The asymmetric effect of review valence on numerical rating. A viewpoint from a sentiment analysis of users of TripAdvisor
| Pages | 283-300 |
| Date | 08 April 2019 |
| Published date | 08 April 2019 |
| DOI | https://doi.org/10.1108/OIR-11-2017-0307 |
| Author | Hsiu-Yuan Tsao,Ming-Yi Chen,Hao-Chiang Koong Lin,Yu-Chun Ma |
| Subject Matter | Library & information science,Information behaviour & retrieval,Collection building & management,Bibliometrics,Databases,Information & knowledge management,Information & communications technology,Internet,Records management & preservation,Document management |
The asymmetric effect of review
valence on numerical rating
A viewpoint from a sentiment analysis
of users of TripAdvisor
Hsiu-Yuan Tsao and Ming-Yi Chen
Department of Marketing, National Chung Hsing University, Taichung, Taiwan, and
Hao-Chiang Koong Lin and Yu-Chun Ma
Department of Information and Learning Technology,
National University of Tainan, Tainan, Taiwan
Abstract
Purpose –The basic assumption is thatthere is a symmetric relationship between reviewvalence and rating,
butwhat if review valenceand rating werelinked asymmetrically?There are f ew studies which have investigated
the situations in which positive and negative online reviews exert different influences on ratings. This study
considers brand strength as having an important moderating role because the average rating of existing reviews
for a particular product is a heuristic cue for decision makers. Thus, the purpose of this paper is to argue that an
asymmetric relationship between review content valence and numerical rating will depend on brand strength.
Design/methodology/approach –The authors have conducted a sentiment analysis via text mining, using
self-developed computer programs to retrieve a data set from the TripAdvisor website.
Findings –This study finds there is an asymmetric relationship between review valence (verbal) and
numerical rating. The authors further find brand strength to have an important moderating role. For a
stronger brand, negative review content will have a greater impact on numerical ratings than positive review
content, while for a weaker brand, positive review content will have a greater impact on numerical ratings
than negative review content.
Practical implications –Marketers could adopt sentiment analysisvia text mining of online reviews as a valid
measure or predictor of consumer satisfaction or numerical ratings. Strong brands should direct more attention
to negative reviews, because in such reviews the negative impact transcends the positive. In contrast, weak
brands should aim to exploit as many positive reviewsas possible to minimizet heimpact of any nega tivereviews.
Originality/value –This study finds there is an asymmetric relationship between review valence (verbal)
and numerical rating and considers brand strength to play an important moderating role. The authors have
used real data from the TripAdvisor website, which allow people to express themselves in an unsolicited
manner, and linked these with the results from the sentiment analysis.
Keywords Sentiment analysis, Online review, Text mining, Asymmetric effect, Brand strength
Paper type Research paper
1. Introduction
We are now well into the era of big data, which comprises everything from ubiquitous
social networking sites, such as Facebook and Twitter, to a myriad of smaller online
information communities, such as that included in this study, namely TripAdvisor.
Online information communities, such as TripAdvisor, are important examples of data hubs
which may help consumers to make decisions about the quality of products or services.
At the same time, the benefit to consumers of reading online reviews is that they provide
real-world accounts of how fellow consumers rate products or services in which they are
interested (Cheung et al., 2008). Thus, the online review is a market phenomenon that plays a
large and increasingly important role in purchase decisions (Chen and Xie, 2008; Chiou et al.,
2014; Lee et al., 2013; Li et al., 2013; Yan et al., 2015), in a process by which consumers inform Online Information Review
Vol. 43 No. 2, 2019
pp. 283-300
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-11-2017-0307
Received 1 November 2017
Revised 23 June 2018
Accepted 11 August 2018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
The authors thank helpful reviewers’comments from 2016 American Marketing Association (AMA)
Summer Marketing Educators’Conference.
283
Asymmetric
effect of
review valence
each other of their opinions about the quality of a product or service. Simultaneously,
through learning about consumer opinions, marketers may choose to apply such lessons to
furthering the appeal of their products to the consumer. Li and Hitt (2008) suggested that
enterprises armed with information of this kind can modify their marketing strategies,
including such factors as pricing, advertising or product design. Thus, marketers in
e-commerce can encourage their targeted consumers to spread positive reports and, by
word-of-mouth, create an early positive buzz for a product or service (Hung, 2017).
Several studies published in recent years have examined the impact of online reviews on
different outcome variables,such as sales (Chevalier andMayzlin, 2006; Floyd et al., 2014) and
the helpfulness of online reviews (e.g. Chen, 2016; Li et al., 2013; Mudambi and Schuff, 2010;
Schlosser, 2011;Yin et al., 2014; Goudas et al., 2015; Lin et al.,2017). However, the relationship
between online review valence (i.e. positive vs negative) and numerical rating is a topic that
has attracted negligible attention to date. This issue is particularly important, because to
determine if a reviewer is able to convey the truth, consumers will likely compare review
comments andthe rating. In this case, disconfirming expectationsshould reduce (not enhance)
credibility. In the absence of thorough empirical testing of the proposition, online reviewing
cannot yet be acceptedas a fully reliable indicator of productquality ratings (Koh et al., 2010).
The basic assumption is that there is a symmetric relationship between review valence and
rating, but what if review valence and rating were linked asymmetrically? There are few
studies which haveinvestigated the situations in which positive and negative online reviews
exert differentinfluences on ratings.This study considersbrand strength to play an important
moderating role, because the average rating of existing reviews for a particular product
is a heuristic cue for decision makers. Furthermore, stronger brands provide more credible
signalsthan weaker brands, becausethey are more susceptibleto the loss of established brand
equity (Erdemand Swait, 1998) and future salesand profit (Wemerfelt, 1988).Thus, this study
argues that an asymmetric relationship between reviewcontent valence and numericalrating
will depend on brand strength. To fill this research gap, the main purpose of this study isto
gain a better understanding of an asymmetric relationship between positive/negative reviews
and numerical rating. Moreover, we have considered an important moderating role: brand
strength.In addressing this issue, wehave performed a sentiment analysis (or opinionmining)
to understand the surrounding positive or negative connotations.
Compared with previous studies, this research provides additional contributions in
three ways. First, it contributes to the literature on the basic assumption that there is a
symmetric relationship between review valence and rating, because it assumes there is an
asymmetric relationship between review valence (verbal) and numerical rating. We
further consider brand strength to play an important moderating role. Second, this study
performs sentiment analyses to gain more insight into the composition of reviews.
Sentiment analysis is a useful technique with which to assist marketers in determining
how a brand or product is perceived in relation to value and quality. For example,
subtasks of sentiment analysis include determining subjectivity, the degree and valence o f
opinion (positive or negative), and classifying the subject matter and author. Sentiment
analysis, through the monitoring of social media or online discussion forums, can change
the way firms measure consumer opinions. Pang and Lee (2008) presented a detailed and
comprehensive review of affective computing and computer technology for the
recognition of emotion and expression. Thus, sentiment analysis is a useful research
method in either text or review mining. The results derived from a sentiment analysis
represent a first step toward a better understanding of the nature of reviews. Third, this
study uses real data from the TripAdvisor website, which provide a complete spectrum of
hotel consumer reviews, including the text of consumers’comments, numerical ratings
and the average rating of existing reviews for a particular brand (i.e. hotel quality or
ranking), by using measures which allow people to express themselves in an unsolicited
284
OIR
43,2
To continue reading
Request your trial