Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.16 No.1 pp.103-108

Consumer Analysis for Increasing Occupancy Rates of Tourism Hotel

Gozaly Jimmy*
Department of Industrial Engineering, Maranatha Christian University, Bandung, Indonesia
Corresponding author :
April 29, 2016 August 25, 2016 October 3, 2016


Tourism is a sector that plays an important role in the economic growth of Indonesia. Bandung as the capital of West Java province is known as the city with diverse tourism potential, both in the attractiveness of the city and surrounding natural beauty. DHR is a three stars resort hotel in the city with a strategic location. As a three stars resort hotel, DHR has been experiencing occupancy rate problems, consequently, it cannot often reach the set targets, both during high season and low season. The purpose of this study is to identify the factors of hotel performance that influences consumer staying back decision in the future. Questionnaires have been distributed to hotel guests to gather information regarding their interest and the performance assessment of the hotel services and facilities, and staying back decision in the future. Discriminant Analysis and Hypothesis Testing are used to determine which hotel performance variables will directly affect consumer staying back decision. The result of this study provide marketing strategy that should be implemented by the hotel management in order to increase its occupancy rate.



    DHR is a hotel in the city which located not far from the city, yet it has a beautiful natural environment. Furnished with rooms in traditional architecture of West Java, the atmosphere of the mountains, and complete facilities, the hotel has a potential as a promising tourist destination. However the hotel has been experiencing occupancy rate problems, which often cannot achieve the set targets, both during high season and low season.

    To overcome the problem, hotel management have to know the level of consumers satisfaction for the hotel performance, and which factors will directly influence their staying decision. The study is conducted to provide inputs for the hotel management about things that need to be considered in relation to increase the level of consumer satisfaction.


    2.1.Customer Satisfaction

    Customer Satisfaction is the internal feelings of every individual which may reflect their satisfaction or dissastifaction resulting from the assesment of service provided to an individual in context to customer’s anticipation by an organization (Leisen and Vance, 2001). Satisfied customers are produced when the service provided (as perceived by the guest) is more than that expected by the guest (Mill, 2002). When customers are satisfied, they remain loyal with the hotel and hence it will affects the purchasing behavior (Saleem and Raja, 2014). Enterprises which are able to rapidly understand and satisfy customer’s needs, make greater profits than those which fail to understand and satisfy their customers (Barsky and Nash, 2003). Providing high quality services and improving customer satisfaction are widely recognized as fundamen- tal factors boosting the performances of companies in the hotel and tourism industry (Barsky and Labagh, 1992).

    2.2.Discriminant Analysis (Hair et al., 2006)

    Discriminant Analysis is an appropriate statistical techniques when a research problem involves a single categorical dependent variable and several metric independent variables. The results of discriminant analysis can assist in profiling the intergroup characteristics of the subjects and in assigning them to their appropriate groups.

    Discriminant analysis involves deriving a variate. The discriminant variate is the linear combination of the two or more independent variables that will discriminate best between the objects in the groups defined a priori.(1)

    Z jk  = a +  W 1 X 1k  +  W 2 X 2k  +  + W n X nk

    where :

    • Zjk:  discriminant Z score of discriminant function j for object k

    • a:  intercept

    • Wi :  discriminant weight for independent variable i

    • Xik :  independent variable i for object k

    2.3.Two-Sample Test of Hypotesis: Independent Samples (Lind et al., 2008)

    Hypothesis testing is a procedure based on sample evidence and probability theory to determine whether the hypothesis is a reasonable statement.

    After stating the null hypothesis (Ho) and the alternate hypothesis (H1), select a level of significance (α), and select the appropriated test statistic, then decision can be made based on decision rule.

    Testing for two-independent samples are :(2)

    Z= X ¯ 1 - X ¯ 2 S 1 2 n 1 + S 2 2 n 2


    The study aims to determine the factors that influence consumer staying back decision. Datas are collected through the questionnaires which are compiled and based on the model of "Seven Ps" (Product, Price, Promotion, Place, Physical Evidence, People, Process). There are three groups of consumer who are involved in the research, namely a group of consumer who decided to stay back, a group of consumers who might decide to stay back, and a group of consumers groups who decide not to stay back. Proposed improvements are given based on factors that simultaneously affect all three groups and are still considered unsatisfied.

