Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:

The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

Research on the Influential Factors of Customer Satisfaction and Post-

Purchase Behavior for Hotels: The Multilayer Perceptrons Neural
Network Approach and Logistic Regression Analysis
Han-Chen Huang
No. 168, Hsueh-Fu Rd., Tanwen Village, Chaochiao Township, Miaoli County, 36143 Taiwan
Department of Leisure Management, Yu Da University, E-mail: hchuang@ydu.edu.tw

Abstract
The majority of previous studies have used logistic regression analysis(LRA) as an analysis tool to
evaluate the impact of service attributes on customer satisfaction and post-purchase behavior.
However, LRA assumes that a linear relation exists between the input and output variables, and that
variables are normally distributed, which places significant limitations on the analysis. This study uses
multilayer perceptrons neural networks(MLPs) to overcome the limitations of LRA. The results show
that among the four hotel service attributes, “personnel services” has the greatest impact on customer
satisfaction and repurchase or recommending the service to others, whereas “business and travel
services” has the lowest impact. MLPs is more accurate than LRA. MLPs achieved an accuracy rate of
93% for predicting customer satisfaction, and 90% for predicting repurchase or recommending the
service to others. Although LRA has an accuracy rate of 87% for predicting customer satisfaction, it
only scores 23.77% for predicting “unsatisfied customers.” MLPs has an accuracy rate of 69.23% for
predicting this category. LRA has an accuracy rate of 80% for predicting repurchase or recommending
the service to others, but only has 9.52% accuracy in the “no intention to repurchase or recommend
the service to others” category. By contrast, MLPs has an accuracy of 71.43% for this category,
indicating that MLPs can predict repurchase or recommending the service to others more effectively.

Keywords: Service Attribute, Customer Satisfaction, Neural Network, Logistic Regression Analysis

1. Introduction

Studies on customer satisfaction are beneficial[1,2]. Numerous studies have shown that customer
satisfaction is a key factor that directly impacts customer intention to repurchase or recommend to
others[3]. Customer repurchase or word-of-mouth behavior increases corporate profit and market share.
After receiving hotel services (dining, room, transportation assistance, business assistance, and travel
assistance), customers evaluate the quality of each service. The importance of each hotel service
attribute to customer satisfaction or repurchase behavior varies. This study investigates the service
attributes that affect customer satisfaction and the degree of impact each service attribute has on
customer satisfaction. Customer opinions are obtained through questionnaires and the data is analyzed
using artificial neural network analysis. The analysis result is compared with logistic regression
analysis. The results enable hotel management to understand future marketing directions and the
service attributes that customers prefer and that may require improvement.

2. Literature review

2.1 Customer satisfaction

Customer satisfaction has numerous definitions. Anton[4] stated that customer satisfaction is a
mental state achieved when a product or service meets or exceeds the requirements and expectations of
customers. This mental state leads to customer repurchase and loyalty to a product or service.
Zellhaml[5] believed that customer satisfaction is customer evaluation on whether a product or service
meets their requirements and expectations. If the product or service fails to meet customer
requirements and expectations, they will feel unsatisfied. Oh[6] stated that customer satisfaction is a
complex human behavior that involves cognitive, emotional, psychological, and physiological
processes. Matzler[7] proposed a definition that “customer satisfaction is a contrast of customer

Advances in information Sciences and Service Sciences(AISS) 442

Volume4, Number10, June 2012
doi: 10.4156/AISS.vol4.issue10.52
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

expectation and actual service performance.” For evaluating purposes, consumers typically compare
expected service quality with actual feeling. For hotels, a more logical definition of customer
satisfaction may be an overall subjective evaluation of every face-to-face service and experience
provided by the hotel following a customer purchase. Therefore, measuring customer satisfaction is a
crucial task for businesses. The assessment results can motivate the service industry to improve their
behavior and enhance their service quality. Customer satisfaction assessment is an indispensable
measure for obtaining competitive advantage in an industry. Enge1[8] shows that customer satisfaction
levels may be affected by customer background, personality, and the stimuli influencing customers
when they receive service. Jaksa[9] maintained that customer satisfaction could be assessed using
service quality because satisfaction is impacted by the effectiveness of services. Therefore, this study
conducts a survey to collect customer evaluations of service quality based on their consumption
experience.
The quantity of recent studies on the relationship between customer satisfaction and business
operation performance has increased. Numerous studies have examined customer behavior in the hotel
and service industries based on the theory of customer satisfaction[10,11]. Previous research shows
that high customer satisfaction has a close relationship with the intention to repurchase, word-of-mouth,
and loyalty. The more satisfaction customers have, the more willing customers are to repurchase. The
reason that customers do not return to a business and choose competitors is typically because they felt
unsatisfied after comparing their perceived value for one business with the services provided by
competitors. Many studies prove that customer satisfaction leads to word-of-mouth. When customers
feel satisfied with corporate service quality, they are willing to introduce it to others. When customers
feel unsatisfied with a service, they will also share negative verbal comments with others that are
significantly damaging to corporate image. The level of customer satisfaction is a key factor that
determines whether customers will return or recommend a product or service to others[12].

