This article was downloaded by: [Tulane University]

On: 28 January 2015, At: 05:04

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Advertising:

The Review of Marketing
Communications
Publication details, including instructions for authors and
subscription information:
[Link]

Online word of mouth and consumer

purchase intentions
a a b
Gerard Prendergast , David Ko & V. Yuen Siu Yin
a
Hong Kong Baptist University
b
Classroom Publications Limited
Published online: 07 Jan 2015.

To cite this article: Gerard Prendergast, David Ko & V. Yuen Siu Yin (2010) Online word of mouth
and consumer purchase intentions, International Journal of Advertising: The Review of Marketing
Communications, 29:5, 687-708

To link to this article: [Link]

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at [Link]
and-conditions
 Online word of mouth and
consumer purchase intentions
Gerard Prendergast and David Ko
Hong Kong Baptist University
Siu Yin V. Yuen
Classroom Publications Limited

Previous researchers have explained how offline word of mouth (WOM) influences
consumer behaviour. Increasingly, however, consumers are turning to online WOM,
Downloaded by [Tulane University] at 05:04 28 January 2015

especially online forums. In an online context, the research discussed in this article rep-
licates the approaches taken to studying offline WOM. Grounded in social comparison
theory, social network analysis and the theory of reasoned action, a mall intercept survey
of consumers in Hong Kong showed that both similarity between a user’s interests and a
forum’s topic and user attitudes towards the forum strongly predict purchase intentions as
well as having an indirect effect through helping determine the forum’s persuasiveness.
Theoretical implications are discussed as well as practical implications for marketers and
recommendations for further research.

Introduction
Consumers search for information prior to making a purchase in order to
reduce perceived risk (Bettman 1973; Srinivasan & Ratchford 1991). One
such source of information is word of mouth (WOM). Since the seminal
book by Katz and Lazarsfeld (1955), marketing researchers have shown
that WOM communication has an impact on consumer attitudes (Brucks
1985), consumer risk taking (Woodside & Delozier 1976), short-term and
long-term product judgments (Bone 1995) and purchase decisions and
choice behaviour (Lau & Ng 2001). It is even argued that since WOM is
consumer-dominated, it is perceived as being more reliable and trustwor-
thy than firm-generated communications (Arndt 1967). Indeed traditional
forms of communication such as advertising appear to be losing their
effectiveness (Nail 2005), possibly because consumers doubt their reli-

International Journal of Advertising, 29(5), pp. 687–708

© 2010 Advertising Association
Published by Warc, [Link]
DOI: 10.2501/S0265048710201427 687
 International Journal of Advertising, 2010, 29(5)

ability and trustworthiness. The credibility of WOM, however, combined

with the fact that consumers will be more involved with WOM than with,
say, advertising, lends itself to the formation of higher order cognition and
stronger effects (Brown et al. 2007).
WOM need not just be communicated face-to-face. Increasingly, con-
sumers are turning to the internet as an information source (Fallows 2005;
Peterson & Merino 2003). This is enabling real-time content due to inter-
activity, response and conversation (Kliatchko 2008). There are consider-
able benefits for consumers in using the internet due to the large amount
of information available, the entertainment available while searching,
freedom from sales pressure, and 24-hour access (Alba et al. 1997; Hoffman
& Novak 1996; Wolfinbarger & Gilly 2001). While internet information
Downloaded by [Tulane University] at 05:04 28 January 2015

is often provided by company-sponsored websites and advertisements,

increasingly, online WOM, in the shape of online communities such as
internet forums, bulletin boards, and newsgroups, is providing consumers
with the ability to share their experiences, opinions and knowledge with
others. As a result of such participatory media, audiences are no longer just
receivers of media content but are simultaneously creators of the content
through texts, pictures, videos, music, and so on (Kliatchko 2008). When
the forums are consumption-related, the members may share enthusiasm
and knowledge about a specific consumption experience or related group
of activities (Kozinets 1999). As with offline WOM, online WOM may be
positive or negative. Chevalier and Mayzlin (2006), for instance, studied
online book reviews and found that the impact of negative reviews on
book sales was greater than for positive reviews.
Although probably 90 per cent of WOM communication takes place
offline (Keller & Berry 2006), a growing role for online WOM is likely as
consumers rely more and more on the internet as a source of information.
Are the theories used to explain the effectiveness of offline WOM appro-
priate for explaining the effectiveness of online WOM? With offline WOM
the communicators are in close proximity and can draw upon a wealth of
social context cues (Knapp & Daly 2002). Online, however, people create
fully formed impressions of others based solely on the linguistic content
of written messages (Brown et al. 2007). Online WOM presents numerous
opportunities for marketers, yet empirical evidence relating online WOM
to firm performance is limited (Trusov et al. 2009). If online WOM does
indeed influence consumer behaviour, then marketers need to know the

688
 Online word of mouth and consumer purchase intentions

mechanisms through which this occurs. To address this need, this study
set out to borrow theory used in the offline WOM context and apply it to
online WOM. More specifically, social comparison theory, social network
analysis, and the theory of reasoned action were invoked to explain how
WOM generated in online forums might influence purchase intentions
about products discussed in the forums. Replicating the approaches taken
to studying offline WOM in an online environment is felt to be both suit-
able and important, since theoretically WOM activity ought to operate
generally in a similar manner in both environments, yet the contextual
issue may result in yet to be discovered variations in the application of
the theory.
Downloaded by [Tulane University] at 05:04 28 January 2015

Background and hypotheses

Offline and online WOM

Traditional offline WOM involves face-to-face two-way communication
within a social relationship (Money et al. 1998). In the offline situation, one
WOM message can reach and potentially influence many receivers, but
usually by passing through a chain or tree of correspondents (Lau & Ng
2001). Online WOM information, in contrast, ought to diffuse faster, since
there may be hundreds, thousands, or even millions of people receiving
and potentially spreading an initial message through email and other inter-
net-based media. Although the decision as to whether to pass on a piece of
information still lies with each recipient, compelling news may spread like
a ‘memetic virus’ on the internet (Gottsch 2001), such that the influence of
online WOM can be greater than that of offline WOM. With online WOM,
social context cues are much more limited. Does this inhibit our ability
to apply WOM theory from an offline context to online communication?
Grounded in attribution theory, an experiment by Lee and Youn (2009)
examined how product reviews generated by various platforms (company
website, independent website and personal blogs) influenced attitude
towards the product, intention to buy and intention to recommend it to
friends. Unexpectedly, they found that participants exposed to the review
on the personal blogs were less likely to recommend the product to friends
than those exposed to the review on the other platforms, supporting the
discounting principle in attribution theory (Kelley 1971, 1973). While Lee

 689
 International Journal of Advertising, 2010, 29(5)

and Youn’s experiment is of great value, the authors admit problems in

terms of external validity, stating that future research should study more
natural online WOM processing, as well as extending the research beyond
student samples and single product categories.

