Investigating the Effect of eWOM in Movie Box Office Success Through an Aspect-Based Approach

Investigating the Effect of eWOM in Movie Box Office Success Through an Aspect-Based Approach

Saurav Mohanty (Department of Decision System Sciences, Saint Joseph's University, Philadelphia, PA, USA), Nicolle Clements (Department of Decision System Sciences, Saint Joseph's University, Philadelphia, PA, USA) and Vipul Gupta (Department of Decision System Sciences, Saint Joseph's University, Philadelphia, PA, USA)
Copyright: © 2018 |Pages: 15
DOI: 10.4018/IJBAN.2018010101

Abstract

This study examines the influence of Electronic Word of Mouth (eWOM) on the box office revenue generation of movies in the U.S domestic market using the technique of Aspect-Based Sentiment Analysis (ABSA) and aspect identification. The analysis was conducted on the sentiment score and frequency of five movie aspects from the user reviews collected from high grossing 2014 movies. This study revealed a significant dependence on the aspect-based sentiment frequency of the movie's Story aspect. Surprisingly, the data also showed a strong dependence of movie success on the negative sentiment frequency on the Casting aspect. The findings of the study suggest that the eWOM present in online movie reviews can be used to predict the performance of a movie at the box office by monitoring the aspect's frequency of sentiment, which can be referred to as a metric of the online “buzz” of the movie.
Article Preview

Introduction

Hollywood, as an industry, had been growing with leaps and bounds having a huge number of successful movie releases each year and generating billions of dollars in box office revenue (Litman, 1983). Considering the magnitude of the business involved, researchers have endeavored on determining indicators that can accurately predict the financial success of movies. Since the early 1970s, research in this direction has caught momentum despite the apparent difficulties involved. Sochay (1994) examined the performance of motion pictures released in United States and Canada between October 1987 and October 1989 by measuring the parameters of domestic rentals and the length of run of the movie, while also taking into consideration the competition from other films. Moving forward, the development of analysis methodology and easier availability of data paved the way for more comprehensive research regarding the financial performance in the movie industry. The measures of performance and the method of analysis evolved in many ways. Many scholarly articles have emphasized the role of critics as being very prominent in the film industry (Eliashberg & Shugan, 1997; Holbrook, 1999; West & Broniarczyk, 1998). The first decade of the 20th century saw the proliferation of the internet into every domain of society. Movies began to be discussed online through various forums, not only by expert reviewers and critics, but also by the general movie audience. Thus, massive volumes of electronic Word of Mouth (eWOM) were generated regarding these films. More recently, the concept of eWOM has become very important in the field of social media mining and is being recognized as a rich source of business intelligence. In fact, Word-of-Mouth (WOM) has been acknowledged as one of the most influential means of conveying information in modern history (Gupta & Gupta, 2016; Godes & Mayzlin, 2004; Maxham & Netemeyer, 2002; Reynolds & Beatty, 1999). The eWOM present in social media has even stronger influence considering the amount of people it reaches in a short amount of time. Research later revealed the importance of eWOM together with expert reviews as indicators to gauge performance of movies at the box office (Basuroy, Chatterjee, & Ravid, 2003; Dellarocas, Zhang, & Awad, 2007; De Vany & Walls, 1996; Duan, Gu, & Whinston, 2008; Elberse & Eliashberg, 2003; Liu, 2006; McKenzie, 2009).

This study analyzes the eWOM of the user reviews online through a new aspect-based sentiment analysis (ABSA) and aspect identification to investigate its effect on the box office revenue collection. Thus, the two methodologies adopted in this research are 1) ABSA and 2) Frequency of Aspect Sentiment (FAS), also called “buzz.” The structure of the paper is as follows. First, background information regarding ABSA and FAS is presented, followed by a literature review of previous work in this area and how this paper contributes to the field. Next, the authors present a methodology section which describes the research framework, data collection, and analysis models employed. Finally, the results are given along with discussions and conclusions.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 6: 4 Issues (2019): Forthcoming, Available for Pre-Order
Volume 5: 4 Issues (2018)
Volume 4: 4 Issues (2017)
Volume 3: 4 Issues (2016)
Volume 2: 4 Issues (2015)
Volume 1: 4 Issues (2014)
View Complete Journal Contents Listing