2.2 Service attributes

Many studies show that each service attribute has varying degrees of importance for customer
satisfaction. Cadotte and Turgeon[13] suggested that each hotel service attribute has varying degrees of
impact on customer satisfaction. The service attributes that directly influence customer choices are the
determinant attributes that motivate customers to repurchase and distinguish a corporation from its
competitors. If a hotel does not pay attention to the service attributes that have the greatest impact on
customer choices, hotels will receive negative feedback from customers and eventually lose
customers[14]. To evaluate the importance of service attributes, numerous relational equations have
been established by previous studies to successfully evaluate the impact of service attributes on
customer satisfaction. Each attribute, such as geographical location, price, personnel attitude, clean and
tidy environment, consistent high-quality services, service personnel, and communication with
customers, has varying degrees of importance[15,16]. The majority of previous studies have used LRA
as an analysis tool. However, this tool assumes that a certain relation (for example, a linear relation)
exists between the input and output variables, and that variables are normally distributed, which places
significant limitations on the analysis. This study uses artificial neural network to overcome the
limitations of LRA and is more suitable for assessing the impact of each service attribute on “customer
satisfaction” and “repurchase or word-of-mouth”.

3. Data collection and analysis

This study uses a questionnaire that is composed of two sections. The first section obtains customer
evaluations for each service provided by hotels. The questionnaire was designed based on previous
studies [1-3, 11, 15]. The 25 questions in the first section(Table 1) use the 5-point Likert scale for
assessment (l is very unsatisfied, 5 is very satisfied). The two questions in the second section are used
to determine if customers are satisfied (yes or no), and if they will return or recommend the service to
others (yes or no). We distributed 600 copies of the questionnaires to customers in four hotels in
northern Taiwan (two five-star hotels and two four-star hotels) between June and September in 2011.
Surveyed customers had already completed their service interaction with the hotel. Four-hundred and
fifty-six copies were returned (76%), although only 425 copies were valid (70.83%). This study used

443
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

the SPSS software package to perform descriptive statistics, factor analysis, and logistic regression
analysis. The NeuroSolutions toolbox was also used to establish a multilayer perceptrons neural
network.

3.1 Descriptive statistics results

The mean value and standard deviation of the 25 questions in the first section of the questionnaire is
shown in Table 1. The mean value of the 25 variables ranges from 3.42 to 4.11, indicating that
customers are satisfied with the services in each of the questions. The standard deviation of the 25
variables ranges between 0.48 and 0.75, indicating the high consistency of the data’s variance level.

Table 1. The mean value and standard deviation of the first section of data from the questionnaire
Code Question Mean Standard Deviation
Q1 Service accuracy of check-in/check-out procedure 3.88 0.52
Q2 Service speed of check-in/check-out procedure 3.97 0.48
Q3 Providing relevant information 3.94 0.49
Q4 Luggage service 3.84 0.51
Q5 Front desk service 3.85 0.62
Q6 Bed comfort 3.92 0.69
Q7 Convenience of room amenities 3.80 0.61
Q8 Completeness of room amenities 3.79 0.67
Q9 Room safety 4.11 0.69
Q10 Room quietness 4.02 0.75
Q11 Room cleanliness 4.09 0.69
Q12 Meeting room arrangement service 3.66 0.70
Q13 Travel arrangement service 3.52 0.68
Q14 Business facilities 3.50 0.73
Q15 Transportation and shuttle service 3.42 0.69
Q16 Food variety in restaurant 3.53 0.65
Q17 Taste of food in restaurant 3.54 0.63
Q18 Comfort of restaurant environment 3.83 0.61
Q19 Cleanliness of restaurant environment 3.85 0.64
Q20 Prompt service in restaurant 3.81 0.68
Q21 Thoroughness of restaurant services 3.83 0.69
Q22 Friendliness of service personnel 4.01 0.61
Q23 Behavior of service personnel 4.00 0.67
Q24 Service vocabulary of service personnel 3.98 0.69
Q25 Appearance and expressions of service personnel 4.03 0.75