The persuasiveness of online forums

An online forum is an online community where visitors may read and post
topics. A source’s intention in providing information is a major factor in the
receiver’s perception of the information’s trustworthiness (Romani 2006).
Product experiences posted by forum members are likely to be perceived as
reliable because the information providers are fellow consumers, who pre-
Downloaded by [Tulane University] at 05:04 28 January 2015

sumably have no interest in marketing the product. The forum members are
sharing their experiences but they do not have the specific objective of influ-
encing the reader. This is unlike advertising, where the company intends to
encourage recipients to buy the product. Furthermore, online forums often
generate empathy (the ability to connect emotionally with another indi-
vidual) among forum members (Galinsky et al. 2008). Participants in online
discussions are, in a sense, ‘performing’ for other forum members through
their contributions to the forum (Baym 1997). Sharing personal experiences
on a forum causes the reader to empathise with the feelings of the writer,
enhancing persuasiveness (Deighton et al. 1989).

Similarity of the forum’s interests to the receiver’s interests

Source similarity or homophily refers to the degree to which individuals
are similar in terms of certain shared social characteristics (Rogers 1983).
Festinger’s (1954) theory of social comparison provides a mechanism
explaining how source similarity influences persuasiveness. This theory
proposes that people tend to compare their attitudes and capabilities
with those of others. The tendency to compare oneself with another
person increases as that person is seen to be similar to oneself, because,
according to Festinger, individuals implicitly assume that similar people
have similar needs and preferences. A source perceived as similar to the
receiver is therefore more persuasive than a source perceived as dissimi-
lar (Eagley & Chaiken 1993). Many related theories have followed from
Festinger’s work. Heider’s (1958) balance theory argues that people tend

690
 Online word of mouth and consumer purchase intentions

to like others who exhibit signs of similarity because it is reinforces their

own self-concept and helps them to predict and understand others. Also,
the source-attractiveness model (Kelman 1961) suggests that receivers can
better identify with and understand sources that are similar to themselves,
and therefore perceived sender similarity increases the persuasiveness of
the information transmitted. Byrne (1971) has proposed a general law of
attraction predicting that attraction towards an individual is a linear func-
tion of attitudinal similarity with that person.
Whatever its basis, similarity can lead to greater persuasive influence
(Hass 1981). The origins of the influence of source similarity, at least
physical similarity, most likely lie in evolutionary psychology. A study
by Alvarez (2005) indicated that facial resemblance was a strong driving
Downloaded by [Tulane University] at 05:04 28 January 2015

force for mating—human couples resemble each other significantly more

than would be expected from random pair formation. Because facial char-
acteristics are inherited, any such ‘self-seeking like’ mechanism should
promote reproduction between genetically similar mates, favouring the
stabilisation of genes supporting social behaviour, as long as other mecha-
nisms deter mating among close kin.
The theories related to social comparison and source similarity were
developed and tested in an offline context. Would they operate online as
well? Online, information providers may or may not have demographics
and lifestyles that are similar to those of the reader, but they are often
similar to the reader in that they share a common interest in the topic
of the forum (Blanton 2001) and to some extent a shared group identity
(Brewer & Weber 1994). This makes the information exchanged more
relevant to the reader, especially since it can be expected to reflect typi-
cal product performance. Supporting this logic, social network analysis
(Haythornthwaite 1999) would suggest that greater similarity between the
interests of the member and those of the forum should result in informa-
tion being perceived as more credible and persuasive (Brown et al. 2007)
because this matching enhances tie strength. The strength of a tie is a
combination of the amount of time it has been in force, the emotional
intensity, the intimacy (mutual confiding) and the reciprocal services
which characterise the tie (Granovetter 1973). Such tie strength, even in
an online context, influences behaviour (De Bruyn & Lilien 2008). This is
especially the case when a participant in a forum shares the same interests
as the other forum members. Consider forums relating to religion (e.g.

 691
 International Journal of Advertising, 2010, 29(5)

MyChurch), pets (e.g. Dogster) or specific hobbies (e.g. Ravelry), Ploderer

et al. (2008) suggest that sites such as these connect mostly people with no
previous offline connection. Instead these online communities tend to be
organised around a common interest providing a hierarchical structure for
people to discuss topics. Active participation in such communities fosters
behaviour in line with the group’s norms to gain social approval, to uphold
group identity and to maintain group harmony (Khaled et al. 2006).
As Brown and his co-workers have argued (Brown et al. 2007), social
network analysis is suitable for studying online WOM behaviour because
its unit of analysis is the exchange of (tangible and intangible) resources
between social actors. Each exchange is considered a social exchange rela-
tion, and individuals who maintain the relation are said to maintain a tie.
Downloaded by [Tulane University] at 05:04 28 January 2015

