Impact of information consistency in online reviews on consumer behavior in the e-commerce industry: a text mining approach
Qinglong Li,Jaeseung Park,Jaekyeong Kim
DOI: https://doi.org/10.1108/dta-08-2022-0342
IF: 1.713
2023-06-13
Data Technologies and Applications
Abstract:Purpose The current study investigates the impact on perceived review helpfulness of the simultaneous processing of information from multiple cues with various central and peripheral cue combinations based on the elaboration likelihood model (ELM). Thus, the current study develops and tests hypotheses by analyzing real-world review data with a text mining approach in e-commerce to investigate how information consistency (rating inconsistency, review consistency and text similarity) influences perceived helpfulness. Moreover, the role of product type is examined in online consumer reviews of perceived helpfulness. Design/methodology/approach The current study collected 61,900 online reviews, including 600 products in six categories, from Amazon.com. Additionally, 51,927 reviews were filtered that received helpfulness votes, and then text mining and negative binomial regression were applied. Findings The current study found that rating inconsistency and text similarity negatively affect perceived helpfulness and that review consistency positively affects perceived helpfulness. Moreover, peripheral cues (rating inconsistency) positively affect perceived helpfulness in reviews of experience goods rather than search goods. However, there is a lack of evidence to demonstrate the hypothesis that product types moderate the effectiveness of central cues (review consistency and text similarity) on perceived helpfulness. Originality/value Previous studies have mainly focused on numerical and textual factors to investigate the effect on perceived helpfulness. Additionally, previous studies have independently confirmed the factors that affect perceived helpfulness. The current study investigated how information consistency affects perceived helpfulness and found that various combinations of cues significantly affect perceived helpfulness. This result contributes to the review helpfulness and ELM literature by identifying the impact on perceived helpfulness from a comprehensive perspective of consumer review and information consistency.
computer science, information systems,information science & library science