Don't be so emotional! How tone of voice and service type affect the relationship between message valence and consumer responses to WOM in social media
| Published date | 13 November 2017 |
| DOI | https://doi.org/10.1108/OIR-08-2016-0219 |
| Pages | 905-920 |
| Date | 13 November 2017 |
| Author | Freya De Keyzer,Nathalie Dens,Patrick De Pelsmacker |
| 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 |
Don’t be so emotional! How tone
of voice and service type affect
the relationship between message
valence and consumer responses
to WOM in social media
Freya De Keyzer, Nathalie Dens and Patrick De Pelsmacker
Department of Marketing, University of Antwerp, Antwerp, Belgium
Abstract
Purpose –The purpose of this paper is to shed light on the boundary conditions of the effect of the valence of
word-of-mouth on social networking sites (sWOM) on consumer responses (attitude toward the service
provider, purchase intention and positive word-of-mouth intention). Specifically, the authors examine two
moderators: the tone of voice (factual vs emotional) of the sWOM and service type (utilitarian vs hedonic)
of the service that the sWOM is about.
Design/methodology/approach –A 2 (message valence: positive vs negative) ×2 (tone of voice: factual vs
emotional) ×2 (service type: utilitarian vs hedonic) full-factorial between-subjects online experiment with
400 respondents was conducted and the data were analyzed using Hayes’PROCESS macro.
Findings –The results show that message valence exerts a greater impact on consumer responses with
factual sWOM messages compared to emotional ones. Furthermore, the impact of messagevalence is stronger
for hedonic services compared to utilitarian services. In contrast to the authors’expectations, there is no
significant impact of matching the tone of voice to the service type.
Practical implications –First, for sWOM senders, factual messages are found to be more influential:
backing an sWOM up with arguments and specific details increases the chance of it affecting consumers’
responses. As a result, marketers, especially of predominantly hedonic services, should encourage their
followers and customers to spread positive factual sWOM about their service.
Originality/value –The study tests two previously unstudied moderating variables that affect the
relationship between message valence and consumer responses to sWOM messages. Moreover, this study
provides interesting insights for marketers and bloggers or reviewers.
Keywords Social media, Facebook, Service type, sWOM, Tone of voice
Paper type Research paper
Introduction
Consumers’increased use of online communications is reflected in their word-of-mouth
(WOM) behavior (Karjaluoto et al., 2016). Sharing daily consumption online is an important
part of modern life (Kim et al., 2015). Hennig-Thurau et al. (2004) define electronic word-of-
mouth (eWOM) as: “any positive or negative statement made by potential, actual, or former
customersabout a product or a company, which is madeavailable to a multitude of peopleand
institutionsvia the Internet.”(p. 39). eWOM can occurthrough review sites (e.g. TripAdvisor),
(micro) blogging platforms (e.g. Twitter), video sharing sites (e.g. YouTube) and social
networkingsites (SNSs, e.g. Facebook). In the presentstudy, we are interested in the latter, for
which the term sWOM has beencoined (Balaji et al., 2016). sWOM is eWOM on SNSs, which
are, “applications that enable users to connect by creating personal information profiles,
inviting friends and colleagues to have access to those profiles, and sending e-mails and
instant messages between each other”(Kaplan and Haenlein, 2010, p. 63).
In general, eWOM is acknowledged to greatly impact consumer attitudes, behavioral
intentions (e.g. Cheung and Thadani, 2012) and sales (Rui et al., 2013). Sociable Labs (2012)
report that 62 percent of online shoppers have read product-related messages
from Facebook connections. The limited research on sWOM seems to suggest that
Online Information Review
Vol. 41 No. 7, 2017
pp. 905-920
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-08-2016-0219
Received 17 August 2016
Revised 19 February 2017
5 June 2017
Accepted 29 June 2017
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
905
WOM in
social media
sWOM, like other forms of eWOM, is perceived by consumers as trustworthy and credible
(Chu and Kim, 2011) and exerts a positive effect on consumers’online brand engagement,
brand awareness, brand attitude and purchase intention (PI) (Karakaya and Barners, 2010;
Schivinski and Dabrowski, 2015; Wang et al., 2012).