Social network analysis suggests that the behaviour of online forum par-
ticipants is affected more by the kinds of ties and exchanges involved than
by individual attributes of the actors themselves (Haythornthwaite 1999).
Indeed, in an online context the individual attributes of the communica-
tors are usually unknown.
Forum members may be anonymous to each other, but they are similar
to the extent that they have voluntarily joined a forum which presumably
deals with their common attitudes and interests. With extended participa-
tion, bonding (ties) would be expected to grow. Individual attributes of the
participants themselves become irrelevant and the forum communication
itself becomes the point of comparison and influence. In other words, the
content of the forum becomes a proxy for the individual members and
therefore the source in the communication process. This is analogous to
the concept of self-brand connections (Chaplin & John 2005). As consum-
ers engage in a matching process to identify products or brands that are
congruent with their self-image (e.g. Birdwell 1968), consumers presum-
ably also engage in a matching process to identify congruent online forums
to follow. Indeed Brown et al. (2007) have suggested it is forum content,
rather than the characteristics of the individual members, which demon-
strates similarity of interests. Here, content refers to the actual textual
content of the forum’s messages without reference to who provided them.
So, source similarity is conceptualised here as similarity between the user’s
interests and the forum’s topic.
What is the mechanism through which similarity between the user’s
interests and the forum’s topic influences a user’s behaviour? Social

692
 Online word of mouth and consumer purchase intentions

comparison­ theory suggests that similarity ought to be positively related

to the persuasiveness of information being communicated, and to behav-
ioural intentions as well. Since persuasiveness is the mechanism through
which similarity influences behavioural intentions, the same logic might
be expected to apply in the online context. While the process is the same,
the underlying reasoning is somewhat different in an online context, since
social networking analysis would suggest that the forum itself becomes
the source rather than individual members, and members will be attracted
to forums that relate to their interests. This shared common interest with
the forum strengthens ties, resulting in the forum’s content having a more
persuasive influence on the behaviour of its members. It is based on this
reasoning that the following hypotheses are proposed:
Downloaded by [Tulane University] at 05:04 28 January 2015

H1: Source similarity has a positive effect on the persuasiveness of an

online forum.

H2: Source similarity strengthens purchase intention with respect to

products discussed in an online forum.

H3: The persuasiveness of an online forum mediates any positive

relationship between perceived source similarity and purchase
intentions with respect to products discussed in the forum.

Receiver’s attitude toward the forum

Attitudes towards advertising in a specific medium appear to explain why
consumers engage in advertising avoidance across a range of media (Lee
& Lumpkin 1992), including the internet (Cho & Cheon 2004). Findings
indicate that more positive attitudes towards an advertising medium
generally lead to less avoidance of ads in that medium. Online forums,
however, are not presenting advertising. How about consumer attitudes
towards them? Would such attitudes predict the influence of the forum on
their behaviour?
Consumers’ attitudes toward interacting with online communities
should be of central concern to marketers (Brown et al. 2007). According
to Brown’s study, participating in an online consumer community can lead
to increased purchasing and facilitate participation in community-related

 693
 International Journal of Advertising, 2010, 29(5)

activities. If an individual has a positive attitude towards a forum, the

intensity of the relationship between the individual and the forum is likely
to increase, as manifested in the individual’s visiting the forum more regu-
larly. At the same time, the influence of any information picked up from
the forum ought to increase. Fishbein and Ajzen’s (1975) research on the
theory of reasoned action (TRA) serves as the basis for much of the work
examining this link between attitude and behaviour. According to the
TRA, attitude is a crucial predictor of behavioural intention. As with other
cognitive decision-making models, the underlying premise of the TRA is
that individuals make decisions rationally and systematically on the basis
of the information available to them, but attitude is nevertheless a core
construct in the TRA. The TRA has proven itself in extensive empiri-
Downloaded by [Tulane University] at 05:04 28 January 2015

cal testing in relation to a diverse range of health, social and consumer

behaviours, and meta-analyses demonstrate strong support for the pro-
posed determinants of intention (Sheppard et al. 1988). Using the TRA,
we predict that a more positive attitude towards a forum should result in
information gleaned there being more persuasive, and this persuasiveness
should influence purchase intentions about any products discussed.

H4: A member’s positive attitude towards an online forum has a posi-

tive effect on the persuasiveness of an online forum.

H5: A member’s positive attitude towards an online forum strength-

ens purchase intention with respect to products discussed in an
online forum.

H6: The persuasiveness of an online forum mediates any positive

relationship between attitude towards an online forum and pur-
chase intentions with respect to products discussed in the forum.

Methodology

Sampling
To test the hypotheses, a mall intercept survey was conducted in Hong
Kong. Hong Kong has one of the largest internet penetrations in the world
and internet use spans all age groups. According to the World Bank (World

694
 Online word of mouth and consumer purchase intentions

Development Indicators 2009) there are 59.1 internet users per 100 peo-
ple in Hong Kong. Point Topic (2008) indicates that Hong Kong is placed
5th in the world in terms of internet penetration, behind Monaco, South
Korea, Singapore and Macau.
When collecting data for regression analysis, the number of cases should
be at least 40 times the number of independent variables (including both
moderating and mediating variables) to allow a reasonable chance of dem-
onstrating any underlying relationships, assuming reliable measurements
and a dependent variable which is not skewed (Tabachnick & Fidell
2001). This study was designed to investigate relationships involving
three independent variables, so a total of 150 interviews were conducted.
Mall intercepts (Sudman 1980) were used because many Hong Kong
Downloaded by [Tulane University] at 05:04 28 January 2015

people pass through shopping malls on weekends, providing a convenient

cross section of Hong Kong’s population. The field work was conducted
on weekends from 3:00 pm to 9:00 pm over a four-week period at major
shopping malls in the three main districts of Hong Kong (Hong Kong
Island, Kowloon and the New Territories). The shopping malls selected
were surrounded by dense residential developments. Every 4th passer-by
was selected to avoid self-selection bias. Internet forums were explained
at the beginning of the interview. Examples of forums were given to
ensure that the interviewees fully understood them. The interviews were
conducted by one of the authors who had been fully briefed. After the
briefing a series of mock interviews were conducted.

