An artificial neural network approach to hotel employee satisfaction: The case of China
Main Article Content
At present, the hotel employment sector in China has a high rate of employee turnover compared to other services. This is not unlike other countries. The reason for the turnover among hotel employees may be lower worker satisfaction resulting in decreased - or no - loyalty to employers. This study was based on an Artificial Neural Network (ANN). The factors influencing employee satisfaction were examined and the impacts of demographic characteristics on hotel employee satisfaction were analyzed. Results show that hotel employee satisfaction in China is low, hotel employee satisfaction differs by age and gender, and that professional development opportunities for employees and the long-term growth prospects of the hotels themselves are the most important contributors to employee satisfaction. On the basis of these findings, several recommendations for improving employee satisfaction, thereby sustaining the long-term economic health of China’s hospitality industry, are provided.
Many empirical studies show a strong positive relationship between employee satisfaction and customer satisfaction (Band, 1988; George, 1990; Schmitt, 1999). This is quite true in service businesses. As Rogers, Clow, and Kash note in their 1994 study (p. 15), “Increasing job satisfaction among service personnel has the potential of generating higher customer satisfaction.” Compared to the manufacture of a tangible product, the “product” of the hotel industry is encounter service. In addition to clean, safe, and special lodging units, the main product is efficient and pleasant interaction with customers. Therefore, the quality of service is measured by the quality of the interaction between front-line employees and hotel guests. For the hotel sector, the product has the characteristics of intangibility, nontransferability, and synchronization of production and consumption (Goeldner, Ritchie, & McIntosh, 1999). So employees’ attitudes and behaviors influence not only actual service quality, but also the customer’s perception of service quality and overall satisfaction. According to the 2005 Chinese Tourism Report prepared by the Chinese Tourism Administration Bureau (CTAB), most of China’s hotel employees share the traits of low salary and a high degree of mobility. Most are quite young and the majority are female. The objectives of the hotel sector and the nature of its employees distinguish it from other sectors, which makes our research particularly relevant.
Numbers from CTAB show that the turnover rate of hotel employees is about 31% (the average turnover rate of other service sectors is only 17%) and hotel revenues declined by 5.03% in 2004. It seems reasonable to propose that employee satisfaction and loyalty to hotels in China are low and the resultant high turnover rate is adversely impacting China’s hospitality industry. Therefore we would like to know the level of employee satisfaction of China’s hotel sector, factors which have the greatest impact on overall satisfaction, and measures to address and reverse the human resource threats facing China’s hotel sector. Another objective of this study is to provide empirical use of an Artificial Neural Network (ANN) in the field of employee satisfaction evaluation.
Based on a literature review of employee satisfaction, a neural network was designed to measure overall employee satisfaction. Then the impact of two important demographic characteristics − gender and age − on employee satisfaction was examined. Results are presented here and followed by conclusions and recommendations.
The issue of employee satisfaction has been considered in numerous studies. The earliest of these was pioneered by Locke and Kornhauser (1969). Locke (1976) defined employee satisfaction as a pleasurable or positive emotional state resulting from the appraisal of employees or employees’ experiences. Robbins and Coulter (1996) stated that employee satisfaction is an employee’s general attitude towards his/her work, and when people speak of an employee’s attitudes, they are likely to be referring to his/her employee satisfaction. In the 1960s, several research institutes developed standard questionnaires to measure employee satisfaction, such as the Minnesota Satisfaction Questionnaire (Weiss, Dawis, England, & Lofquist, 1967) and the Job Descriptive Index (Smith et al., 1969). Since then, many scholars have devoted themselves to the field and made substantial progress. Although the literature is voluminous concerning employee satisfaction in general, there is a lack of literature specifically related to employee satisfaction in the hospitality industry. One of the small number of studies conducted on hotel employee satisfaction was done by Pizam and Chandrasekar (1983) who studied overall and factor-specific satisfaction. They found there was a high percentage of satisfaction, with 75% of the respondents voicing overall satisfaction with their positions. However, the bulk of respondents were highly educated young executives who were not reflective of the vast majority of hotel sector employees. Kent (1981) argued that there were no significant differences in job satisfaction among three categories of managerial employees: directors of sales, directors of personnel and general managers. When the size of the hotel was controlled, there were differences in satisfaction scores between general managers and the other two levels, with general managers of larger hotels expressing higher satisfaction levels. Pavesic and Brymer (1990), however, determined dissatisfaction with income, particularly in relation to the number of hours worked, to be a primary reason for managers with hospitality administration degrees leaving the hospitality industry. But no studies were found that examined the importance of job factors among nonmanagerial hotel employees (Smith, 1996).
