The Factors Affecting Writing Reviews in Hotel Websites

Abstract

We aim to understand the people's motivation for writing the reviews in order to better explain drivers of WOM activities in the online channel for the successful implementation of the hotel marketing. For this purpose, we choose one of the popular tourist destinations: Paris. We collect data for empirical analysis from one of the biggest online hotel reservation website. Our analysis reveals that higher rating and lower price increases the propensity to write reviews. However, while the extreme rating and star score and star rating does not have any effect on the propensity to write a review. We have also found the evidence for the negative effect of larger room size on the propensity to write a review. These results imply that satisfaction rather than dissatisfaction increases people's motivation more for writing reviews in hotel websites. We briefly discuss the managerial implications of results. (C) 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the 8th International Strategic Management Conference

Full text

Procedia - Social and Behavioral Sciences 58 ( 2012 ) 980 – 986
1877-0428 © 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the 8th International Strategic Management Conference
doi: 10.1016/j.sbspro.2012.09.1077
8th International Strategic Management Conference
The Factors Affecting Writing Reviews in Hotel Websites
a , Asunur Cezara
TOBB University of Economics and Technology, Ankara, 06530, Turkey
Abstract
We aim on for writing the reviews in order to better explain drivers of WOM activities in the
online channel for the successful implementation of the hotel marketing. For this purpose, we choose one of the popular touri st
destinations: Paris. We collect data for empirical analysis from one of the biggest online hotel reservation website. Our analysis
reveals that higher rating and lower price increases the propensity to write reviews. However, while the extreme rating and star
score and star rating does not have any effect on the propensity to write a review. We have also found the evidence for the negative
effect of larger room size on the propensity to write a review. These results imply that satisfaction rather than dissatisfaction
increases people more for writing reviews in hotel websites. We briefly discuss the managerial implications of our
results.
2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of The 8th International Strategic
Management Conference
Keywords: online hotel reservation; online word of mouth; online reviews
1. Introduction
Customers increasingly use e-commerce sites for purchasing many products and services and internet becomes
preferred sales channel for many industries. Travel industry is one of the first and successful industries to use Internet
for this purpose and studies show that online travel sales keep growing. With a 16% share, hotel accommodation is the
second largest sales item after air travel among online travel sales and revenue generated through online hotel booking
increases (Marcussen 2008).
It has been reported in many cases that WoM is very effective marketing tool. In hotel industry, most customers
choose hotels based on recommendation of a friend and industrial report shows that word of mouth is one of the
important factors in hotel selection decision (Barsky & Nash 2008). Word of mouth of the product has long been
discussed as the free marketing tools of the products. This medium of marketing is more convincing than traditional
marketing tools as people experiencing the product information has no incentive to spread wrong information about
the product. It has been found that WoM marketing was seven times as effective as newspapers and magazines, four
times as effective as personal selling, and twice as effective as radio advertising in influencing people choices(Brown
and Peter 1997).
Corresponding author. Tel. + 90-312-292-4218 fax. +90-312-292-4104
Email address: hogut@etu.edu.tr
Available online at www.sciencedirect.com
© 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the 8th International Strategic
Management Conference Open access under CC BY-NC-ND license.
Open access under CC BY-NC-ND license.