Measurements
All scales are shown in the Appendix. At the start of the questionnaire,
each interviewee was requested to specify the one consumption-related
online forum (i.e. a forum that includes discussion of products/services)
that they visited most frequently and to answer all questions in relation to
that forum.
Similarity between the individual’s interests and the forum group’s
interest was measured using a four-item, seven-point semantic differen-
tial scale developed by Wolfinbarger and Gilly (1993), where 1 = strongly
dissimilar and 7 = strongly similar. This scale was originally developed to
measure similarity between people. ‘This person’ was replaced by ‘forum
group’. The fifth item in the original scale was deleted as it was not

 695
 International Journal of Advertising, 2010, 29(5)

relevant­ to the current context (it measured gift givers and gift receivers’
tastes in gifts). Attitude towards the forum was measured using an adapted
version of the six-item, five-point Likert scale developed by Chen and
Wells (1999). This scale was originally used to measure attitudes toward
a website. The persuasiveness of the forum’s information was measured
using a three-item, seven-point semantic differential scale developed by
Gürhan-Canli and Maheswaran (2000). Purchase intention was measured
using a four-item, seven-point Likert-type measure. This was the same
scale used by Coyle and Thorson (2001), who combined three items from
Putrevu and Lord (1994) with one item originally developed by Kim and
Biocca (1997).
Downloaded by [Tulane University] at 05:04 28 January 2015

Covariates
Internet familiarity may influence perceptions of forum information. For
instance, those less familiar with the internet may be relatively naïve
about the reliability of internet information, resulting in a forum’s informa-
tion persuading them too easily (or too little). Therefore internet familiar-
ity was controlled for using a three-item, seven-point semantic differential
developed by Roehm (2001) to measure each respondent’s familiarity with
the internet.
Demographic variables were also controlled for. Education was dummy
coded, with 1 representing tertiary-level education and 0 representing sec-
ondary school or below. Sex was dummy coded with 1 representing males.
Monthly income was dummy coded, with one dummy (income dummy 1)
representing a monthly income of HK$10,000 or less and another (income
dummy 2) representing more than HK$15,000. An income between these
thresholds was used as the reference category. Age was dummy coded
with one dummy for ages of 29 or less (age dummy 1), the other for ages
of 40 or more (age dummy 2), and an age between these thresholds used
as the reference category.
As the majority of Hong Kong people are Cantonese speakers, the ques-
tionnaire was developed in English and translated into Chinese through
a translation and back-translation procedure (Brislin et al. 1973; Tamanin
et al. 2002). The questionnaire was then pretested prior to launching the
full survey.

696
 Online word of mouth and consumer purchase intentions

Finally, to limit the possibility of common method variance (Podsakoff

et al. 2003), varying response formats were used. Also, several distracter
items (‘I believe that some people are lucky and some people are
unlucky’; ‘I think Hong Kong is an attractive tourist destination’; ‘Hong
Kong is notorious for its air pollution’; ‘If SARs came back to Hong Kong,
Hong Kong would recover quickly’) were inserted into the question-
naire to prevent respondents from guessing the survey’s objective and
underlying hypotheses. Harman’s (1967) single factor test for common
method variance was applied to the data collected. This test assumes that
if significant common method variance exists, a single factor, or a general
factor that explains a significant amount of the variance, will emerge from
factor analysis when all the items are entered. The results of the factor
Downloaded by [Tulane University] at 05:04 28 January 2015

analysis revealed that there were four factors with eigenvalues above 1.0,
which together explained 67% of the variance. The largest single factor
explained 21% of the variance, which is significantly less than half.

Results
54.7% of the interviewees were male and 45.3% were female. The 20–29
and 30–39 age groups occupied 34.7% and 39.3% of the total sample
respectively. For education level, 32% and 27.3% claimed to have high
school and diploma/certificate level education respectively. The largest
group of respondents (35.3%) said they earned HK$5,001–$10,000 per
month (roughly US$700–1400). The second largest group (21.3%) claimed
HK$10,001–$15,000 per month incomes.

Validity and reliability

The factor structure of each scale was analysed separately. Principal com-
ponents analysis (PCA) using a varimax rotation revealed unidimensional
structures for all scales, with explained variance ranging from 55–88%.
The mean scores, standard deviations, reliabilities, and correlations are
shown in Table 1.

 697
 Table 1: Descriptive statistics
Mean SD 1 2 3 4 5 6 7 8 9 10
1. Purchase intention 3.68 0.89 (0.81) –
2. Persuasiveness of
4.28 1.03 (0.88) 0.470 (**)
the forum
International Journal of Advertising, 2010, 29(5)

3. Similarity to the
3.83 0.99 (0.87) 0.563 (**) 0.498 (**)
forum
4. Attitude toward
3.70 0.63 (0.82) 0.449 (**) 0.414 (**) 0.495 (**)
the forum
5. Gender dummy 0.55 0.50 –0.010 –0.059 –0.098 0.013
6. Education dummy 0.41 0.49 0.026 0.029 0.054 0.009 –0.079
7. Age dummy 1 0.41 0.49 0.034 –0.096 –0.031 –0.083 0.057 0.120
8. Age dummy 2 0.19 0.30 –0.100 –0.030 –0.178 (*) –0.035 0.073 –0.240 (**) –0.411 (**)
9. Income dummy 1 0.54 0.50 –0.093 –0.133 –0.080 –0.131 –0.008 –0.312 (**) 0.286 (**) 0.012
10. Income dummy 2 0.25 0.43 –0.013 0.032 –0.026 0.035 0.086 0.242 (**) –0.449 (**) 0.151 –0.620 (**)
11. Familiarity 4.52 1.14 (0.93) 0.119 0.073 0.054 0.010 0.045 0.229 (**) 0.218 (**) –0.472 (**) –0.074 0.048
**Indicates significance at the p ≤ 0.01 confidence level (2-tailed).
n = 150
Notes: Purchase intention was measured on a 7-point Likert scale. A higher score indicates higher purchase intention. Persuasiveness of the forum’s information was measured on a 7-point semantic
differential scale. A higher score indicates greater persuasiveness of a forum’s information. Source similarity was measured on a 7-point Likert scale. A higher score indicates greater perceived
similarity between the respondent and the forum’s interests. Attitude towards the forum was measured on a 5-point Likert scale. A higher score indicates a more positive attitude towards the forum.
Familiarity with the internet was measured on a 7-point semantic differential scale. A higher score indicates greater familiarity with the internet.
Scale reliabilities are shown in parentheses.