With regard to research method, of course, there have been some traditional quantitative analysis methods, such as Multiple Regression Analysis and Structure Equation Modeling (SEM), being used to evaluate employee satisfaction. They all have merits, but the obvious weakness is that they presume a linear relationship between dependent variables and explanatory variables. However the relationship between employee satisfaction and its drivers may be nonlinear, because some employees may value monetary return and others promotion, and the importance of job factors is quite different. Using a nonlinear model to evaluate employee satisfaction is reasonable. Nowadays scholars agree that Artificial Neural Network (ANN) is a very effective tool to use when addressing a nonlinear question (Haykin, 1999). However, we could not identify any literature or studies that have utilized the ANN approach to evaluate satisfaction of hotel industry employees, managerial or other.
Method
Data Collecting and Description Analysis
Hotel employee satisfaction was investigated by using a questionnaire. A 10- point Likert scale ranging from 1 – Extremely Dissatisfied (or Strongly Disagree) to 10 – Extremely Satisfied (or Strongly Agree) was used. The survey consisted of three parts. In part one, the first question was related to overall job satisfaction and set apart on the instrument from the other questions. Part two investigated facet satisfaction of 29 variables (from V1 to V29). The third part was used to collect demographic information on hotel employees, including age, gender, educational level and so on. Scale validation was tested using Cronbach alpha and the internal reliability was measured at 0.86, exceeding Nunnally’s (1976) threshold of 0.70. For purposes of analysis, we selected one Chinese hotel chain (JinJiang Hotel International) with more than 500 employees. This hotel chain provides full service and ranks 26th (medium level) within China’s hotel industry. Both natives and foreigners are its customers and it represents the average service quality of the Chinese hotel sector. So, we thought that this hotel was most appropriate for this study. In all, 413 respondents answered our questionnaire correctly. The statistical software SPSS12.0 was used to analyze the responses. Our first step included preparation of a description analysis. The means and standard deviations of 29 variables are listed in Table 1. The data indicate that the variable mean scores range between 5 and 7 and their standard deviations are small, suggesting that most employees have medium estimating scores and the scores are relatively concentrated. The variables with which employees are most dissatisfied are: V7 - Satisfaction With Benefits in the Hotel (M = 5.01); V29 - Satisfaction with Employee Empowerment (M = 5.20); V8 - Satisfaction with Spiritual Motivation (M = 5.33). The results demonstrate low employee satisfaction with respect to both monetary and spiritual aspects of their jobs. On the contrary, the variables with which employees are most satisfied are: V3 - Satisfaction with Physical Environment (M = 7.55); V11 - Challenge of Your Work (M = 7.10). This implies overall satisfaction with hotel facilities, ambience and work environment, and that employees view their positions as challenging.