    3.1.Data Collection

    Consumer information is obtained through questionnaires distributed to 130 local hotel guests. Foreign hotel guests are excluded in this survey with a consideration of differences in their assesment standard. Consumer are asked to rate the performance of the hotel based on the marketing mix variables, and a decision to stay back or not.

    3.2.Research Model

    The model used in this study is seen as follows:

    3.2.1.Consumer Performance Assessment and Interest Rate

    Consumer performance assessment and interest rate for the hotel are done by using variables that are developed and based on the model 7Ps (Product, Price, Place, Promotion, People, Physical Evidence, and Process). (Zeithaml et al., 2013).Figure 1

    The variables are :

    • VAR01:  Diversity of room type

    • VAR02:  Adequate room facilities

    • VAR03:  Room facilities are functioning properly

    • VAR04:  Other supporting facilities (meeting room, beauty spa, etc.)

    • VAR05:  Food and beverages quality

    • VAR06:  Hotel is easily reached with the help of GPS / signpost

    • VAR07:  Ease in terms of transportation

    • VAR08:  Adequate parking area

    • VAR09:  Prices fit for a three-star resort hotel

    • VAR10:  Discount for several rented rooms

    • VAR11:  Promotion through print media

    • VAR12:  Room cleanliness

    • VAR13:  Good air circulation

    • VAR14:  Security in the hotel

    • VAR15:  Leisure in the hotel

    • VAR16:  Beauty of surrounding

    • VAR17:  Employee appearance

    • VAR18:  Good room lighting

    • VAR19:  Sports facilities

    • VAR20:  Interesting interior design

    • VAR21:  Recreational facilities

    • VAR22:  Hotel landscaping

    • VAR23:  Employee hospitality

    • VAR24:  Employee responsiveness

    • VAR25:  Ease of payment process

    • VAR26:  Ease of booking process

    • VAR27: The speed of check in and check out process

    The scales used for this question are:

    Performance Assesment

    • 1 : Very Bad

    • 2 : Bad

    • 3 : Good

    • 4 : Very Good

    Interest Rate

    • 1 : Very Not Important

    • 2 : Not Important

    • 3 : Important

    • 4 : Very Important

    3.2.2.Consumer Stay Back Decision

    The data of consumer decision to stay back 1s obtained through the following questions: Do you want to stay back at DHR hotel?

    The scale used for this question is:

    • 1: Not interested

    • 2: Probably

    • 3: Assuredly

    3.3.Data Processing

    3.3.1.Discriminant Analysis

    This method aims is to find independent variables that significantly affect and distinguish between groups of dependent variable.

    Variables used in Discriminant Analysis is as follows:

    • Independent Variables consisting of 27 variables on hotel performance assessment by consumer.

    • Dependent Variable is decision to stay back from the consumer.

    The data will be processed by using SPSS software.

    3.3.2.Hypotesis Testing

    Interest rate datas and DHR Hotel’s level of performance are processed together with two sample hypothesis testing to determine consumer’s satisfactory level for each variable that is derived from Discriminant Analysis process.


    4.1.Assessing Variables’ Feasibility for Discriminant Analysis

    Initial Discriminant Analysis processing, will be find out any independent variables that significantly affect and distinguish between groups of consumer (not interested to stay back, probably stay back, and assuredly stay back). To assess the feasibility of independent variables used in Discriminant Analysis, tests are carried out as follows:

    The results in Table 1 show that only two of independent variables (VAR02, VAR10) have Sig.value < 0.05, which means that those variables are feasible for further processing due to significantly differentiate groups of consumer stay interest. The Wilks’ Lambda value show the importance of independent variable to the discriminant function (a smaller value indicate more important).