698
Downloaded by [Tulane University] at 05:04 28 January 2015
 Online word of mouth and consumer purchase intentions

Hypothesis testing
Regressions were used to highlight any relationships between source simi-
larity and attitude towards the forum and (i) level of persuasiveness and
(ii) purchase intention, and also any potential mediating relationships. At
each step of the mediation analysis, demographics and internet familiarity
were entered first as controls.

Source similarity – persuasiveness – purchase intention

The three-step mediated regression approach recommended by Baron
and Kenny (1986) was applied in testing for any mediating effects of per-
Downloaded by [Tulane University] at 05:04 28 January 2015

suasiveness in the source similarity–purchase intention linkage. Source

similarity was found to significantly predict persuasiveness (β = 0.495,
p ≤ 0.001). This result satisfies the first condition of the mediation, and
supports H1. To test the second mediation condition, a new model was
estimated which specified only a direct relationship between source simi-
larity and purchase intention. Without the persuasiveness mediator, source
similarity was significantly related to purchase intention (β = 0.570, p ≤
0.001). This result satisfies the second mediation condition and supports
H2. To examine the third mediation condition, persuasiveness was added
as a mediator. The results indicated that persuasiveness is significantly
related to purchase intention (β = 0.246, p ≤ 0.01). Focusing on the signifi-
cant path, source similarity still significantly predicted purchase intention
(β = 0.449, p ≤ 0.001) in the presence of persuasiveness, but there was a
reduction in β (from 0.570 to 0.449). This suggests that persuasiveness
partially mediates the impact of source similarity on purchase intentions
toward products discussed in the forum. The third mediation condition
was thus met, and H3 was supported.

Attitude – persuasiveness – purchase intention

The same three-step mediated regression approach was then used to
examine the mediating effects of persuasiveness in the attitude–pur-
chase intention linkage. To meet the first mediation condition, a positive
attitude towards an online forum was found to have a positive effect on
the persuasiveness of forum communications (β = 0.387, p ≤ 0.001). This

 699
 International Journal of Advertising, 2010, 29(5)

supports H4. To test the second mediation condition, a new model was
estimated in which only the direct path between attitude and purchase
intention was specified. Without persuasiveness as a mediator, a positive
attitude significantly strengthened purchase intention (β = 0.443, p ≤
0.001). This result satisfies the second mediation condition, and supports
H5. After adding the persuasiveness mediator, the results indicated that
persuasiveness was significantly related to purchase intention (β = 0.340,
p ≤ 0.001). Focusing on the significant path, a positive attitude still dis-
played significant predictive power for purchase intention (β = 0.308, p ≤
0.001) in the presence of persuasiveness, but there was a reduction in β
(from 0.443 to 0.308). This suggests that persuasiveness partially mediates
the impact of a positive attitude towards the forum on the intention to
Downloaded by [Tulane University] at 05:04 28 January 2015

purchase products discussed in the forum. The third mediation condition

was therefore met, and H6 was supported.

Discussion
Considerable previous research has studied the effects of offline WOM on
consumer behaviour. With the substantial growth in online forums, there is
a need for marketers to better understand how online WOM might influ-
ence consumer behaviour. The persuasiveness of online WOM, however,
has until now received limited attention. This research has replicated
the approaches taken to studying offline word of mouth but in an online
context. Building on this previous research in the offline context, Brown
et al.’s (2007) qualitative work relating to online WOM and the recent
experiment by Lee and Youn (2009), this research examined how simi-
larity between a forum’s topic and a user’s interests, as well as the user’s
positive attitude toward the forum, influence the forum’s persuasiveness.
In turn, the influence of a forum’s persuasiveness on purchase intention
toward products discussed in the forum was examined. The results show
partial mediation in both instances.
These results demonstrate that theories used to explain how offline
WOM influences consumer behaviour in daily life can also be used in an
online context. Similarity between forum topics and the receiver’s interests
has a direct relationship with purchase intention, as well as an indirect rela-
tionship through determining the forum’s persuasiveness. Persuasiveness
is a mechanism through which similarity between the forum’s interests

700
 Online word of mouth and consumer purchase intentions

and those of the receiver influences purchase intention. Similarity in

this context refers to Brown’s notion of shared group interests and group
mind-set (Brown et al. 2007). As Brown has suggested, online communities
can act as a proxy for individual identification. Consumers interact with
the forum rather than individuals when processing information delivered
online. The relationship between source similarity and persuasiveness
and purchase intention is explained by Festinger’s (1954) theory of social
comparison and its premise that individuals implicitly assume that similar
people have similar needs and preferences. This explanation is amplified
by social network analysis which predicts that greater similarity between
the interests of the member and those of the forum should result in infor-
mation being perceived as more credible and persuasive (Brown et al.
Downloaded by [Tulane University] at 05:04 28 January 2015

2007) because this matching enhances tie strength. Further explanation

for these findings is found in Kelman’s (1961) source-attractiveness model
and Byrne’s (1971) general law of attraction, which suggests that receivers
can better identify with and are more attracted to sources that are similar
to themselves.
As with similarity, attitude towards a forum has a direct effect on
purchase intention as well as an indirect effect through the forum’s per-
suasiveness. Persuasiveness is a mechanism through which attitude influ-
ences purchase intention. This is consistent with Fishbein and Ajzen’s
(1975) theory of reasoned action. If a person has a positive attitude toward
a forum, this has a positive influence on the persuasiveness of the infor-
mation contained in that forum, which in turn increases the receiver’s pur-
chase intention to products and services discussed favorably in that forum.
What are the practical implications of these findings? It seems that
online WOM can significantly influence consumer behaviour, so market-
ers should not ignore the proliferation of online consumer-to-consumer
communication. The results of this research suggest that marketers wish-
ing to capitalise on online forums as a marketing communications tool
need to ensure that the group’s interests closely match those of their
product’s users. They might then consider funding concrete steps to foster
positive attitudes towards the forum.
A marketer considering creating a forum matching target consum-
ers’ interests should first undertake primary research to identify specific
interests common among the intended users. Once the forum has been
established, discussion topics could be monitored to identify which