981
Hulisi Öğü t and Asunur Cezar / Procedia - Social and Behavioral Sciences 58 ( 2012 ) 980 – 986
One of the forms of word of mouth in the cyber world is online reviews. Recent studies show that travel reviews are
increasingly becoming an important factor in hotel selection by travelers. As indicated by Milan (2007), millions of
travelers log on daily to Travel websites like Tripadvisor.com and experience web content through hotel generated
photos, written text and hotel reviews by past customers. Milan (2007) indicates that 84% of people visiting a Travel
website hosting consumer generated content have their hotel choices affected by what they see and online hotel
shoppers find reviews and hotel and room photos much more convincing than other features of hotels.
Although there are some similarities between two forms of online reviews and WoM, there are significant differences
between online review and WoM.
verbal communication, online reviews can reach all the people having access to the internet. Furthermore, the effect of
online reviews does not fade away with time and distance and it can be more detailed and durable as it reflects the
opinions of more than one person in written form. Another difference is that while it is very difficult to measure the
effectiveness of WoM, the metrics related to online reviews can easily be established (Bhatnagar & Ghose 2004, Duan
et al. 2008). Thus, it is important to understand drivers of WOM activities in the online channel for the successful
reviews.
The remainder of the paper is organized as follows. First, we present relevant literature. Then, we discuss the
hypothesis and their background. Next, we present our data and empirical results. Last section concludes the paper.
2. Literature Review
Many recent studies analyze the impact of online reviews on product sales by considering the review volume and
review valence measured as customer rating or positive/negative user ratings. (Sen and Lerman 2007; Senecal and
Nantel 2004). These studies in the movie and online book industry show mixed results. Some of the studies show that
both the volume and review valence affect future sales. Among the earlier studies, Chevaliear and Mayzlin (2006)
examine the effect of consumer reviews on relative sales of books at Amazon.com and BarnesandNoble.com and they
find that an improvement in a book's customer rating in the website causes an increase in sales at that site. However,
Chen et al. (2004) find the review valence does not affect future sales using the same data set from Amazon.com. In
the movie industry, Liu (2006) has found that review volume is the driver of future box office sales. By separating the
effect of online review as both the originator and the result of sales, Duan et al. (2008) find that both a sales and
review valence significantly leads to higher review volume and higher review volume in turn results in higher sales.
Few rece By using
consideration set theory, Vermeulen and Seeger (2009) conduct an experimental study to analyze the effect of internet
customer reviews to consumer decision making. Consideration set theory (Roberts & Lattin, 1991) states that
customer decision making is multi-staged and at each subsequent stage (awareness/consideration/choice stage
respectively) a customer narrows down available alternatives until she makes her final decision. They consider review
valence (positive versus negative), hotel familiarity and the reviewer expertise as the construct of their study. They
find that both negative and positive reviews enhance consumer awareness for hotels. Furthermore, positive reviews
this result is especially valid for lesser-known hotels as the exposure to reviews have limited effect for well known
hotels. They also show that the impact of reviewer expertise is positive, albeit minor. Dickinger and Mazanec (2008)
show that recommendations of friends and online reviews are the most two important drivers of online hotel booking.
Ogut and Tas (2011) investigate the impact of star rating and customer rating on hotel room sales and prices and find
that higher customer rating increases the online sales. However, higher star do not increases the online hotel room
sales. They also show that there is a positive relationship between customer ratings and prices. Ghose et al (2011)
illustrates how social media can be mined in order to generate a new ranking system in product search engines and
propose a hotel ranking system that recommends products that provide on av
money.
Our paper is mostly related to Dellarocas and Narayan (2006) and Dellarocas et al. (2010). Dellarocas and Narayan
propensity to engage in post-purchase online word-of-mouth in the movie industry. They

982 Hulisi Öğü t and Asunur Cezar / Procedia - Social and Behavioral Sciences 58 ( 2012 ) 980 – 986
Dellarocas et al. (2010) find that people are more likely to contribute for movies that are less and more successful.
Our paper is different from these papers from the following respects. First, we investigate one of the most successful
applications of e-commerce which is hotel industry. Second, the sales of rooms and writing the reviews take place at
the same website. Thus, people can write their ex-ante and ex-
investigate the effect of price and hotel specific characteristics such as room size. Fourth, the quality of service is
measured in two different ways: star and customer rating.
3. Hypothesis
One of the determinants of satisfaction and dissatisfaction is the quality of the service. In our context, the quality of
the hotels can be measured in two different ways. These are traditional way of star rating and digitized way of online
customer ratings. National rating agencies have been established by local authorities to evaluate hotels on the basis of
their intrinsic qualities and rank them according to a five or four star scale. As a quality measure, online customer
ratings complement star feature by considering subjective quality dimensions such as how nice hotel staff is, comfort
and cleanliness of the hotel room, facilities/services offered to customers ,value provided versus the price of the hotel
and location. Thus, we used both star and customer rating as a proxy for satisfaction. For this reason, we expect that
Hypothesis 1: The motivation for writing review will be high for hotels that have either low or high ratings.
Hypothesis 2: The motivation for writing review will be high for hotels that have either low or high stars.
Online reviews are considered as the counterpart of the word of mouth (WoM) in the cyber world as they share many
similarities. However, we believe that there is subtle difference between online reviews and traditional WoM since it is
possible to observe the quality of the product prior to purchase in online reviews. Thus, they will have lower (higher)
expectation if the customer rating of the product is low (high). We expect the same affect for the star rating as well. As
the positive and negative disconfirmation will be higher for high customer and star rating, we expect the propensity to
write the review will be high for hotels that have high ratings. Dellocras et al.(2010) find the similar results in movie
industry and stated that there is J type relationship between customer rating and the propensity to write review. For
this reason, we expect that
Hypothesis 3: The motivation for writing review will be high for hotels that have high ratings.
Hypothesis 4: The motivation for writing review will be high for hotels that have high stars.
The relationship between the word of mouth communication and price has not been thoroughly investigated in the
literature. As the price of the hotel decreases, people will become more satisfied or dissatisfied and they are more
likely to write reviews on web page. As the price of the hotel increases, the people are less likely to write review as the
quality of the hotel increases. For this reason, we expect negative relationship between price and the propensity to
review. Thus, we can hypothesize that
Hypothesis 5: For the given quality, the motivation for writing review will be high for hotels that have lower price.
Model
We collected the data from one of the biggest online hotel web booking sites: www.booking.com. We choose Paris
hotel as the source of our data sets since it is one of the most popular tourist destinations in the world. The dependent
variable, the propensity to write a review is the likelihood that customer can write a review after staying in the hotel.
Number of Review at Specific Time Interval
Likelihood of Writing Review= Number of Booking at Specific Time Interval
Since number of booking at specific time interval can not be greater than number of booking at that time, this variable
is restricted to be between 0 and 1. We can use the logistic function to model the relationship between dependent