    4.2.Forming Variables for Discriminant Analysis

    At the next stage of datas processing it appears that the discriminant function is only formed by two variables, namely VAR10 and VAR02 (Table 2) , with significant value 0.002 and 0.001 (<0.05) (Table 3).

    4.3.Discriminant Function for Discriminant Analysis

    At this stage the model will be obtained and discriminant validity of the model determined.

    The results in Table 4 show that two discriminant functions are formed with Chi-square value of 18.513 and Sig. 0.001, which indicate there is a significantly differences in performance assesment for VAR02 and VAR10 between 3 groups of consumer with different staying back decision.

    The results in Table 5 show that two discriminant functions are formed, which are :

    ZScore _1 = -5,475+(0,796VAR02)+(1,056VAR10) ZSscore_2 = -0,573+(1,168VAR02)+(-0,963VAR10)

    The two functions above can be used to predict consumers decision to stay back based on their performance assesment for VAR02 and VAR10. Higher variable value in each function shows how strong those variables in dis- criminating the groups. For the first function, VAR10 is stronger than VAR02, and for the second function VAR02 is stronger than VAR10.

    4.4.Territorial Map for Discriminant Analysis

    The Discriminant analysis results indicates that there are two variables that influence consumer staying back interest, which are VAR02 (adequate room facilities) and VAR10 (discount for several rented rooms).

    Territorial map (Figure 2) showed the mapping of the boundaries of each consumer group by the Function 1 (X axis) and Function 2 (Y axis), and it can be used to predict consumer group based on their performance assessment score.

    To determine the consumer group in territorial map, ZScore 1 will be plotted on the X axis and ZScore 2 will be plotted on the Y axis in the territorial map.


    A consumer who assesses the performance of VAR02 Very Good (4) and VAR10 Good (3), will generate discriminant score as follows:

    ZScore _1 = -5,475 + (0,796VAR02) + (1,168VAR10) = -5,475 + (0,796×4) + (1,168×3) = 1,213

    ZScore_2 = -0,573 + (1,168VAR02) + (-0,963VAR10) = -0,573 + (1,168×4) + (-0,963×3) = 1,21

    Based on territorial map, consumer is included in area 3 (assuredly stay back).

    4.5.Model Validity for Discriminant Analysis

    The results in Table 6 show how well the discriminant function work for each group of dependent variable. Cases classified correctly is 71,4% for Not Interested group, 25% for Possibly group, and 74,6% for Assuredly group. Overall, 54,6% of the cases are correctly classified. The lowest correct classification result is for “Possibly” group, thus resulting in a low overall classified value. Ideally, only middle value of VAR02 and VAR10 performance assesment should generate “Possibly” stay back decision, but there are some customers with a low or high assesment values who cannot decide whether to stay back or not in the future, so they choose “Possibly” answer. This consumers hesitancy will cause the low value of “Possibly” group classification result.

    4.6.Group Statistics

    The results in Table 7 show that consumer with low stay back interest tends to have low performance assessment result. The Total of 3 groups mean values both for VAR02 and VAR10 still are less than 3 (good category), which shows that both of variables must be improve by the hotel management.

    4.7.Two-Sample Test of Hypotesis

    The level of consumer satisfaction is known by comparing the average rate of interest with the average performance level for each variable derived from Discriminant Analysis process.

    The hypotheses are:

    Ho: μ1 = μ2

    (No difference between consumer performance level with interest level, so that visitors are satisfied)

    Hi : μ1 μ2

    (Consumer performance level lower than interest level. so that visitors are dissatisfied)

    Level of significance (α): 0.05


    Z= X ¯ 1 - X ¯ 2 S 1 2 n 1 + S 2 2 n 2

    Region of rejection: Z < - 1,645 (taken from Normal distribution table)

    The results of Hypothesis Testing in Table 8 show that consumers in Not Interested group and Possibly group feel disstastified regarding with VAR02 and VAR10, yet only consumer in Assuredly group feel satisfied about those variables.


