BACKGROUND
Electronic Word-of-Mouth (eWOM) has become an important communication method and has
received considerable interest. Social networking sites enable consumers to receive product reviews and opinions held by members of
their social networks and the degree to which the consumer want to be influenced varies. Some people prefer to
rely on their own judgment whereas others tend to accept friends’ opinions if they value fitting into the
group. The degree to which the consumer want to be influenced reflects the concept of self-construal.
Through this study, we aim to better understand how consumers’ difference in self-construal may
influence how they perceive eWOM from members of their social networks and how it affect their online purchase decisions.
HYPOTHESIS
H1: Self-construal moderates the relationships between direction of eWOM and user’s intention to
click (H1a), attitude toward the ad (H1b), product attitude (H1c) and purchase intention (H1d).
Specifically, the effects of eWOM direction on intention to click, attitude toward the ad, product attitude and purchase intention were stronger for interdependent individuals than for independent
individuals.
H2: Self-construal moderates the relationships between sender’s tie strength and receiver’s intention
to click (H2a), attitude toward the ad (H2b), product attitude (H2c) and purchase intention (H2d).
Specifically, the effects of tie strength on intention to click, attitude toward the ad, product attitude and
purchase intention were stronger for interdependent individuals than for independent individuals with
strong tie; whereas, the effects were stronger for independent individuals than for interdependent
individuals with weak tie.
EXPERIMENTAL DESIGN
Study 1: Effects of Direction of eWOM and Self-construal
To test the proposed hypotheses, a laboratory experiment was conducted. The independent variables
were self-construal (independent vs. interdependent) and direction of eWOM (positive vs. negative).
Both independent variables were between-subjects variables; hence, there were in total of 4 conditions
in study 1. Participants were randomly assigned to one of four conditions.
In order to manipulate the self-construal of participants, two techniques introduced by Trafimow et al.
(1991) were adopted: the similar/different prime technique and the “anecdote” prime technique. In the similar/different prime technique, participants in the IndSC condition were asked to think about what
makes them different from their family members and friends; whereas, participants in the InterSC
condition were asked to think about what makes them similar to their family members and friends. In
the “anecdote” prime technique, participants read a story about a feudal ruler selecting a commander
to lead a detachment of soldiers to aid the ruler’s overlord in a war. For participants in the IndSC
condition, the story continued with the ruler choosing a talented but unrelated general to increase his
personal prestige; whereas, for participants in the InterSC condition, the story continued with the ruler
choosing a family member to increase the family’s prestige. Through using both prime techniques, we
tried to ensure that participant’s self-construal were successfully manipulated.
The direction of eWOM
was presented by showing participants either a positive or negative review about a product on a
webpage. The webpage was created by the experimenters based on real Facebook fan pages to
maintain a natural layout.
Participants were instructed to seek product information on the webpage
and evaluate their intention to click, attitude toward the ad, product attitude, and purchase intention,
which were the dependent variables in this study.
Study 2: Effects of Tie Strength and Self-construal
The experimental design of study 2 was exactly the same as study 1 except that the independent
variables were self-construal (independent vs. interdependent) and tie strength of Facebook friend
(strong tie vs. weak tie). Participants received the same manipulation for self-construal, but got
recommendation from friends of different tie strength instead of positive or negative reviews.
To manipulate participants’ tie strength of Facebook friend, we directly used the friends listed on
participants’ Facebook. We found that friend list on Facebook is sorted by contact frequency. The
Facebook friends with more interactions with the user (strong tie) usually appear in the front of his/her
friend list. Therefore, participants assigned to the strong tie conditions were instructed to report the
name of the friend that appeared first on friend list and participants assigned to the weak tie condition
were instructed to write down the name of the friend that appeared last. The friend’s name would later
be used to indicate his/her attitude toward the advertisement of shampoo and running shoes.
EXPERIMENTAL MATERIALS
EMPIRICAL RESULTS
The empirical results showed that the effects of
eWOM direction on intention to click, attitude toward the product ad, product attitude, and purchase
intention were stronger for consumers with interdependent self-construal than for those with
independent self-construal.
Besides, the results also showed that the effects of tie strength between
reviewer and consumer on intention to click, attitude toward the product ad, product attitude and
purchase intention were stronger for interdependent consumers than for independent consumers when
receiving eWOM from their strong ties; whereas, the effects were stronger for independent consumers
than for interdependent consumers when receiving eWOM from their weak ties.
The full experiment data can be found here.
INSIGHTS
This study proposed self-construal as an important factor that could influence the relationship between
direction of eWOM, tie strength, and consumer purchase decisions. As suggested by the
findings, interdependent individuals reacts stronger than independent individuals to direction of
eWOM. Online advertisers could design special mechanism to temporarily activate consumer’s
interdependent self-construal in order to enhance the effectiveness of the spreading of eWOM favoring
their products. Moreover, the results show that tie strength has a positive eWOM effects on
interdependent consumers whereas negative effects on independent consumers. Thus, online
advertisers should customize their marketing strategy on SNS according to consumer’s self-construal.