Artificial Neural Network: Introduction
There is a nonlinear relationship between overall employee satisfaction and its drivers. So, it is relevant to apply Artificial Neural Network (ANN), a new technology derived from the study of neural science, to estimate employee satisfaction. ANN provides the advantages of highly nonlinear calculation, learning ability and strong error-correction capabilities, using nonlinear parallel nodes similar to the structures and functions of human brains. There are some symbolic modes of neural networks, such as Perceptron and Hopfield networks (Rumelhart & Novig, 1986), but Back-propagation networks are the most popular among researchers. A Back-propagation algorithm is used for practical implementation of ANN with supervised learning (Kramer & Sangiovanni- Vincentelli, 1989). Back-propagation refers to the process of calculating errors by working backwards from the output. Thresholds and weights linking the input layer, hidden layer and output layer are adjusted according to the errors (i.e., differences between the given values of the nodes in the output layer and values predicted by the network). This learning process can be viewed in psychological terms, but can also be viewed as the basis for a parameter estimation procedure, which searches the overall parameter space for a set of values that minimize a suitable error criterion (Moutinho, 1996). So far, no studies have been done using ANN to research employee satisfaction, but there are several studies on the application of ANN in the area of customer satisfaction evaluation. Most of them used several impact factors of satisfaction as the inputs of neural network to gauge customer satisfaction (Goode & Davies, 2005). However, all the studies used discrete accurate values as research data, which are not precise for satisfaction estimation, because employee satisfaction is a kind of psychological feeling and has fuzzy traits. In this study continuous scope values were used to reduce such fuzzy evaluation.
Table 1. Mean and SD of Each Variable
Results
The main aim of factor analysis is to identify a few factors hidden in a large number of variables. After completion of factor analysis, the original variables are concentrated into a few factors that can be used to replace the original ones for further analysis. As with data collection and description analysis, SPSS12.0 is used to carry out factor analysis. At first, two tests − a Bartlett Test (Bartlett Test of Sphericity) and a KMO Test (Kaiser-Meyer-Olkin Measure of Sampling Adequacy) − are done to see if the samples are fit to conduct factor analysis. Usually, levels of 0.01 for the Bartlett test and 0.7 for the KMA test are considered significant. In this case, samples with a χ2 value of 0.000 < 0.01 and KMO value of 0.889 indicate that the variables are correlative and fit for factor analysis (Roxy, 2001). Six factors produced by factor analysis that had eigenvalues greater than 1.0 are shown in Table 2. These six factors accounted for 72.67% of the variance information from the samples.
Table 2. Effect of the Six Factors on Total Variance
Note: Rotation Method: Varimax with Kaiser Normalization
Factor rotation is conducted to produce the factor load matrix with the variance maximum method of orthogonal rotation. Variables belonging to the six factors that had questions loading at 0.55 or above were used to name the factors (see Table 3).
Table 3. Factor Naming
The task of naming factors requires both scientific and artistic effort. According to the principles of higher factor load, correlation and consistency, the common features of the variables are extracted in order to name the factors. Factor 1 consists of V20, V21, V22, and V25, each influencing both employees and the hotel in the future. So, Factor 1 can be named as a developmental element of both employees and the hotel. Factor 2, including V8, V9, V23, and V24 represents psychological motivation from the work, an element perhaps more aptly referred to as “spiritual reward.” Alternatively, Factor 3, showing the material attributes of V7 and V10, can be referred to as “monetary reward.” Factor 4 can be referred to as “soft environment of work” because V15, V19, V26, V27, and V29 have in common elements of the appeal of the work to human sensibility. In much the same way, Factor 5 can be termed “social evaluation” of the work because V4 and V6 represent society’s reaction to the work. Factor 6, controlled by V1 and V2, is termed “image of enterprise,” for obvious reasons.
Construction, Training and Test of Ann Model
For most research, a three-layer structure network is appropriate. Accordingly, a three-layer BP neural network was constructed to analyze the data with the help of software Matlab7.1. This is a back-propagation network with six input nodes (N), corresponding to the six factors. The output layer has one node (M) representing overall employee satisfaction. One hidden layer of 19 nodes was used according to the popular formula: G = (0.43NM+0.12M2+2.54N+0.77M+0.35)1/2 +0.51 = 19 (Zhao et al., 2005). Figure 1 depicts the ANN model for this study.