    The research shows that there are two hotel performance factors (adequate room facilities and discount for several rented rooms) that influence customer decision to stay back in the future. Customer decision can be predicted by using two function formed by discriminant analysis which are : ZScore _1 = -5,475+(0,796 VAR02) +(1,168 VAR10) and ZScore_2 = -0,573+(1,168 VAR02) +(-0,963 VAR10). Based on validation results, the model validity is 54,6%, therefore discriminant function formed is considered quite appropriate to classify customers based on the performance assessment.

    There is a clear difference between the performance assesment results from the three group of consumers (not interested in staying back, probably staying back, and assuredly staying back), consumer with high interest to stay back (“Assuredly” group) tend to have a higher assesment value and feel satisfied regarding the two variables of hotel performance, even though the average value of performance assesment from the three groups is still slightly below the expectedly good limit.

    This research suggest enhancing the level of customer satisfaction for room facilities and discount rooms in order to increase consumers in “Assuredly” group. Based of field study, several marketing strategies that can be proposed to hotel management are: provide television with good image quality and variety channel selection in all room types; improve the speed and easiness of internet connection in all rooms and hotel environments; increase the number of rooms with connecting facilities, provide special discounts for several rented rooms by classifying discounts according to the number of rented rooms. Hotel management must constantly observe the level of customers satisfaction to provide the best service quality for them. Satisfied customers most likey will stay back and increase the occupancy rate of the hotel.



    Research Model.


    Territorial map.


    Tests of equality of group means

    Variables entered

    Variables significance

    Chi square

    Canonical Discriminant Function Coefficients

    Unstandardized coefficients

    Classification results

    NI = Not Interested P = possibly A = Assuredly.

    Group Statistics

    Results of Hypothesis Testing


    1. Barsky J , Labagh R (1992) A strategy for customer satisfaction , The Cornell Hotel and Restaurant Administration Quarterly, Vol.35 (3) ; pp.32-40
    2. Barsky J , Nash L (2003) Customer satisfaction: applying concepts to industry-wide measures , The Cornell Hotel and Restaurant Administration Quarterly, Vol.44 (4) ; pp.173-183
    3. Gozaly J (2015) Consumer analysis for increasing occupancy rates of tourism hotel , Proceedings of Abstracts and Papers of the 16th Asia Pacific Industrial Engineering & Management Systems Conference (APIEMS 2015): VNU-HCMC Press and Authors, ; pp.1567-1572
    4. Hair J F , Black W C , Babin B J , Anderson R E (2006) Multivariate Data Analysis, Pearson International Edition, ; pp.101269-164354
    5. Leisen B , Vance C (2001) Cross-national assesment of service quality in telecommunication , Journal of Marketing, Vol.11 (5) ; pp.307-317
    6. Lengkoan M P R (2015) Proposal of Marketing Strategy Based on Customer Satisfaction to Increase Occupancy Rate, Department of Industrial Engineering, Maranatha Christian University,
    7. Lind Marchal Wathen (2008) Statistical Techniques in Business and Economics With Global Data Sets 13th edition , McGraw-Hill International Edition, Vol.13 ; pp.388-391
    8. Mill R C (2002) A comprehensive model of customer satisfaction in hospitality and tourism: Strategic implications for management , International Business & Economic Research Journal, Vol.1 (6) ; pp.7-18
    9. Saleem H , Raja N S (2014) The impact of service quality on customer satisfaction, customer loyalty, and brand image: Evidence from hotel industry of pakistan , Journal of Business and Management, Vol.16 (1) ; pp.117-122
    10. Santoso S (2012) Aplikasi SPSS Pada Statistik Multivariat , Elex Media Komputindo, ; pp.153-204
    11. Sekaran U (2003) Research Methods for Business, A Skill Building Approach 4th edition, John Wiley & Sons, Vol.185-193 ; pp.236-244
    12. Zeithaml V A , Bitner M J , Gremler D D (2013) Services Marketing 6th Edition, John Wiley & Sons, ; pp.24-27