983
Hulisi Öğü t and Asunur Cezar / Procedia - Social and Behavioral Sciences 58 ( 2012 ) 980 – 986
variable(likelihood of writing review) and independent variables as in Dellarocas and Narayan (2006). Then, logit
transformation of this relationship is given as
ln 1
yXu
y
One of the problems with this approach is that if y variable is 0 and 1, then logistic transformation of odds will be
infinity. The other problem is that this function is not reversible in order to estimate the equation in 1 (Papke and
Wooldridge (1996)). Papke and Wooldridge (1996) use a quasi-likelihood regression model for continuously
ln 1 1 ln
ii i i
lyGx y Gx
where
1
i
i
fx
ifx
e
Gx e
and
2
13
2
45 6
i0 i2 i i
N-1
ii ik,D i
k=1
f x = + ln(Price) + Customer Rating + Customer Rating Mean Rating
Star+ Star Mean Star + ln RegionDummy
(Size)
In the equation above
i
y
is the likelihood of writing review. In order to test the effect of extreme satisfaction and
dissatisfaction, we added the square root of difference between quality variable and the mean value of it. In this way,
U type relationship between quality variable (i.e. star and customer rating) and independent variable is captured. We
predicted the parameters of independent variables using Maximum likelihood approach. Size and region dummy is
added as explanatory variable for control purposes. We take the logarithmic transformation of price and size of the
affect as the standard deviations of these variables are large compared to other variables.
4. Data
The data are from one of the biggest online hotel web booking sites: www.booking.com. After customers enter
information on location, check-in and check-out dates to the website, available hotels are listed and it is possible to
obtain information on hotel star, type and price of hotel rooms and average hotel customer rating in this listing. If
facilities, hotel policies, number of hotel rooms and detailed guest reviews. From these websites, we gathered the
information on hotel star, region of the hotel in the city, room price per night, average customer rating, number of
hotel rooms, number of customer reviews and number of hotel reservation. Data are collected in 2011 and the average
values of price and customer rating are computed for the final data set. We used the price of a standard double room as
the room price since some hotels do not have the price information for single rooms.
Individual customer rating, is calculated in the following way. First, customers rate hotel quality in terms of hotel
staff, services/facilities, cleanness of hotel room, comfort, value for money and location. The score in these
dimensions can be poor, fair, good or excellent and counts for 1, 2, 3 and 4 points respectively. All these points are
added and divided by 2.4 for the final individual score
obtained based on classification of booking.com. Table I displays the descriptive statistics of the variables used in our
study. Table 2 displays the correlation between the variables.
Table 1. Descriptive Statistics
Variable
Observation
Mean
Standard
Deviation
Minimum
Maximum
Likelihood to Review
810
0.187503
0.067889
0.0384615
0.375
Price
810
136.3558
69.77797
36.59096
668.25
Star
810
2.887654
0.767685
1
5
Rating
810
7.509877
1
4.7
9.4
(Rating-Mean Rating)2
810
0.528321
1
0.0000256
7.868379

