A Study of the Participative Motivation, Satisfaction and Loyalty of the Members at the Taekwondo Training Hall in Taipei County

Abstract

The purpose of this study was to explore the differences among the taekwondo training hall members’ demographic variables as they related to participative motivation, satisfaction, and loyalty. A secondary aim is to verify the cause and effect relationship of participative motivation, satisfaction, and loyalty. For this study, a total of 358 members were selected from 15 taekwondo training halls in Taipei County. The instruments utilized in this research include a participative motivation scale, a satisfaction scale, and a loyalty scale. The data were statistically analyzed utilizing descriptive statistics (including a frequency distribution percentage, the mean and the standard deviation), a t-test, a one-way ANOVA, the scheffe method and structural equation modeling. The results were as follows: (a) As it related to the demographics of the members at the taekwondo training halls in Taipei county, the descriptive statistics indicated that a majority of the members were males between 9-12 years old; their total family income was around NT 40,001~NT 60,000; and a majority of the members had practiced taekwondo for less than one year. (b) The results of the analysis of the member’s demographic variables showed that a member’s gender, age, and time spent learning taekwondo indicated statistically significant differences on his or her participative motivation and satisfaction. A member’s gender, age, family income, and time spent learning taekwondo also indicated statistically significant differences on his or her loyalty. (c) According to the analysis conducted by the structural equation modeling, participative motivation had a positive influence on satisfaction and loyalty, and satisfaction had a positive influence on loyalty. Based on these findings, the researchers have provided some suggestions for taekwondo training halls.

Introduction

Taekwondo was introduced from Korea to Taiwan more than four decades ago. The military has long established a special battle tactic training team that utilizes taekwondo, and the police force has also practiced it in regular training. In addition, most schools have taekwondo clubs on campus, and hundreds of private establishments are scattered throughout the island. Today, there are 737 taekwondo training halls in Taiwan (Chien, 2007). Taekwondo has been an official event in the Olympic Games since 2000. This has been a medal-winning event for the Taiwanese national team members. These factors have led to the acceleration of the development of taekwondo in Taiwan, as evidenced by the establishment of the R.O.C. Taekwondo Association and numerous private taekwondo training halls and taekwondo schools. The outstanding performance of the athletes is a testament to the sport. Therefore, the private taekwondo training halls and taekwondo schools are of great significance to the development of taekwondo in Taiwan (Yeh, 1999).

In recent years, a number of researchers have explored participative motivation in the field of taekwondo. Chen (2001) did research on the participative motivation of taekwondo members, and the results indicated that the major participative motivations were “learning the skill of taekwondo” and “developing a healthy body.” Sung (2005) pointed out that there were four major factors related to the taekwondo training hall members’ participative motivation, (i.e., learning the skill of taekwondo, relieving stress, encouraging self-fulfillment, and developing a healthy body). S. H. Yang (2005) also pointed out that the participative motivation factors of the taekwondo club members at high schools were, based on a factor analysis, health demands, achievement demands, psychological demands, leisure demands, and social demands. Therefore, different researchers have focused on different aspects of participative motivation, and there is no universal, parsimonious, or all encompassing definition. A number of researchers have explored demographic variables related to participative motivation. For example, gender, age, training, family income, and the period of learning have made significant differences to participative motivation (Chen, 2001; Huang, 2005; Sung, 2005; W. C. Yang, 2005).

The degree of a member’s satisfaction plays an important role in determining the quality of a taekwondo training hall. Satisfaction signifies the loyalty of a consumer. Using these four dimensions (i.e., the professional ability of the coach, the relationships of the members, the training hall environment, and the promotion system), Chen (2001) and Sung (2005) made efforts to quantify the level of satisfaction. They developed an instrument to assess satisfaction empirically, which relied on the differences in the scores of consumers’ perceived performances. Additionally, Yang (2006) indicated that significant differences were found in members’ participative satisfaction levels based on their demographic variables. Sung (2005) pointed out that the demographic variables, including gender, age, time spent studying, and the duration of the learning period, were of significance, but there was not a significant difference in family income. W. C. Yang (2005) also pointed out that significant differences were found in training satisfaction based on gender, grade, the number of years they had been training, when they started learning taekwondo, their degree or level, and their sport performance.