Figure 1. A Three-Layer Ann Model.
We randomly selected 200 employees as training samples and 50 employees as testing samples. Using a tool-box of software Matlab7.1, the authors devised a neural network computing program (see the Appendix). Activation function of the first layer is Transig-function, its output scope is (-1, +1), the second layer activation function is log-sigmiod and its output scope is (0, 1). Some parameters were set as follows: the error between actual output and expected output is smaller than 10-3 (i.e., Goal = 10E-3; Learning rate: Lr = 0.05; Times of epochs = 3000). Figure 2 demonstrates the iterative procession and shows the network begins to converge after 105 epochs and the error has reached the goal, so the network is trained. After finishing training network with the samples, 40 training samples were selected to test the training accuracy rate, requesting error =|actual output- expected output|<0.05. The result tells us that the training accuracy rate is 95%. Next, 50 testing samples were introduced to the network to check the prediction accuracy rate. Samples displaying errors less than 0.0599 are considered correct. The prediction accuracy rate of the ANN test is 91%. In other words, the ANN model in question correctly predicts 91% of the cases and the predictive validity of the ANN modeling is judged as being quite strong. So, the network has been trained successfully and can be used.
Figure 2. Epochs of the Networks.
Application of the Ann and Research Findings
Measurement of Overall Employee Satisfaction The trained ANN model could then be used to evaluate hotel employee satisfaction. Because the means of samples are the unbiased estimator of population, means of six factors should be able to reflect the average variance of total employees in the enterprise. If the means of six factors are introduced into the network, its output should be the overall employee satisfaction level in the given hotel. In this case, the mean of each factor is assumed as μn (μn refers to the mean of Fn, n = 1, 2, 3, 4, 5, 6). Thus, the means of six factors of hotel employees take the following form:
μ = (μ1, μ2, μ3, μ4, μ5, μ6)T
Because the factor analysis for the samples yields a standard factor matrix with a mean of each factor: 0 and the standard deviation: 1,
μ = (0, 0, 0, 0, 0, 0) T
When this vector is introduced into the ANN, the output is total employee satisfaction of this hotel. That is to say, Y0 = 0.5848 ¨(0.55, 0.65). Care must be taken that the network is only adapted to evaluate the employee satisfaction of this particular hotel chain because the training samples originated from it. In order to measure that of a different, competing hotel chain, this network would require training using samples drawn from that chain.
Impact of Demographic Features on Employee Satisfaction The optimal network is employed to measure total employee satisfaction based on gender and age (under 26 and over 26), with the objective being to discover some meaningful results through comparison of the satisfaction change.
First, using software SPSS to classify the samples according to gender, “one” (1) represents male and “two” (2) female. The means of male employees concerning six factors are:
μ(1) = (-0.0618875, 0.0086414, 0.0999411, -0.0147719, -0.0644828, - 0.0243742)T.
μ(1) is put into the ANN model to obtain male employee satisfaction, the output (total employee satisfaction of male) is:Y1 = 0.5802. Similarly, means of six factors regarding female employees are:
μ(2)=(0.0304819,-0.0042562,-0.0492247,0.0072757,0.0317602,0.0120052)T.
Placing μ(2) into the ANN to get total female employee satisfaction, yields Y2 = 0.5871. The result shows that the employee satisfaction of both males and females falls into the same estimating score zone (0.55, 0.65), which means that gender does not have a significant impact on hotel employee satisfaction.
In the same way, SPSS 12.0 was used to classify samples according to age, “three” (3) represents those whose age is under 26, “four” (4) represents those aged above 26. The means of six factors concerning employees whose age is under 26 and over 26 are available:
μ(3) = (-0.0795134, -0.0057348, -0.0959628, -0.0251924, 0.0551146, - 0.0581538)T.
μ(4)=(0.3622276, 0.0261251, 0.4371636, 0.1147654,-0.2510776, 0.2649230) T.