 701
 International Journal of Advertising, 2010, 29(5)

discussions appear to be of most interest to the target consumers, and

topics of less interest could be discouraged. Developing a forum which
matches the shared interests of target consumers would ensure that they
find the forum relevant, thereby increasing the persuasiveness of the
information it delivers and, if the information is positive, increasing the
likelihood of purchase.
Having found or created an online group whose interests match those of
the target consumers, marketers need to foster positive attitudes towards
the forum. Users need to be convinced that using the forum is a good use
of their time and that the forum is better than other similar groups. Most
forums have rules and moderators, and presumably these are important in
determining attitudes towards the forum.
Downloaded by [Tulane University] at 05:04 28 January 2015

In all of this, marketers need to remember that online forums are con-
sumer-generated and, in principle at least, marketers are on the sidelines.
As Brown has pointed out (Brown et al. 2007), marketers need to be aware
of the risks involved in attempting to influence online WOM. As soon as
a forum’s discussion moves from being consumer-controlled to marketer-
controlled, the forum can easily lose credibility and users. Related to this,
marketers have the option of adding a forum feature to their company
website (where, obviously, the creator of the forum is identified), or creat-
ing a stand-alone forum (see Procter and Gamble’s ‘[Link]’). It is
possible that consumer remarks posted on a standalone forum will be more
persuasive than remarks posted on a forum that is part of a company web-
site, although this is uncertain and warrants further research. Companies
might experiment with establishing stand alone forums while also main-
taining a forum on their own website to determine which approach is more
likely to achieve their objectives. Marketing academics and practitioners
often refer to brands being co-created by consumers and firms (Deighton
& Kornfeld 2007), but in fact little is known about how firms can promote
such co-creation. The firm’s role in maintaining some control of ostensibly
consumer-managed forums is a crucial area needing investigation.

Limitations and further research

This research has used theory from an offline environment to explain
how online forums influence consumer behaviour. The results extend the
work of Brown et al. (2007) and Lee and Youn (2009). However, several

702
 Online word of mouth and consumer purchase intentions

limitations should be recognised. First, the research did not specify if the
forum was part of a company website or a standalone forum. As pointed
out by Lee and Youn (2009), these two different approaches to forum
placement may produce different levels of persuasiveness. Second, the
variables explored in this study do not offer an exhaustive explanation
of how online word of mouth influences consumer purchase behaviour.
Other variables would presumably offer additional explanation. Third,
we only considered the influence of positive online WOM on purchase
intentions. Again as pointed out by Lee and Youn (2009), online WOM
can of course also be negative and therefore could have negative effects
on purchase intention. Finally, our study looked at perceived influences
of online word of mouth rather than actual influences and behaviour. The
Downloaded by [Tulane University] at 05:04 28 January 2015

difficulty in researching offline WOM is that word of mouth cannot easily

be traced. However in an internet setting it ought to be possible to track
online interactions (Trusov et al. 2009), so future research might consider
employing actual rather than perceived measures.

Conclusions
Several theories have been used by previous researchers to explain how
WOM influences consumer behaviour in daily life. This study has taken
some of those theories and applied them to the online environment,
thereby extending our understanding of WOM in a new context. The
results confirm that source similarity and attitude towards an online forum
have a direct relationship with intention to purchase products discussed
in the forum, as well as an indirect effect through helping determine
the forum’s persuasiveness. Understanding how online WOM influ-
ences consumer behaviour is undoubtedly important, and will become
even more important in the future as the internet continues to grow. To
develop further theoretical and practical insights into this area, it is hoped
that the findings reported here will be verified and built upon by future
researchers.

Appendix
Think of one internet forum (it must contain product/service information)
that you visit most frequently to answer all of the following questions.

 703
 International Journal of Advertising, 2010, 29(5)

Variable Items
Source similarity 1. Considering your outlook on life, how similar are you and the forum group’s interests?
(Wolfinbarger & 2. Considering your likes and dislikes, how similar are you and the forum group’s interests?
Gilly 1993) 3. Considering your values and experiences, how similar are you and the forum group’s interests?
4. How similar are this forum group’s tastes in products compared to yours?
Persuasiveness 1. I feel the product information given on the forum is strong.
(Gürhan-Canli & 2. I feel the product information given on the forum is convincing.
Maheswaran 2000) 3. I feel the product information given on the forum is persuasive.

4. I feel the product information given on the forum is powerful.

Attitude towards 1. This forum makes it easy for me to build a relationship with the online community.
the forum (Chen & 2. I would like to visit this forum again in the future.
Wells 1999) 3. I’m satisfied with the services provided by the forum.
4. I feel comfortable in surfing the forum.
5. I feel surfing this forum is a good way for me to spend my time.
Downloaded by [Tulane University] at 05:04 28 January 2015

6. Compared with other forums, I would rate this one as the best one.
Internet 1. In general, how is the Internet to you? (familiar/not at all familiar)
familiarity (Roehm 2. In general, how is the Internet to you? (recognise it right away/don’t recognise it right away)
2001) 3. In general, how is the Internet to you? (know it very well/don’t know it well at all)
Behavioral After considering the product information on the forum that I visit most often:
intention (Coyle & 1. It is very likely that I will buy the product.
Thorson 2001) 2. I will purchase the product next time I need a product.
3. I will definitely try the product.
4. Suppose that a friend called you last night to get your advice in his/her search for a product.
Would you have recommended him/her to buy the product?