3.2 Factor analysis

Factor analysis (FA) is employed to reduce the number of observation variables to a smaller number
of factors. Researchers can use these factors to replace original variables and conduct further analysis.
We performed FA for the first set of 25 questions in the questionnaire (Bartlett’s test of
sphericity=7737.61, p＝0.000; KMO measure of sampling adequacy＝0.896). The results show that
the questionnaire is suitable for FA. Table 2 shows that the 25 variables can be categorized into four
factors. The cumulative total variance explained is 74.82%. The factors are titled F1: personnel
services, F2: dining quality, F3: room quality, F4: business and travel services.

3.3 Artificial neural network

This study uses multilayer perceptrons neural network (MLPs) that contain three layers: an input
layer, a nonlinear hidden layer, and an output layer (Figure 1). The hidden layer and the output layer
apply the tanh transfer function. NeuroSolutions software (Neuro Dimension Inc., Gainsville, Florida,
USA) is employed to construct the required MLPs.

444
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

Table 2. Factor analysis results

F1: F2: F3: F4:
Code* Personnel services Dining quality Room quality Business and travel
services
Q2 0.895
Q1 0.843
Q4 0.789
Q5 0.756
Q3 0.732
Q23 0.720
Q24 0.720
Q22 0.694
Q25 0.610
Q18 0.827
Q16 0.738
Q17 0.744
Q20 0.709
Q21 0.614
Q19 0.597
Q10 0.759
Q7 0.737
Q11 0.71l
Q9 0.675
Q8 0.676
Q6 0.586
Q14 0.802
Q15 0.763
Q13 0.742
Q12 0.599
% of variance 45.21% 13.86% 9.42% 6.33%
Cumulative % 45.21% 59.07% 68.49% 74.82%
*. For an explanation of these codes, please refer to Table 1.

Input layer Hidden layer Output layer

Personnel services

Dining quality .
. Customer satisfaction
.
. or
.
.
Room quality Repurchase or word-of-mouth

Business and travel services

Figure 1. Multilayer perceptrons neural network

The MLPs has been trained based on 250 pieces of questionnaire data. The cross validation process
of the network uses a data set of 75 pieces of questionnaire data. Cross validation is an optional method
that is highly recommended for terminating network training when independent data set errors begin to
increase. This is considered the point of best generalization. The testing process is defined as data used
to evaluate performance after training is complete. The trained network was then tested using 100
pieces of questionnaire data that were not used in the training set. The number of nodes of the network
is the number of exemplars (rows) of the training set equal to 250 and 1,000 epochs. The number of
nodes is configured automatically by the NeuralBuilder. To validate the proposed system, positive
predicted value (PPV) was computed in each case as:

445
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

(1)

3.3.1 Training, verification, and testing results

For each service attribute’s (F1 to F4) predicting ability for customer satisfaction, the learning curve
of MLPs training and verification (Figure 2) shows effective error convergence without overfitting.
The results are shown in Table 3. Overall PPV for the test is 93%, indicating effective predicting
ability. A sensitivity analysis was conducted to distinguish the impact of each input factor (F1 to F4)
on customer satisfaction (Figure 4). The importance of each hotel service attribute to customer
satisfaction, from highest to lowest, is personnel services (37.34%), room quality (28.69%), dining
quality (24.33%), and business and travel services (9.64%).

Figure 2. The MLPs learning curve

For each service attribute’s (F1 to F4) predicting ability for repurchase or word-of-mouth, Table 3
shows that the overall PPV for the test is 90%. Figure 3 shows that the influence of each hotel service
attribute (F1 to F4) on repurchase or word-of-mouth, from highest to lowest, is personnel services
(34.68%), room quality (24.11%), dining quality (27.99%), and business and travel services (13.22%).