Realizing the importance of consumer loyalty, students in the field of recreation and leisure have paid growing attention to loyalty-related issues in recent decades (Gahwiler & Havitz, 1995; Park & Kim, 2000). This section reviewed the development of the exploration of consumer loyalty in the past, especially in the field of sport or recreation and leisure. This study reviewed the development of the exploration of consumer loyalty in the past, especially in the field of recreation and leisure. Most researchers focused only on one dimension of consumer loyalty—either behavioral or attitudinal (Chen, 2005; Cheng & Yuang, 2007; Hung, 2007). Therefore, this study designed four items to measure loyalty.
Both W. C. Yang’s (2005) and Sung’s (2005) results indicated that there was a significant correlation between the members’ participative motivation and their satisfaction with the taekwondo training halls. Hung (2007) indicated that satisfaction and loyalty were positively and significantly correlated. Cheng and Yuang (2007) indicated that participant motive was a significant influence on satisfaction and that satisfaction was a significant influence on loyalty, but that participant motive was not significant to loyalty. Furthermore, this research was designed to determine whether or not the relationship between motivation and loyalty is a suppressed relationship, and if so, whether or not the results indicated a suppressed relationship in the whole model, that is, whether or not the involvement of satisfaction suppressed the influence of the relationship between participative motivation and loyalty. Therefore, this study investigated the causal relationships among taekwondo members’ perceptions of participative motivation, satisfaction, and loyalty. In this study, the researcher identified loyalty as an outcome variable, the consumer perception of participative motivation as an independent variable, and satisfaction as a mediate variable.

Purpose

The purpose of this study was as follows:

1. To understand the relationships among participative motivation, satisfaction, and loyalty.

2. To understand the demographic differences among the participants as they related to participative motivation, satisfaction, and loyalty.

Research Questions

Based on a review of the existing literature, the researcher constructed the following research questions:

1. Were there any significant demographic differences among the participants as they related to participative motivation?

2. Were there any significant demographic differences among the participants as they related to satisfaction?

3. Were there any significant demographic differences among the participants as they related to loyalty?

4. Were there any significant relationships among participative motivation, satisfaction, and loyalty?

Methodology

Subjects and Sampling

The target population included people who participated in the taekwondo training halls in Taipei County. Four hundred questionnaires were distributed to 15 taekwondo training halls randomly selected from a pool of 146 taekwondo training halls. Three hundred and seventy-eight questionnaires were collected. This represented a response rate of 94.5 percent. The survey analysis required a data set without any missing values. Therefore, the questionnaires with missed values were dropped. Three hundred and fifty-eight questionnaires of the questionnaires had no missing values. The return-rate valid questionnaire was 89.5 percent.

Instruments

Three instruments were used in this study to gather information on the participants’ perceptions of participative motivation, satisfaction, and loyalty in the taekwondo training halls, as follows:

1. The Participative Motivation Scale of Participant Taekwondo Training Halls (PMSPTTH), a modified version of Chen’s (2001), Sung’s (2005), and S. H. Yang’s (2005) scale of participative motivation, measured the perception of participative motivation among the participants. It consisted of five primary dimensions (i.e., health demands, achievement demands, psychological demands, leisure demands, and social demands), and the PMSPTTH consisted of 20 items measured by a six-point Likert scale ranging from (1) “Strongly Disagree” to (6) “Strongly Agree”. The validity and reliability of the PMSPTTH was conducted by the LISREL 8.80 software. The result showed that the PMSPTTH passed the extrinsic quality test with the overall fit measures of the model as follows: χ2 = 450.47, df = 165, GFI = .90, RMSEA = .08, SRMR = .05, CFI = .98, NNFI = .98, χ2 / df = 2.73. The test result of the intrinsic fit measures of the model, from the composite reliability of the overall PMSPTTH, was 0.93, revealing high reliability. In terms of the convergent validity, all of the observed variables and latent factors reached “significance.” This indicated that the model had convergent validity (Anderson & Gerbing, 1988). Accordingly, the foregoing findings identified the validity and reliability of the PMSPTTH.

