SESSION 7: ELECTRONIC WORD OF MOUTH
You, Ya, Gautham G. Vadakkepatt, and Amit M. Joshi (2015). A Meta-Analysis of Electronic Word-of-
Mouth Elasticity. Journal of Marketing, 79, 19-39.
You, Vadakkepatt and Joshi (2015) conducted a meta-analysis on electronic word-of-mouth (eWOM).
Goal: Research the impact of electronic WoM on sales.
Elasticities Volume: Refers to the total number of eWoM messages
Elasticities Valence: Refers to the preference or ton of comments, typically expressed as
positive/neutral/negative.
Durable vs non-durable goods: durable goods (e.g., automobiles) are more complex and
have longer interpurchase intervals than nondurable goods (e.g., movies)
Product Trialability: Product trialability (low vs. high). Product trial plays a significant role in influencing
consumer beliefs and purchase intentions mostly because it provides a low risk option through which consumers
can more accurately evaluate product attributes and fit (Agarwal and Prasad 1997; Wright and Lynch
1995). In other words, product trials affect sales positively by minimizing product uncertainty (Bawa and
Shoemaker 2004).
Community based vs. Online Product Review Site: Brown and Reingen (1987) show that WOM is at its
most influential if there is a strong relationship between the sender and the recipient of the message.
Threat of public consumption on eWOM: Public consumption of products can weaken the effect of eWOM
metrics on sales through three routes. First, observing a product in use provides potential consumers with an
alternative route to gather information about the product, which in turn
reduces the benefits of information obtained through eWOM. Second, observability of product consumption
could result in mimicking behavior. Often called observational learning or social learning (Bikhchandani,
Hirshleifer, and Welch 1992; Chen, Wang, and Xie 2011),
this mimicking behavior could arise because consumers perceive other users’ final choice as
more reliable information than their own private information. Third, products help people not only create self-
identities (e.g., Belk 1988; Berger and Heath 2007; Escalas and Bettman 2003) but also infer identities of
others.
#Potential of privately consumed products: However, products that are privately
Consumed provide potential consumers with very limited opportunity to learn through observation. For these
products, eWOM can have a significant impact on sales by making information from private consumption
more readily available, therefore enabling consumers to
evaluate whether the product matches their own preferences.
a. What is a meta-analysis? What is the difference between a meta-analysis and a single research
study? And what is its significance for managers?
• Meta-analysis is to synthesize the results of multiple studies to validate the effect size through multiple
studies
• Meta analysis make claims about the distribution of effect sizes in a set of eligible studies
, • Significance for managers is the summary of results several studies. Understanding the
• average effect across studies and its variability, leading to more informed decisions about important
policy issues.
• Yij is the eWOM volume (or valence) elasticity from study j
• β0j is the intercept for the j study
• βj is the parameter estimate of the influencing factors for the j study
• eij is random error associated with i elasticity in study j
• γ0 is overall intercept
• μj is the study level
residual error term
Level 1 Impact of Contextual Data: The Level 1 equation describes the
You, Ya, Gautham G. Vadakkepatt, and Amit M. Joshi (2015). A Meta-Analysis of Electronic Word-of-
Mouth Elasticity. Journal of Marketing, 79, 19-39.
You, Vadakkepatt and Joshi (2015) conducted a meta-analysis on electronic word-of-mouth (eWOM).
Goal: Research the impact of electronic WoM on sales.
Elasticities Volume: Refers to the total number of eWoM messages
Elasticities Valence: Refers to the preference or ton of comments, typically expressed as
positive/neutral/negative.
Durable vs non-durable goods: durable goods (e.g., automobiles) are more complex and
have longer interpurchase intervals than nondurable goods (e.g., movies)
Product Trialability: Product trialability (low vs. high). Product trial plays a significant role in influencing
consumer beliefs and purchase intentions mostly because it provides a low risk option through which consumers
can more accurately evaluate product attributes and fit (Agarwal and Prasad 1997; Wright and Lynch
1995). In other words, product trials affect sales positively by minimizing product uncertainty (Bawa and
Shoemaker 2004).
Community based vs. Online Product Review Site: Brown and Reingen (1987) show that WOM is at its
most influential if there is a strong relationship between the sender and the recipient of the message.
Threat of public consumption on eWOM: Public consumption of products can weaken the effect of eWOM
metrics on sales through three routes. First, observing a product in use provides potential consumers with an
alternative route to gather information about the product, which in turn
reduces the benefits of information obtained through eWOM. Second, observability of product consumption
could result in mimicking behavior. Often called observational learning or social learning (Bikhchandani,
Hirshleifer, and Welch 1992; Chen, Wang, and Xie 2011),
this mimicking behavior could arise because consumers perceive other users’ final choice as
more reliable information than their own private information. Third, products help people not only create self-
identities (e.g., Belk 1988; Berger and Heath 2007; Escalas and Bettman 2003) but also infer identities of
others.
#Potential of privately consumed products: However, products that are privately
Consumed provide potential consumers with very limited opportunity to learn through observation. For these
products, eWOM can have a significant impact on sales by making information from private consumption
more readily available, therefore enabling consumers to
evaluate whether the product matches their own preferences.
a. What is a meta-analysis? What is the difference between a meta-analysis and a single research
study? And what is its significance for managers?
• Meta-analysis is to synthesize the results of multiple studies to validate the effect size through multiple
studies
• Meta analysis make claims about the distribution of effect sizes in a set of eligible studies
, • Significance for managers is the summary of results several studies. Understanding the
• average effect across studies and its variability, leading to more informed decisions about important
policy issues.
• Yij is the eWOM volume (or valence) elasticity from study j
• β0j is the intercept for the j study
• βj is the parameter estimate of the influencing factors for the j study
• eij is random error associated with i elasticity in study j
• γ0 is overall intercept
• μj is the study level
residual error term
Level 1 Impact of Contextual Data: The Level 1 equation describes the