Then μ(3) and μ(4) are respectively put into the ANN model to evaluate total employee satisfaction for both age bands. The outputs are:
Y3 = 0.5660¨(0.55,0.65); Y4 = 0.6708¨(0.65, 0.75).
It is an amazing finding that the employee satisfaction for those over 26 is obviously higher than that of employees under 26, leading to a conclusion that age does have a significant impact on hotel employee satisfaction.
Impact Degree of Six Factors on Hotel Employee Satisfaction Through factor analysis, the research identified six factors that are determinants of employee satisfaction: (1) Development Prospects; (2) Spiritual Reward; (3) Monetary Reward; (4) Soft Environment; (5) Social Evaluation; (6) Image of Enterprise. The researcher endeavored to determine whether or not they have the same impact on hotel employee satisfaction, because hoteliers are most concerned about the relative importance of these factors. Next, the trained ANN was used to study the impact degree of six factors on overall employee satisfaction. We evaluated the impact of drivers on employee satisfaction by improving one standard deviation for each factor mean and compared overall employee satisfaction change.
According to factor analysis, Mean of each factor: μ = 0; SD of each factor:
σ = 1.
We have determined that: μ = (μ1, μ2, μ3, μ4, μ5, μ6 )T = (0, 0, 0, 0, 0, 0)T which was put into the network and overall employee satisfaction was obtained:
Y0 = 0.5848 ¨(0.55,0.65)
When the mean of F1 is increased by one standard deviation, keeping other means the same, the input vector becomes:(1, 0, 0, 0, 0, 0)T. If this result is applied to the network, the output is employee satisfaction when the mean of F1 increased by one SD. Similarly, if the input vector is (0, 1, 0, 0, 0, 0)T, the output is the employee satisfaction with the mean of F2 increased by one SD, keeping other means of factors unchanged. All input vectors with the mean of each factor increased by one SD are shown in Table 4.
Table 4. Employee Satisfaction Change (Output Value)
From the output values in Table 4, we can see that F1 (development prospects) makes the largest contribution to employees’ total satisfaction because the output value (0.9542) is the largest one in the six output values when each mean of six factors is increased by one SD and other means are kept constant. So, F1 yields the greatest potential to improve hotel employee satisfaction. In much the same way, we know that F5 generates least improvement to employee satisfaction. The order of six factors based on their influence on overall satisfaction is F1, F2, F6, F3, F4 and F5 in succession. On the basis of these findings, we can offer some recommendations to hotel managers.
Recommendations for Hotel Managers
Offer Brighter Prospects for Hotel Employees
The overall employee satisfaction of the hotel chain is relatively low at 0.5848. If measures are not implemented to correct the situation, the trends already in evidence (i.e., low employee loyalty, high absenteeism, and high turnover) will continue and could, potentially, worsen. In order to improve employee satisfaction, it is recommended that hoteliers explore ways to increase the F1 satisfaction level (development prospects). It is a key determinant of overall employee satisfaction. First of all, the hotel should offer better employee training and a system of merit-based promotion and advancement so that employees can avail themselves of opportunities to improve both their skills and their long-term career prospects. Only when employees have viable long-term advancement prospects will employee satisfaction and loyalty improve. Although a hotel cannot promote all of its employees to higher levels, it seems that offering viable opportunities to its employees could result in a happier and more stable workforce which, in turn, enhances the value of both the product and services of the hotel.
Improve Monetary and Spiritual Rewards
According to the description analysis of the data (Table 1), hotel employees have low satisfaction with regard to both monetary and spiritual return. The vast majority of hotel employees in China receive a low salary compared to other service businesses and this is significant because monetary return is one of the most important drivers contributing to satisfaction. It is therefore recommended that hotel managers should revise their compensation packages to make them more competitive with those of other industries that can lure away hotel employees by offering more lucrative compensation. At the same time, with the improvement of Chinese living standards, people no longer regard their jobs as simply a means of living, but rather as a means of self-fulfillment. It is recommended that hotel operators should also strive to offer spiritual motivation and a sense of pride to employees through image-building campaigns featuring employees as the core of the hotel chain’s success and revitalization efforts. These key steps, together, will greatly improve both employee satisfaction and the brand appeal of the hotel chains.