984 Hulisi Öğü t and Asunur Cezar / Procedia - Social and Behavioral Sciences 58 ( 2012 ) 980 – 986
Table 2. Correlation between Variables
5. Estimation Results
Before testing the hypotheses, several econometric specifications are checked. We first identify outliers of the log
transformation of price and sales per room variable in the data set using the Grubbs methodology. Grubbs' test
(Grubbs (1969) and Stefansky (1972)) is used to detect one outlier at a time in the univariate data set. The outlier is
removed from the data set and the test is iterated until no outliers are detected. This test is also known as the maximum
normed residual test. The multicollinearity of explanatory variables is investigated using the variance inflatio n factor
(VIF). VIF values range from 1.27 to 3.10. Since VIF values are smaller than the recommended value of 10 (Belsley,
Kuh, & Welsch, 1980), we conclude that there is no multicollinearity problem in our regression analysis.
Table 3. Regression Results
The significances of the coefficients of star and customer rating test our third (H3) and fourth (H4) hypotheses. For all
of the regression specifications with different sets of explanatory variables, the coefficient of the customer rating is
significant at 1% and the sign of the coefficient is positive. Thus, the regression results validate our third hypothesis
(H3) by showing that an increase in the customer rating of a hotel leads to a significant increase in the propensity to
write a review. Even though we are able to validate H3, we could not find supportive evidence from the regression
result for the fourth hypothesis. The regression results show that the coefficient of star rating is insignificant even at
the 10% significance level. Thus, the hypothesis that higher star rating result in higher propensity to write a review is
not supported. In order to test U type relationship between the propensity to review and quality variables proxied by
star and customer rating, we added the square of the difference between the value of these variables and their means as
explanatory variables. The significances of these variable test our first (H1) and second (H2) hypotheses. However, the
coefficients of these variables are not significant .Thus, we could not validate the hypotheses that extreme higher or
(Star-Mean Star)2
810
0.588685
0.817987
0.0145996
4.497916
Room Size
810
57.80864
77.14124
9
1025
Likelihood to
Review
Price
Star
(Star
-Mean Star)2
Rating
(Rating-
Mean
Rating)2
Room
Size
Likelihood to Review
1
Price
-0.097
1
Sstar
-0.063
0.744
1
(Star-Mean Star)2
-0.114
0.360
0.060
1
Rating
0.079
0.583
0.481
-0.0111
1
(Rating-Mean Rating)2
-0.044
0.048
-0.09
0.2782
-0.319
1
Room Size
-0.072
0.220
0.277
0.1466
0.104
-0.087
1
Coefficent
Standart Deviation
z-statistics
P value
ln(Price)
-0,18772
0,083318
-2,25
0,024
ln(Room Size)
-0,08308
0,028513
-2,91
0,004
Rating
0,134557
0,032072
4,2
0
(Rating-Mean Rating)2
0,006063
0,022925
0,26
0,791
Star
0,017593
0,03657
0,48
0,63
(Star-Mean Star)2
-0,02953
0,02134
-1,38
0,166