References
Alba, J., Lynch, J., Weitz, B., Janiszewski, C., Lutz, R., Sawyer, A. & Wood, S. (1997)
Interactive home shopping: consumer, retailer, and manufacturer incentives to
participate in electronic marketplaces. Journal of Marketing, 61(3), pp. 38–53.
Alvarez, L. (2005) Narcissism guides mate selection: humans mate assortatively,
as revealed by facial resemblance, following an algorithm of self seeking like.
Evolutionary Psychology, 2, pp. 177–194.
Arndt, J. (1967) Role of product related conversation in the diffusion of a new
product. Journal of Marketing Research, 4(3), pp. 291–295.
Baron, R.M. & Kenny, D.A. (1986) The moderator-mediator variable distinction in
social psychological research: conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, pp. 1173–1182.
Baym, N.K. (1997) Interpreting Soap Operas and Creating Community: Inside an
Electronic Fan Culture, in Sara Kiesler (ed.) Culture of the Internet. Mahwah, NJ:
Lawrence-Erlbaum.
Bettman, J. (1973) Perceived risk and its components: a model and empirical test.
Journal of Marketing Research, 10, pp. 184–190.
Birdwell, A.E. (1968) A study of the influence of image congruence on consumer
choice. Journal of Business, 41, pp. 76–78.

704
 Online word of mouth and consumer purchase intentions

Blanton, H. (2001) Evaluating the self in the context of another: the three-selves
model of social comparison assimilation and contrast, in G.B. Moskowitz (ed.),
Cognitive Social Psychology: The Princeton Symposium on the Legacy and Future of Social
Cognition. Mahway, NJ: Lawrence Erlbaum Associates, pp. 75–87.
Bone, P. (1995) Word-of-mouth effects on short-term and long-term product
judgements. Journal of Business Research, 32(3), pp. 213–223.
Brewer, M. & Weber, J. (1994) Self-evaluation effects of interpersonal versus
intergroup social comparison. Journal of Personality and Social Psychology, 66,
pp. 268–275.
Brislin, R., Lonner, W. & Thorndike, R. (1973) Cross-cultural Research Methods. New
York: Wiley.
Brown, J., Broderick, A. & Lee, N. (2007) Word of mouth communication within
online communities: conceptualizing the online social network. Journal of
Interactive Marketing, 21(3), pp. 2–20.
Downloaded by [Tulane University] at 05:04 28 January 2015

Brucks, M. (1985) The effects of product class knowledge on information search

behaviour. Journal of Consumer Research, 12(June), pp. 1–16.
Byrne, D. (1971) The Attraction Paradigm. New York: Academic Press.
Chaplin, L.N. & John, D.R. (2005) The development of self-brand connections in
children and adolescents. Journal of Consumer Research, 32(June), pp. 119–129.
Chen, Q. & Wells, W. (1999) Attitude toward the site. Journal of Advertising Research,
39(5), pp. 27–38.
Chevalier, J. & Mayzlin, D. (2006) The effect of word of mouth on sales: online book
reviews. Journal of Marketing Research, 43(3), pp. 345–354.
Cho, C.H. & Cheon H.J. (2004) Why do people avoid internet advertising? Journal of
Advertising, 33(4) pp. 89–97.
Coyle, J. & Thorson, E. (2001) The effects of progressive levels of interactivity and
vividness in web marketing sites. Journal of Advertising, 30(3), pp. 65–77.
De Bruyn, A. & Lilien, G.L. (2008) A multi-stage model of word-of-mouth influence.
International Journal of Research in Marketing, 25(3), pp. 151–163.
Deighton, J. & Kornfeld, L. (2007) Digital interactivity: unanticipated consequences
for markets, marketing, and consumers. Working paper no. 08-017, Harvard
Business School.
Deighton, J., Romer, D. & McQueen, J. (1989) Using drama to persuade. Journal of
Consumer Research, 16(December), pp. 335–343.
Eagley, A. & Chaiken, S. (1993) The Psychology of Attitudes. Fort Worth (TX): Harcourt-
Brace.
Fallows, D. (2005) Tracking Online Life: How Women and Men Use the Internet. Pew
Internet and American Life Report. Available online at: [Link]
org/Reports/2005/[Link] (accessed 1
November 2010).
Festinger, L. (1954) A theory of social comparison processes. Human Relations, 7(2),
pp. 117–140.
Fishbein, M. & Ajzen, I. (1975) Belief, Attitude, Intention and Behaviour. Reading MA:
Addison-Wesley.

 705
 International Journal of Advertising, 2010, 29(5)

Galinsky, A., Maddux, W., Gilin, D. & White, J. (2008)� Why it pays to get inside the
head of your opponent: the differential effects of perspective taking and empathy
in negotiations. Psychological Science, 19(4), pp. 378–384.
Gottsch, J. (2001) Mutation, selection, and vertical transmission of theistic memes
in religious canons. Journal of Memetics – Evolutionary Models of Information
Transmission, 5. Available online at: [Link]/2001/vol5/gottsch_jd.html
(accessed 1 November 2001)
Granovetter, M.S. (1973). The strength of weak ties. The American Journal of Sociology,
78(6), pp. 1360–1380.
�Gürhan-Canli, Z. & Maheswaran, D. (2000) Comparative advertising in the global
marketplace: the effects of cultural orientation on communication. Working Papers
series no. 328, Stephen M. Ross Business School, William Davidson Institute,
University of Michigan.
Harman, D. (1967) A single factor test of common method variance. Journal of
Downloaded by [Tulane University] at 05:04 28 January 2015

Psychology, 35, 359–378.

