Evaluating online review helpfulness based on Elaboration Likelihood Model: the moderating role of readability

Li, Boying; Hou, Fangfang; Guan, Zhengzhi; Chong, Alain Yee-Loong; Pu, Xiaodie

Abstract

It is important to understand factors affecting the perceived online review helpfulness as it helps solve the problem of information overload in online shopping. Moreover, it is also crucial to explore the factors’ relative importance in predicting review helpfulness in order to effectively detect potential helpful reviews before they exert influences. Applying Elaboration Likelihood Model (ELM), this study first investigates the effects of central cues (review subjectivity and elaborateness) and peripheral cues (reviewer rank) on review helpfulness with readability as a moderator. Second, it also explores their relative predicting power using the machine learning technique. ELM is tested in online context and the results are compared between experience and search goods. Our results provide evidence that for both types of products review subjectivity can play a more significant role when the content readability is high. Furthermore, this study reveals that the dominant predictor is varied for different product types.

Citation

Li, B., Hou, F., Guan, Z., Chong, A. Y., & Pu, X. (2017). Evaluating online review helpfulness based on Elaboration Likelihood Model: the moderating role of readability.

• Conference Name: 21st Pacific Asia Conference on Information Systems (PACIS 2017)
• End Date: Jul 20, 2017
• Acceptance Date: Jan 1, 2017
• Publication Date: Jul 16, 2017
• Deposit Date: Jun 26, 2018
• Publicly Available Date: Jun 26, 2018
• Peer Reviewed: Not Peer Reviewed
• Keywords: Review helpfulness; Elaboration Likelihood Model (ELM); readability; search goods; experience goods
• Public URL: https://nottingham-repository.worktribe.com/output/873040
• Publisher URL: http://aisel.aisnet.org/pacis2017/257/