985
Hulisi Öğü t and Asunur Cezar / Procedia - Social and Behavioral Sciences 58 ( 2012 ) 980 – 986
test our fifth hypothesis that lower price results in higher the propensity to review. Our results provide support for this
hypothesis.
6. Conclusion
activities in the online channel for the successful implementation of the hotel marketing. For this purpose, we choose
one of the popular tourist destinations: Paris. We collect data for empirical analysis from one of the biggest online
hotel reservation website. Our analysis reveal that higher rating and lower price increases the propensity to write while
the extreme high or low rating and star score, star rating does not have any effect on the propensity to write a review.
We have also found the evidence for the negative effect of larger room size on the propensity to write a review. These
results imply that satisfaction rather than dissatisfaction affects people more for writing reviews in hotel websites.
In our paper, we test whether online customer review is subject to self selection bias meaning that customers having
extreme satisfaction or dissatisfaction are more likely to post reviews compared to other past customers. Our results
did not provide supportive evidence for this argument. We also show how to increase posting of online comment. This
is especially more important for the hotels having fewer reviews as limited reviews are more likely to represent the
customer rating. Furthermore, higher online number of reviews may increases sales as it
implies that people frequently prefer this hotel.
References
Barsky, J. & Nash, L. (2008). Word of mouth should be top-of-mind for hotel operators.
http://www.marketmetrix.com/en/default.aspx?p=research1&s=research,Accessed 23.03.2012.
Belsley, D., Kuh, E., & Welsch, R. E. (1980). Regression Diagnostics: Identifying Influential Data and Sources of
Collinearity. New York: John Wiley.
Bhatnagar, A. & Ghose, S.(2004). Online Information Search Termination Patterns Across Product Categories and
Consumer Demographics, Journal of Retailing, 80(3), 221-228.
Brown, J. J. and Peter H. R. (1987).Social Ties and Word-of-Mouth Referral Behavior. Journal of Consumer
Research,14 (3), 350-362.
Chen, P., Shin-yi W., & Jungsun Y. (2004). The Impact of Online Recommendations and Consumer Feedback on
Sales. in Proceedings of the International Conference on Information Systems, 711 724.
Chevalier, J. A., & Mayzlin, D. (2006). The effect of word of mouth on sales: online book reviews. Journal of
Marketing Research, 43(3), 345 354.
Dellarocas, C.N., Neveen, F. A., & Zhang X. (2007), Exploring the Value of Online Product Reviews in Forecasting
Sales: The Case of Motion Pictures. Journal of Interactive Marketing, 21 (4), 23 45.
Dellarocas C. and Narayan R. (2006). A Statistical Measure of a Propensity to Engage in Post-Purchase
Online Word-of-Mouth. Statistical Science, 21 (2), 277-285.
Dellarocas C., Gao G. and Narayan R. (2010). Are consumers more likely to contribute online reviews for hit products
or niche products? Journal of Management Information Systems, 27 (2), 127-157
Dickinger, A. & Mazanec, J. (2008): Consume
-254.

986 Hulisi Öğü t and Asunur Cezar / Procedia - Social and Behavioral Sciences 58 ( 2012 ) 980 – 986
Duan, W., Gu, B., & Whinston A.B. (2008). Do online reviews matter? An empirical investigation of panel data.
Decision Support Systems, 45(4), 1007-1016.
Ghose, A., Ipeirotis, P.G., & Li, B.(2012). Designing Ranking Systems for Hotels on Travel Search Engines by
Mining User-Generated and Crowd-Sourced Content, Marketing Science, forthcoming
Grubbs F. (1969). Procedures for Detecting Outlying Observations in Samples, Technometrics, 11(1): 1-21.
Marcussen C. (2008), Trends in European Internet Distribution - of Travel and Tourism Services. <
http://www.crt.dk/UK/staff/chm/trends.htm >. Accessed 23.03.2012.
Milan, R. (2007), Travel reviews - consumers are changing your brand and reputation online.
<http://www.travelindustrywire.com/article29359>. Accessed 23.03.2012.
Liu, Y. (2006). Word-of-mouth for movies: Its dynamics and impact on box office receipts?, Journal of Marketing,
70(3), 74 89.
Papke L, Wooldridge J (1996). Econometric methods for fractional response variables with an application to 401(K)
plan participation rates. Journal of Applied Econometrics, 11, 619 32
Roberts, J. H. & Lattin, J. M. (1991). Development and Testing of a Model of Consideration Set Composition. Journal
of Marketing Research, 28(4), 429-440.
). The Influence of Internet Customer Reviews on the Online Sales and Prices in Hotel
Industry. Service Industries Journal, 32(2), 197-214.
Sen, S., & Lerman, D. (2007). Why are you telling me this? An examination into negative consumer reviews on the
Web. Journal of Interactive Marketing, 21(4), 76 94.
Journal of Retailing, 80, 159 169.
Stefansky W. (1972). Rejecting Outliers in Factorial Designs, Technometrics, 14, 469-479.
Tripadvisor.com. (2006). Fact sheet. <http://www.tripadvisor.com/pages/factsheet.html>. Accessed 23.03.2013.
Vermeulen, I. E. & Seegers D.(2009). Tried and tested: The impact of online hotel reviews on consumer consideration.
Tourism Management, 30, 23-127.