Hass, R. (1981) Effects of source characteristics on cognitive responses and
persuasion, in R.E. Petty, T.M. Ostrom & T.C. Brock (eds), Cognitive Responses in
Persuasion. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 141–172.
Haythornthwaite, C. (1999) A Social Network Theory of Tie Strength and Media Use:
A Framework for Evaluating Multi-level Impacts of New Media. Technical Report
UIUCLIS—2002/1+DKRC, Graduate School of Library and Information Science.
Champaign, IL: University of Illinois at Urbana-Champaign.
Heider, F. (1958) The Psychology of Interpersonal Relations. New York: John Wiley.
Hoffman, D. & Novak, T. (1996) Marketing in hypermedia computer-mediated
environments: conceptual foundations. Journal of Marketing, 60, pp.50–68.
Katz, E., & Lazarsfeld, P.F. (1955) Personal influence: the part played by people in theflow
of mass communication. Glencoe, IL: Free Press.
Keller, E. & Berry, J. (2006). Word-of-mouth: the real action is offline. Advertising Age,
77, p. 20.
Kelley, H. (1971) Attribution in Social Interaction, Morristown, NJ: General Learning
Press.
Kelley, H. (1973) Processes of causal attribution. American Psychologist, 28(February),
pp. 107–128.
Kelman, H. (1961) Processes of opinion change. Public Opinion Quarterly, 25,
pp. 57–78.
Khaled, R., Barr, P., Noble, J. and Biddle, R. (2006) Investigating social software
as persuasive technology, in Proceedings of the First International Conference on
Persuasive Technology for Human Well-Being (Persuasive06), Amsterdam.
Kim, T. & Biocca, F. (1997) Telepresence via television: two dimensions of
telepresence may have different connections to memory and persuasion. Journal of
Computer-Mediated Communication, 3(2), Available online at: http:// [Link]/
jcmc/vol3/issue2/[Link] (accessed 1 November 2010).
Kliatchko, J. (2008) Revisiting the IMC construct: a revised definition and four
pillars. International Journal of Advertising, 27(1), pp. 133–160.
Knapp, M. & Daly, J. (eds) (2002), Handbook of Interpersonal Communication (3rd ed.).
Thousand Oaks, CA: Sage.

706
 Online word of mouth and consumer purchase intentions

Kozinets, R. (1999) E-tribalised marketing? The strategic implications of virtual

communities of consumption. European Management Journal, 17(3), pp. 252–264.
Lau, G. & Ng, S. (2001) Individual and situational factors influencing negative word-
of-mouth behaviour. Canadian Journal of Administrative Sciences, 18, pp. 163–178.
Lee, S. & Lumpkin, J.R. (1992) Differences in attitude towards TV advertising: VCR
usage as a moderator. International Journal of Advertising, 11(4), pp. 333–342.
Lee, M. & Youn, S. (2009) Electronic word of mouth (eWOM): how eWOM platforms
influence consumer product judgement. International Journal of Advertising, 28(3),
pp. 473–499.
Money, R., Gilly, M. & Graham, J. (1998) Explorations of national culture and word-
of-mouth referral behaviour in the purchase of industrial services in the United
States and Japan. Journal of Marketing, 62, pp. 76–87.
Nail, J. (2005) What’s the buzz on word-of-mouth marketing? Social computing and
consumer control put momentum into viral marketing. Available online at: http://
Downloaded by [Tulane University] at 05:04 28 January 2015

[Link]/rb/Research/whats_buzz_on_word-of-mouth_marketing/q/
id/36916/t/2 (accessed 30 April, 2010).
Peterson, R. & Merino, M. (2003) Consumer information search behaviour on the
internet. Psychology & Marketing, 20(2), pp. 99–121.
Ploderer, B., Howard, S. & Thomas, P. (2008) Being online, living offline: the
influence of social ties over the appropriation of social network sites, in Proceedings
of the ACM 2008 Conference on CSCW. San Diego, CA.
Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y. & Podsakoff, N.P. (2003) Common
method biases in behavioral research: a critical review of the literature and
recommended remedies. Journal of Applied Psychology, 88(5): pp. 879–903.
Point Topic (2008) Worldwide broadband penetration by household. Available online
at: [Link] (accessed on 30 April 2010).
Putrevu, S. & Lord, K. (1994) Comparative and noncomparative advertising:
attitudinal effects under cognitive and affective involvement conditions. Journal of
Advertising, 23(June), pp. 77–90.
Roehm, M. (2001) Instrumental vs vocal versions of popular music in advertising.
Journal of Advertising Research, 41(3), pp. 49–58.
Rogers, E.M. (1983) Diffusion of Innovations (3rd edn). New York: Free Press.
Romani, S. (2006) Price misleading advertising: effects on trustworthiness toward
the source of information and willingness to buy. Journal of Product & Brand
Management, 15(2/3), pp. 130–138.
Sheppard, B.H., Hartwick, J. & Warshaw, P. (1988) The theory of reasoned action: a
meta-analysis of past research with recommendations for modifications and future
research. Journal of Consumer Research, 15, pp. 325–343.
Srinivasan, N. & Ratchford, B. (1991). An empirical test of a model of external search
for automobiles. Journal of Consumer Research, 18(2), 233–242.
Sudman, S. (1980) Improving the quality of shopping center sampling. Journal of
Marketing Research, 17(November), pp. 423–431.
Tabachnick, B. & Fidell, L. (2001) Using Multivariate Statistics (4th edition). Boston
MA: Allyn and Bacon.

 707
 International Journal of Advertising, 2010, 29(5)

Tamanin, T., Ancona, C. & Rodrigues-Netto, N. (2002) Translation, Validation and

Cross-Cultural Adaptation into Portuguese Language of the ‘King’s Health Questionnaire.
Heidelberg: International Continence Society Annual Conference.
Trusov, M., Bucklin, R.E. & Pauwels K. (2009) Effects of word-of-mouth versus
traditional marketing: findings from an internet social networking site. Journal of
Marketing, 73(September), pp. 90–102.
Wolfinbarger, M. & Gilly, M. (1993) The encoding and decoding of gift symbolism.
Working paper, University of California at Irvine.
Woodside, A. & Delozier, W. (1976)� Effects of word of mouth advertising on
consumer risk taking. Journal of Advertising, 5(4), pp. 12–19.
World Bank – World Development Indicators (2009). Available online at: [Link]
[Link]/[Link]?q=poorest&d=WDI&f=Indicator_Code%[Link].P2
(accessed on 30 April 2010).
Downloaded by [Tulane University] at 05:04 28 January 2015

About the authors

Gerard Prendergast is Professor of Marketing at Hong Kong Baptist
University. His research interests are marketing communication and mar-
keting ethics.
David Ko is Lecturer in Marketing at Hong Kong Baptist University. His
research interests are entrepreneurship and Chinese family businesses.
Siu Yin V. Yuen is a marketing executive with Classroom Publications
Limited. Ms Yuen has interests in marketing communication and social
media.
Address correspondence to: Gerard P. Prendergast, School of Business,
Hong Kong Baptist University, Kowloon Tong, HONG KONG.
Email: Gerard@[Link]

708