Information cascades in online reading: an empirical investigation of panel data

Purpose - The purpose of this paper is to examine information cascades in the context of users' e-book reading behavior and differentiate it from alternative factors that lead to herd behavior, such as network externalities and word-of-mouth effects. Design/methodology/approach - This paper constructed panel data using information concerning 226 e-books in 30 consecutive days from Sina. com's reading channel (Book.Sina.com.cn) from October 2, 2013, to October 31, 2013 of the same year in China. A multinomial logit market-share model was employed. Findings - E-books' ranking has a significant impact on their market share, as predicted by informational cascades theory. Higher ranking e-books' clicks will see a greater increase as a result of an increase in clicks ranking. Due to the information cascades effect, review volume had no impact on the market share of popular e-books. Total votes had a powerful impact on the market share of e-books, showing that once information cascade occurred, it could be enhanced by the increase in total votes. The total clicks of e-books had a significant impact on their market share, suggesting that online reading behavior would be influenced by network externalities. Practical implications - As important information, the ranking or popularity of e-books should be carefully considered by online reading web sites, publishers, and authors. It is not enough for the authors and publishers of e-books to simply pay attention to the content. They should design their marketing strategies to allow network externalities and informational cascades to work for them, not against them. Online reading web sites should also focus on eliminating certain behavior, such as "brush clicks" and "brush votes," in order to prevent an undesirable information cascade due to false information. Originality/value - To the best of the knowledge, this is the first study to examine information cascades in the context of users' e-book reading behavior. Moreover, this study can help other researchers by utilizing a large sample of daily data from one of the earliest online reading platforms in China.