2. Satisfaction was measured by the Satisfaction Scale of Participant Taekwondo Training Halls (SSPTTH), which was a modified version of Sung’s (2005) and Teng’s (2005) satisfaction scales. The satisfaction scale consisted of four primary dimensions (i.e., the professionalism of the coach, the relationships of the members, the training hall environment, and the promotion system). The SSPTTH consisted of 16 items measured by a six-point Likert scale ranging from (1) “Strongly Disagree” to (6) “Strongly Agree.” The validity and reliability of the SSPTTH was conducted by the LISREL 8.80 software. The results showed that the SSPTTH passed the extrinsic quality test with the overall fit measures of the model as follows: χ2 = 255.47, df = 100, GFI = .91, RMSEA = .07, SRMR = .04, CFI = .99, NNFI = .98, χ2 / df = 2.55. The test results of the intrinsic fit measures of the model and the composite reliability of the overall SSPTTH was 0.94, revealing high reliability. In terms of the convergent validity, all of the observed variables and latent factors reached “significance.” This indicated that the model had convergent validity (Anderson & Gerbing, 1988). Accordingly, the foregoing findings identified the validity and reliability of the SSPTTH.

3. Loyalty was measured by the Loyalty Scale of Participants in Taekwondo Training Halls (LSPTTH), which was a modified version of Chen (2005), Cheng and Yuang’s, (2007) and Hung’s (2007) satisfaction scales. The LSPTTH consisted of four items measured by a six-point Likert scale ranging from (1) “Strongly Disagree” to (6) “Strongly Agree.” The validity and reliability of the LSPTTH was tested using LISREL 8.80 software. The results showed that the LSPTTH passed the extrinsic quality test with the overall fit measures of the model as follows: χ2 = 7.03, df = 2, GFI = .99, RMSEA = .08, SRMR = .02, CFI = .99, NNFI = .98, χ2 / df = 3.52. The test results of the intrinsic fit measures of the model from the composite reliability of the overall LSPTTH was 0.85, revealing high reliability. In terms of the convergent validity, all of the observed variables reached “significance.” This indicated that the model had convergent validity (Anderson & Gerbing, 1988). Accordingly, the foregoing findings identified the validity and reliability of the LSPTTH.

Data Analysis

The data were analyzed by the SPSS for Windows 12.0 program and the LISREL 8.80 software. The SPSS for Windows 12.0 program was used to run the descriptive statistics (including the frequency distribution percentage, mean, and standard deviation), a t-test, a one-way ANOVA, and the scheffe method. LISREL 8.80 software was used to run the proposed model. For the assessment of the fit for the proposed model, this study utilized the suggestions of Bagozzi and Yi (1988), Browne and Cudeck (1993), Byrne (1998), Chou and Bentler (1995), Jöreskog and Sörbom (1993), Hair, Anderson, Tatham, & Black (1998) and Kline (1998), which indicated that the proposed model should be evaluated based on the overall model fit as well as the fit of the internal structure.

Results

The descriptive statistics of the demographics of the members in the taekwondo training halls in Taipei county showed that the majority of the members were males between 9-12 years old; their total family income was around NT 40,001~ NT 60,000; the majority of the members have practice taekwondo for less than one year (see Table 1).

Table 1. Descriptive statistics of the demographics

Variables

N

Gender

male

254

70.9

female

104

29.1

Age

9-12

165

46.1

13-15

100

27.9

16-18

 82

22.9

19 or older

 11

 3.1

Family income

NT 40,000 or less

 66

18.4

NT 40,001-60,000

153

42.7

NT 60,001-80,000

109

30.4

NT 80,001 or more

 30

 8.4

Years spent learning Taekwondo

Less than 1

111

31.0

1-2

 89

24.9

2-3

 64

17.9

3-4

 68

19.0

More than 4

 26

 7.3

Note: N=358

The results of the analysis of the members’ demographic variables as they related to participative motivation showed that the females’ participative motivation was higher than the males’ participative motivation. Members 19 or older scored higher than members 9 to 18 years old on participative motivation. Members who had practiced taekwondo for one to three years scored higher on participative motivation than members who had practiced taekwondo for less than one year. The member’s demographic variables as they related to satisfaction showed that females indicated higher satisfaction than males. Members 19 years or older scored higher on satisfaction than members 9 to 18 years old. Members who had practiced taekwondo for one to three years scored higher on satisfaction than members who had practiced taekwondo for less than one year. The member’s demographic variables as they related to loyalty showed that females indicated higher levels of loyalty than males. Members 19 years or older scored higher on loyalty than members 9 to 18 years old. Members with family incomes of NT 80,001 scored higher on loyalty than members with family incomes of 80,000. Members who had practiced taekwondo one to four years scored higher on loyalty than members who had practiced less than one year (see Table 2).