Table 3. PPV of MLPs training, verification, and testing

Output: Customer satisfaction Output: Repurchase or word-of-mouth
Training Validation Testing Training Validation Testing
dataset dataset dataset dataset dataset dataset
PPV 97.2% 98.67% 93% 91.6% 94.67% 90%

446
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

Figure 3. The importance of input variables to output variables

3.3.2 Logistic regression analysis

To compare the results derived using MLPs, we used binary logistic regression analysis to establish
the relationship between each service attribute (F1 to F4) and customer satisfaction, and the
relationship between each service attribute (F1 to F4) and repurchase or word-of-mouth. The results are
shown in Table 4 and Table 5, respectively.
Table 4 shows that the p value of the factor “Business and Travel Services” is 0.269 for customer
satisfaction and 0.498 for repurchase or word-of-mouth. Both are greater than 0.05, indicating that
“Business and Travel Services” does not have a significant impact on customer satisfaction and
repurchase or word-of-mouth. Table 5 shows that the PPV for LRA is 87% for customer satisfaction
and 80% for repurchase or word-of-mouth, both amounts are lower than those for MLPs results.

Table 4. Binary logistic regression analysis results

Output: Customer satisfaction Output: Repurchase or word-of-mouth
β S.E. Wald p β S.E. Wald p
Personnel
0.389 0.092 17.873 0.000 0.229 0.070 10.761 0.001
Services
Room Quality 0.601 0.113 28.288 0.000 0.233 0.077 9.051 0.003
Dining Quality 0.142 0.068 4.403 0.036 0.176 0.058 9.329 0.002
Business and
0.116 0.105 1.222 0.269 0.056 0.083 0.460 0.498
Travel Services
Constant -26.212 4.974 27.770 0.000 -14.656 3.535 17.193 0.000

Table 5. PPV of LRA training, verification, and testing

Output: Customer satisfaction Output: Repurchase or word-of-mouth
Training Validation Testing Training Validation Testing
Dataset Dataset Dataset Dataset Dataset Dataset
PPV 86% 85.3% 87% 80% 87% 80%

447
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

4. Discussion

4.1 Service attributes that influence customer satisfaction and their impact levels

The importance of each hotel service attribute to customer satisfaction, from highest to lowest, is
personnel services (37.34%), room quality (28.69%), dining quality (24.33%), and business and travel
services (9.64%).
(1) Personnel services
The personnel services attribute has the most significant impact on customer satisfaction. The face-
to-face contact between personnel and customers is a crucial factor that affects customer satisfaction.
This result is in accordance with the view of Cadotte and Turgeon[13] that when customers evaluate
overall service quality, personnel service is the most significant factor. In the hotel industry, which
includes intense competition, the amenities and services provided by each hotel are similar. The
enhancement of personnel services should be a critical management subject for hotels attempting to
differentiate themselves from their competitors. Management should address the vocabulary, behavior,
manner, and appearance of service personnel.
(2) Room quality
The impact of room quality on customer satisfaction has the second-highest significance for
customer satisfaction. Although there is debate on whether room quality affects customer satisfaction
[16], this study shows that room quality still had a reference value for hotels participating in the study.
Management should consider customers perspectives and use hotel resources to enhance room quality
factors such as room cleanliness, full amenities, and room safety. When creating a promotional plan,
management should place emphasis on tangible evidence of hotel quality such as hotel room amenities
to retain existing customers and attract new customers.
(3) Dining quality
The importance of dining quality to customer satisfaction is 24.33%. McCleary[17] argued that the
dining service quality in hotels is not significant to customers. Most customers do not think that dining
in a hotel is necessary. This viewpoint differs from the results of this study. When staying at a hotel,
customers have the opportunity to dine in the hotel. Even one dining experience affects customer
evaluation of the hotel. Therefore, management should make efforts to improve dining quality, increase
the variety of the food, enhance the taste of the food, maintain a comfortable and clean dining
environment, and provide prompt and thorough services.
(4) Business and travel services
This attribute has the least impact on customer satisfaction, with an importance of 9.64%. This study
presumes that the reason for this phenomenon is that customers have arranged their business or travel
activities before checking into the hotel; therefore, hotels only assist to fulfill or complete these
arrangements. When there is a demand from customers, prompt and thorough assistance can satisfy
customers’ expectations.

4.2 Service attributes that influence repurchase or word-of-mouth and their impact
levels

The importance of each hotel service attribute to repurchase or word-of-mouth, from highest to
lowest, is personnel services (34.68%), dining quality (24.11%), room quality (27.99%), and business
and travel services (13.22%). The total importance of the top three service attributes is 86.78%,
indicating that business and travel services have an insignificant impact on repurchase or word-of-
mouth behavior. To motivate customers to repurchase or recommend services to others, hotel
management should focus on improving three aspects: personnel services, dining quality, and room
quality. For business and travel services, they should maintain this service to a certain degree and seek
to improve it when resources are available.

