese-language reviews from the Douban platform for 224 movies released in China between January 2019 and September 2022. Using natural language processing techniques, we calculated various readability metrics for each movie's reviews. We then analyzed how these readability scores correlated with the movies' economic and ratings performance. The results show that movies with more readable reviews tend to have higher box office returns, better audience reviews, and larger viewership. However, this effect is weakened when movies receive coverage from more authoritative official media sources or when the movies are not entertaining enough. We also find that the number of "useful" upvotes on reviews partially mediates the relationship between readability and movie outcomes. This study contributes novel insights into how the linguistic features of user-generated content can impact the success of digital entertainment products. The findings can help platforms design more effective review systems and assist studios in marketing movies online. Future work could extend this approach to other domains like book reviews or video game feedback.
computer science, cybernetics, interdisciplinary applications, software engineering