TWO ESSAYS ON USER-GENERATED CONTENTS IN MARKETING

Abstract

This dissertation studies two topics regarding user-generated content (UGC) in social media context. Due to its rapid proliferation, UGC has received much attention among both practitioners and researchers. Although past studies have investigated diverse aspects of UGC and their influences on consumer decision process, few studies have examined (1) status bias in online peer-feedback on UGC with the presence of reputation system, and (2) the impact of video UGC, a relatively new type of UGC, on product purchase and usage decisions.

In Essay 1, we investigate how a reputation system affects peer-evaluations in an online community. In contrast to previous research on reputation systems, which has predominantly shown that reputation systems can induce posting activity and quality contributions, we study how reputation markers (“badges,” for example) may change peer-evaluations. We rely on a unique and detailed data set and employ a difference- in-differences approach, combined with propensity score matching, that suggests that, all else equal, posters receive disproportionately-higher evaluations of their posts

after they earn a reputation marker, irrespective of post quality. This suggests a “rich- get-richer” process induced by a reputation system that may have the unintended effect of favoring some participants at the expense of equally-skilled others.

In Essay 2, we examine the impact of a relatively new type of user-generated content, video user-generated content (VUGC), on consumer’s purchase and product usage decision. Due to its highly visual and auditory characteristics, VUGC has been considered as an effective marketing tool. However, as VUGC can have the risk of revealing too much information about the entertainment products, firms need to limit the amount of original content in VUGC. Under these circumstances, game developers need to have an understanding on how VUGC can impact sales and usage. To examine the impact, we collected VUGC activities and game characteristics from diverse sources. Our results indicate that VUGCs can affect sales and game usage, and the impact can vary by VUGC and game product characteristics. Our findings suggest that game developers should consider both VUGC characteristics and their product characteristics jointly when developing and implementing policies on users’ VUGC uploading, sharing, and monetizing.