Table 2. Difference tests of member demographic variables as they related to participative motivation, satisfaction and loyalty

Demographics Variables

Participative Motivation

Satisfaction

Loyalty

Gender

(a) Male (n=254)

4.33±.67

3.97±.68

4.39±.85

(b) Female (n=104)

4.49±.73

4.20±.72

4.59±.87

t value

-2.00*

-2.79*

-2.07*

Age

(a) 9-12

4.34±.67

4.00±.69

4.39±.89

(b) 13-15

4.40±.69

3.99±.66

4.49±.89

(c) 16-18

4.32±.56

4.01±.59

4.40±.66

(d) 19 or older

5.04±1.39

4.98±1.19

5.32±1.00

F value

3.91*

7.31*

4.31*

scheffe method

d > a, b, c

d > a, b, c

d > a, b, c

Family income

(a) NT 40,000 or less

4.46±.75

4.09±.80

4.44±.98

(b) NT 40,001-60,000

4.34±.68

4.05±.68

4.27±.80

(c) NT 60,001-80,000

4.29±.62

3.94±.63

4.26±.78

(d) NT 80,001 or more

4.65±.78

4.21±.79

5.04±.89

F value

2.57

1.42

6.88*

scheffe method

d > a, b, c

Years spent learning taekwondo

(a) Less than 1

4.15±.63

3.80±.62

4.10±.77

(b) 1-2

4.57±.59

4.26±.57

4.75±.70

(c) 2-3

4.57±.95

4.21±.97

4.59±1.05

(d) 3-4

4.35±.50

4.01±.62

4.51±.87

(e) More than 4

4.22±.61

3.91±.52

4.38±.66

F value

6.82*

7.11*

8.56*

scheffe method

b, c > a

b, c > a

b, c, d > a

Note: mean ± standard deviation; *p < .05

Model Evaluation

A number of fit indicatives were generated to determine the model that best fit with the observed data. Seven indicatives of fit were examined. First, a chi-square test was conducted to test the fit between the sample covariance matrix and the matrix implied by the model. A large chi-square value and a statistically significant result (p < .05) indicated a poor fit in that there was a substantial proportion of variance in the data not explained by the model. As this statistic was somewhat sensitive to sample size, a second calculation could be made that involved dividing the chi-square value by the degrees of freedom (Kline, 1998). In general, GFI, NNFI and CFI of greater than .90 were considered to be acceptable (Kline, 1998; Bentler & Bonnett, 1980; Tabachnick & Fidell, 1996). SRMR and RMSEA of less than .08 were considered to be acceptable (Browne & Cudeck, 1993). of less than three were considered to be acceptable (Kline, 1998).

Table 3. Model fit statistics

Table 3

Note: GFI= goodness of fit index; SRMR= standardized root mean square residual; RMSEA= root mean square error of approximation; CFI= comparative fit index; NNFI= non-normative fit index.

Figure 1
Figure 1. Completely standardized solution of the model
Note: *P < .05

Table 3 presented the fit indicatives for the hypothesized model: χ2 = 602.00, df = 203; GF I= .90 was greater than .90; SRMR = .06 was less than .08; RMSEA = .06 was less than .08; CFI = .98 was greater than .90; NNFI = .97 was greater than .90; = 2.97 were below 3. These results indicated that the hypothesized model fit the empirical data very well (see Figure 1).

When the full model is acceptable, then the quality of measurement and structural models can be examined. The factor loadings for the participative motivation indicators (λ1~λ5) were greater than 0.56 and significant at p < .05, indicating convergent validity for participative motivation (Anderson & Gerbing, 1988). Factor loadings for the satisfaction indicators (λ6~λ9) were greater than 0.65 and significant at p < .05, indicating convergent validity for satisfaction. Factor loadings for the loyalty indicators (λ10~λ13) were greater than 0.70 and significant at p < .05, indicating convergent validity for loyalty.