Eliminate Discrimination from Hotel Management
According to our research, the overall satisfaction of male employees is 0.5802; and that of females is 0.5871, a relatively insignificant gender difference. However, employee satisfaction based on age is quite a different matter. The overall satisfaction of employees aged under 26 is 0.5660; and that of those whose age is over 26 is 0.6708. These show that gender is not a very influential factor in hotel employee satisfaction if employees are at the same age, but age is a remarkable demographic factor in terms of employee satisfaction regardless of gender. When trying to raise the level of employee satisfaction, emphasis should be placed on employee age over gender. In Asian countries, including China, there is a tendency within the hospitality industry, such as hotels and travel services, to employ young, beautiful girls. However, young employees also bring added human resources risks to the enterprise, because in general they are more ambitious, have a higher degree of personal expectations than their older counterparts, are easily dissatisfied, and are more likely to leave the hotel to pursue other opportunities. In comparison, older employees provide greater workforce stability and loyalty. So, it is recommended gender and age discrimination should be eliminated to help hotels build a more balanced and effective workforce based on skill and customer service, rather than just youth and beauty.
Build a Humanitarian Environment for Employees
Hotel employees have frequent personal interactions with various people, including hotel guests, coworkers, and supervisors. Employees seem to experience greater job satisfaction when they perceive that they are closely monitored and constructively led by their supervisors. Therefore, by building an environment of mutual trust, respect, support, helpfulness, and friendliness, hotels can produce a positive effect on employee satisfaction. Furthermore, employees should be empowered by supervisors to do whatever it takes to satisfy customers. Empowerment means management trusts that employees have the skills to arrive at solutions that not only meet and exceed customer needs but also enhance the image, goodwill, and profitability of the hotel chain. In other words, the success of any hotel is built upon a foundation of happy and satisfied employees. Hospitality firms must become sensitive to their employees’ needs and, by helping to bring about a humanitarian, merit-based workplace, help their brand by creating superior customer service.
Conclusion and Limitations
This study based on an ANN examined the factors influencing employee satisfaction as well as their relative levels of importance. The researchers also analyzed the impact of demographic characteristics on hotel employee satisfaction. Study results show that hotel employee satisfaction in China is low; gender and age impact employee satisfaction differently; developmental prospects of both the hotel and individual employees are the most important contributor to hotel employee satisfaction. According to these findings, hotel managers in China should eliminate gender and age discrimination, improve both monetary and spiritual rewards, and offer brighter long-term advancement prospects for hotel employees.
Of course, this study is subject to several limiting conditions. First, the small sample size may cause data bias and skewed analysis. Second, the determination of whether or not the three layer model of an ANN is the best model for this analysis requires further examination. Third, we used a sample from only one hotel chain. This may create problems of generalizability as such findings may be due to unique ownership type, location, and so on. Finally, it would be useful to repeat this study and compare the employee satisfaction differences of two or more different hotel chains to derive a broader sample more representative of the national lodging market.
Appendix
The Program for the Ann Model
net=newff (minmax (G),[19,1],{‘tansig’, ‘logsig’},‘trainbr’); net.trainParam.show=100;
net.trainParam.lr=0.05; net.trainParam.epochs=3000; net.trainParam.goal=1e-3; [net,tr]=train(net, G, g);
T=input(‘input indicators of target employees of the hotel’) ‘Evaluation Result: ’
Result=sim(net,T)
Band, W. (1988). How to gain competitive advantage through customer satisfaction. Sales and Marketing Management, 5, 30-42.