It may, however, be important to distinguish sWOM from the broader concept of eWOM.
First, research on eWOM already suggests that not all digital WOM is created equal. In an
explicit comparison between consumer reviews and microblogs, Marchand et al. (2016)
document differential effects of the valence and volume of the two types of eWOM on video
game sales. Second, some research documents differences between eWOM and sWOM
(e.g. Lin et al., 2017). From a sender’s perspective, one of the major implications of sWOM is
that the senderis identifiable and can be heldaccountable, which resultsin a higher social risk
(Balaji et al.,2016; E isingerich et al., 2015), while other forms of eWOM (e.g. online reviews) may
be anonymous. From a reader’s perspective, sWOM has a unique advantage of referability:
the contributor’s SNS profileprovides for a higher levelof source credibility,benefiting sWOM
adoption (Hajli, 2016). Identification with the sender is also a relatively unique aspect
of SNSs that drives readers’PI (Wang et al., 2012). Bachleda and Berrada-Fathi (2016)
suggest that negative sWOM from a Facebook friend is less influential than negative eWOM
from a consumer review site, because readers place less trust in sWOM.
Users’motivation to visit SNSs and review sites also differs, which could result in
different responses to WOM on these platforms (Gvili and Levy, 2016). SNSs are mainly
used to pass time and for amusement and social exchange (Ku et al., 2013), whereas review
sites are consulted to read information on a product or service in which the user is already
interested (Reichelt et al., 2014). Exposure to brand information in sWOM is more voluntary
(Chu and Kim, 2011). The meta-analysis of BabićRosario et al. (2016) indicates that the effect
of sWOM on sales is weaker than the effect of eWOM through e-commerce platforms.
sWOM does, on the other hand, entail an increased risk that customer complaints go viral
(i.e. shared on a massive scale on SNSs), causing a potential public relations crisis for a firm.
It is therefore essential that the determinants and consequences of sWOM are examined
(Balaji et al., 2016).
The beneficial effect of positive message valence on attitudes (Purnawirawan et al., 2015),
brand loyalty and perceived brand quality (Schivinski and Dabrowski, 2015), PIs (Bae and
Lee, 2011) and even sales (Floyd et al., 2014) of the message recipients is consistently found
in the eWOM literature (e.g. Floyd et al., 2014; Lee and Koo, 2012; Purnawirawan et al., 2015),
and has been replicated in one sWOM study (Rui et al., 2013). However, this effect can be
reinforced or weakened by moderating factors.
The current study offers a number of contributions. According to Barger et al.’s (2016)
conceptual framework on consumer engagement in social media, an antecedent that needs
more extensive research is the “content factor.”Our study responds to this call by inspecting
how the tone of voice of a message (whether the sWOM message takes on a more factual
or emotional tone) moderates the effect of message valence on consumer responses.
Moreover, we also inspect the moderating effect of service type (whether the service mainly
fulfills utilitarian or hedonic buying motivations) on the effect of sWOM valence, which in
the framework proposed by Barger et al. (2016), can be categorized as a “product factor.”
By studying the combination of these variables, we are able to explore the boundary
conditions of the effects of sWOM message valence.
This study sets out to corroborate findings on the role of message valence and tone of
voice previously found in an eWOM context in an sWOM, context and further tests the
moderating effect of service type. We test our propositions for services, since sWOM has
been found to be more influential when services are discussed (compared to products)
(BabićRosario et al., 2016). We focus on both traditional brand responses (attitude and PI)
(e.g. Park and Lee, 2008; Purnawirawan et al., 2015), and consumer engagement, more
906
OIR
41,7
To continue reading
Request your trial