4.3 The relationship between customer satisfaction and repurchase or word-of-mouth

The cross analysis of the 425 questionnaires in this study (Table 6) shows that up to 89.63% of
satisfied customers expressed their intention to repurchase or recommend the service to others. By

448
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

contrast, up to 91.18% of unsatisfied customers expressed their unwillingness to repurchase or

recommend the service to others. The correlation coefficient (r) of customer satisfaction and
repurchase or word-of-mouth is 0.747. This shows a highly positive correlation between customer
satisfaction and repurchase or word-of-mouth. Hotel management should address high customer
attrition rates that result from unsatisfied customers. Management must avoid causing customer
dissatisfaction. When receiving customer complaints, they should actively engage in service recovery
to meet or possibly exceed customer expectations, which may regain customer trust and reduce the rate
of customer attrition.

Table 6. Cross analysis for customer satisfaction and repurchase or word-of-mouth

Satisfied customers Unsatisfied customers
Will repurchase or recommend 320/357＝89.63% 6/68＝ 8.82%
Will not repurchase or recommend 37/357＝10.37% 62/68＝91.18%

4.4 Comparison of MLPs and LRA

Table 3 and Table 5 show that MLPs have greater PPV than LRA for training, verification, and
testing. For predicting customer satisfaction and repurchase or word-of-mouth, MLPs have greater
overall accuracy than LRA. Table 7 shows the detailed prediction results of MLPs and LRA for the test
data. For customer satisfaction, out of the 100 test data, 87 customers rated their hotel as “satisfactory,”
and 13 customers rated theirs as “unsatisfactory.” MLPs and LRA have the same accuracy for
predicting customers who gave a “satisfactory” rating, which was 84/87 (96.55%). However, when
predicting customers who gave a rating of “unsatisfactory,” the accuracy of MLPs was 9/13 (69.23%)
and the accuracy of LRA was 3/13 (23.08%), which is was unacceptable. A similar situation occurs
when predicting if customers will repurchase or recommend to others. 79 customers expressed their
intention to repurchase or recommend the service to others, and 21 customers expressed their
unwillingness to repurchase or recommend the service to others. For predicting customers who will
repurchase or recommend the service to others, the accuracy of MLPs is 75/79 (94.94%) and the
accuracy of LRA is 78/79 (98.73%). This shows that LRA has greater accuracy. However, for
predicting customers who will not repurchase or recommend the service to others, the accuracy of
MLPs is 15/21 (71.43%) and the accuracy of LRA is 2/21(9.52%).

Table 7. Testing data set classification

Output: Customer satisfaction Output: Repurchase or word-of-mouth
Model Class Satisfactory Unsatisfactory Percentage Yes No Percentage
True 84 9 93% 75 15 90％
MLPs
False 3 4 7% 4 6 10％
True 84 3 87% 78 2 80％
LRA
False 3 10 13% 1 19 20％

5. Conclusions

The study results show that of the four categories of hotel service attributes (personnel services,
room quality, dining quality, and business and travel services), personnel services has the greatest
impact on hotel customer satisfaction, followed by room quality, dining quality, and business and
travel services. To increase customer satisfaction, management should improve personnel services first
by implementing measures such as personnel training, authorization, performance appraisal, reward
and punishment systems, improving work environments, and respecting employees. Management
should also allocate resources to improve room quality. Room factors valued by guests include room
comfort, full amenities, safety, quietness, and room cleanliness. Management should focus on these
factors to maintain high standards of service quality. For marketing strategies, hotels should advertise
based more on their competitive advantages of better personnel services and room quality to
differentiate themselves from their competitors.

449
 Research on the Influential Factors of Customer Satisfaction and Post-Purchase Behavior for Hotels:
The Multilayer Perceptrons Neural Network Approach and Logistic Regression Analysis
Han-Chen Huang

For analyzing the impact of service attributes on customer satisfaction and repurchase behavior, this
study used MLPs neural network and LRA. For predicting overall customer satisfaction and repurchase
or word-of-mouth, the accuracy of MLPs is superior to that of LRA. For predicting customer
satisfaction, the overall accuracy of MLPs is 93%. For predicting repurchase or word-of-mouth
behavior, the overall accuracy of MLPs is 90%. Although the overall accuracy of LRA is 87% for
predicting customer satisfaction, the accuracy is only 23.77% for predicting unsatisfied customers,
whereas the accuracy of MLPs is 69.23%. For predicting repurchase or word-of-mouth behavior, the
overall accuracy of LRA is 80%. However, for predicting customers who will not repurchase or
recommend a service to others, the accuracy of LRA is only 9.52%, compared to the accuracy of MLPs,
71.43%. Therefore, MLPs is more suitable than LRA for predicting customer satisfaction and
repurchase behavior.