Chinese Tourism Administration Bureau. (2005). Human resources report of China tourism. Retrieved 7 January, 2006, from http://www.cta.gov.cn
George, W. R. (1990). Internal marketing and organizational behavior: A partnership in developing customer conscious employees at every level. Journal of Business Research, 20, 63-70.
Goeldner, C. R., Ritchie, J. R., & McIntosh, R. W. (1999). Tourism: Principles, practices, philosophies (8th ed.). New York: John Wiley & Sons, Inc.
Goode, M. M. H., & Davies, F. (2005). Determining customer satisfaction from mobile phones: A neural network approach. Journal of Marketing Management, 21, 755-778.
Haykin, S. (1999). Neural networks: A comprehensive foundation (2nd ed.). Indianapolis, IN: Pearson Education Inc.
Kent, W. E. (1981). Interrelationships of attitudes, demographics, and job aspects with job satisfaction among hotel managers. Unpublished doctoral dissertation. Georgia State University, Atlanta, GA.
Kornhauser, A. (1933). The technique of measuring employee attitudes. Personnel Journal, 9, 99- 107.
Kramer, A. H., & Sangiovanni-Vincentelli, A. (1989). Efficient parallel learning algorithms for neural network. Advances in Neural Information Processing Systems, 1, 843-848.
Locke, E. A. (1969). What is job satisfaction? Organizational Behavior and Human Performance, 4, 309-336.
Moutinho, L. (1996). The impact of gender on car buyer satisfaction and loyalty. Journal of Retailing and Consumer Services, 3(3), 135-144.
Nunnally, J. C. (1976). Psychometric theory. New York: McGraw-Hill.
Pavesic, D. V., & Brymer, R. A. (1990). Job satisfaction: What’s happening to the young managers? Cornell Hotel and Restaurant Administration Quarterly, February, 90-96.
Pizam, A., & Chandraseker, V. (1983). You do like your work. Cornell Hotel and Restaurant Administration Quarterly, August, 5-10.
Robbins, S. P., & Coulter, M. (1996). Find management. Bellingham, WA: Prentice Hall College Division.
Rogers, J. D., Clow, K. E., & Kash, T. J. (1994). Increasing job satisfaction of service personnel. The Journal of Services Marketing, 8, 14-26.
Roxy, P. (2001). Introduction to statistical data analysis. London, UK: Thomson Learning.
Rumelhart, S. J., & Novig, P. (1986). Artificial intelligence: A modern approach. Upper Saddle River, NJ: Prentice-Hall.
Schmitt, B. (1999). Experiential marketing. Journal of Marketing Management, 15(1), 53-67.
Smith, K. (1996). Becoming an employer of choice: Assessing commitment in the hospitality workplace. An International Journal of Contemporary Hospitality Management, 8(6), 3-9.
Weiss, D., Dawis, D., England, D., & Lofquist, L. (1967). Manual for the Minnesota Satisfaction Questionnaire. Minneapolis: University of Minnesota Press.
Table 1. Mean and SD of Each Variable
Table 2. Effect of the Six Factors on Total Variance
Note: Rotation Method: Varimax with Kaiser Normalization
Table 3. Factor Naming
Figure 1. A Three-Layer Ann Model.
Figure 2. Epochs of the Networks.
Table 4. Employee Satisfaction Change (Output Value)
Both authors are interested in human resource management.
The authors are grateful to Chirs Estrada for his constructive editing of this manuscript. They also extend thanks to Dr. Michael Xie for his excellent research support.
Appreciation is due to reviewers including
Jirong Cai
PhD
Management College
Southwest JiaoTong University
Chendu 610031
China
Darren Lee-Ross
School of Business
James Cook University
P.O. Box 6811
Cairns QLD 4870
Australia
Xizhou Tian, Tourism School, ChongQing Technological & Business University, Nan-an District, ChongQing, China 400067. Phone: +86 23 6276 9331; Fax: +86 23 62769447; Email: [email protected], Yongjian Pu’s email address is: [email protected]