6. References

[1] D. Zhang, X. Zeng, W. He, "Research on Customer Satisfaction of E-Commerce Website with
Uncertain Linguistic Variables", JCIT, Vol. 7, No. 1, pp. 165 ~ 171, 2012
[2] Y. H. Hsieh, "A Dynamic Model of Service Recovery", JCIT, Vol. 6, No. 10, pp. 391 ~ 400, 2011
[3] Y. Yang, N. Li, M. Liang, S. Li, "Study on the Effects of Logistics Service Quality on Consumers'
Post-Purchase Behavior of Online Shopping", AISS, Vol. 3, No. 11, pp. 241 ~ 247, 2011
[4] J. Anton, Customer Relationship Management: Making Hard Decisions with Soft Numbers,
Prentice-Hall, USA, 1996
[5] V. A. Zeithaml, L. L. Berry, A. Parasuraman, "The Nature and Determinants of Customer
Expectations of Service", Journal of the Academy of the Marketing Science, Vol. 21, No. 1, pp. 1~
12, 1993
[6] H. Oh, C. S. Parks, "Customer Satisfaction and Service Quality: A Critical Review of the
Literature and Research Implications for the Hospitality Industry", Hospitality Research Journal,
Vol. 20, No. 3, pp. 36 ~ 64, 1997
[7] K. Matzler, F. Bailom, H. H. Hinterhuber, B. Renzl, J. Pichler, "The Asymmetric Relationship
between Attribute-Level Performance and Overall Customer Satisfaction: A Reconsideration of
the Importance-Performance Analysis", Industrial Marketing Management, Vol. 33, No. 4, pp. 271
~ 277, 2004
[8] J. F. Engel, R. D. Blackwell, P. W. Miniard, Consumer Behavior, Doyden press, UAS,1990
[9] K. Jaksa, R. Inbakaran, J. Reece, "Consumer Research in the Restaurant Environment, Part 1: A
Conceptual Model of Dining Satisfaction and Return Patronage", International Journal of
Contemporary Hospitality Management, Vol. 11, No. 5, pp. 205 ~ 222, 1999
[10] J. Abalo, J. Varela, V. Manzano, "Importance Values for Importance-Performance Analysis: A
Formula for Spreading Out Values Derived from Preference Rankings", Journal of Business
Research, Vol. 60, No. 1, pp. 115 ~ 121, 2007
[11] M. Rathi, A. Priyadarshini, A. Rastogi, "Predictive Analysis for Customer Relationship
Management", JDCTA, Vol. 4, No. 2, pp. 95 ~ 99, 2010
[12] K. Matzler, E. Sauerwein, K. A. Heischmidt, "The Factor Structure of Customer Satisfaction: An
Empirical Test of the Performance Grid and the Penalty-Reward-Contrast Analysis", International
Journal of service industry Management, Vol. 23, No. 2. pp. 112 ~ 129, 2002
[13] E. R. Cadotte, N. Turgeon, "Key Factors in Guest Satisfaction", The Cornell Hotel and Restaurant
Administration Quarterly, Vol. 34, No. 1, pp. 45 ~ 51, 1988
[14] B. R. Lewis, "Getting the Most from Marketing Research", The Cornell Hotel and Restaurant
Administration Quarterly, Vol. 25, No. 4, pp. 82 ~ 96, 1985
[15] D. C. Gilbert, L. Morris. "The Relative Importance of Hotels and Airlines to the Business
Traveler", International Journal of Contemporary Hospitality Management, Vol. 7, No.6, pp. 19 ~
23, 1994
[16] E. W. Anderson, V. Mittal, "Strengthening the Satisfaction-Profit Chain", Journal of Service
Research, Vol. 3, No. 2, pp. 107 ~ 120, 2000
[17] K. W. McCleary, P. A. Weaver, J. C. Hutchinson, "Hotel Selection Factors as they Relate to
Business Travel Situations", Journal of Travel Research, Vol. 32, No. 2, pp. 42 ~ 48, 1993

450