The structural path coefficients were used to test the direct and indirect effects of the theoretical considerations of this study. The participative motivation had a positive and direct influence on loyalty (γ = .42, t = 5.93, p < .05); the participative motivation had a positive and direct influence on satisfaction (γ = .61, t = 10.50, p < .05); satisfaction had a positive and direct influence on loyalty (γ = .31, t = 4.42, p < .05). Finally, the indirect influence of participative motivation to satisfaction to loyalty was 0.19 (.61*.31 = .19, t = 4.25, p < .05).

Discussion

The evidence in this study reported that the demographics of the taekwondo members in Taipei County indicated that the ratio of the male members was higher than the female members. Additionally, the majority of the members were between 9 and 12 years old. Also, most of the members were in elementary schools. The results supported the previous finding about members learning taekwondo (Chen, 2001; Teng, 2005; Sung, 2005; C. H. Yang, 2006). Among the members, the total family income was around NT 40,001~ NT 60,000. The results supported Teng (2005) and C. H. Yang (2005), but not Chen (2001). The reason for this may have been the living standards in the different counties and cities. Finally, the majority of the members had practiced taekwondo for less than one year, which was the same time period in the studies by Teng (2005) and Sung (2005).

The results indicated that the respondents’ demographic variables showed that female students 19 years or older that had practiced taekwondo for one to three years had higher perceived motivation than the other groups. Additionally, the discoveries in this study were similar to other related research in the papers of Chen (2001), Huang (2005), Sung (2005) and W. C. Yang (2005).
Different demographic variable tests related to satisfaction have found that females score higher than males on satisfaction. These findings are similar to related research (W. C. Yang, 2005) that has explored gender as the main factor of training satisfaction in taekwondo. These results were different from those found in the studies of Sung (2005) and Teng (2005). Both found that males had higher satisfaction than females. The subjects in those studies were participating in taekwondo training halls in Kaohsiung County and Taipei County. In addition, members 19 or older scored higher on satisfaction than 9 to 18 year old members. Table 1 showed that members 19 or older scored lower numbers than other groups. Table 2 showed that this group had more motivation, satisfaction, and loyalty than other groups. Therefore, this group belongs to higher motivation, satisfaction, and loyalty. Members who had practiced taekwondo for one to three years scored higher on satisfaction than members who had practiced taekwondo for less than one year. The results were consistent to C. Yang’s (2005) and Yang’s (2006) research.

Females scored higher than males on loyalty. This result was different than that found in the studies of Chen, Chang, and Li (2005), which explored consumer loyalty in indoor heated swimming pools and Huang (2006), which explored health and fitness club members. The results were no different as they related to gender. So, each service industry had different results. Members 19 or older scored higher on loyalty than members 9 to 18 years old. Members with family incomes of NT 80,001 or more scored higher on loyalty than members with family incomes of 80,000 or less. Members who had practiced taekwondo for one to four years scored higher on loyalty than members who had practiced taekwondo for one year.

The results of the structural equation modeling showed that participative motivation had a positive influence on satisfaction and loyalty and that satisfaction had a positive influence on loyalty. These results confirmed the findings of previous studies indicating a strong relationship among participative motivation, satisfaction, and loyalty in taekwondo activities (Hung, 2007; Sung, 2005; W. C. Yang, 2005).

Suggestions

1. The target market for the taekwondo training halls were 9-12 years olds. Huang (2005) indicated that the priority of participation motivation among taekwondo members was exercising to become healthier. It also pointed out that the first participation motivation was the factor of body needs, then knowledge needs and achievement needs. Therefore, the taekwondo programs need to enhance their related fitness activities, and try to promote the health of their members.

2. The coaches need to realize that member’s traits (i.e., gender, age) are relevant. Nineteen year old females have more motivation, satisfaction, and loyalty than other groups. The coaches need to maintain their movement interest and strengthen other member groups; for example, they could increase the degree of difficulty for the male members in order to create challenge, and coaches could offer certificates to encourage interest and a sense of achievement.

References

References

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411-423.

Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16, 74-94.

Bentler, P. M., & Bonnett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606.

Brown, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing Structural Equation Models (pp. 445-455). Newbury Park. CA: Sage.

Byrne, B. M. (1998). Structural Equation Modeling with LISREL, PRELIS and SIMPLIS: Basic Concepts, Applications and Programming. Mahwah, NJ: Lawrence Erlbaum Associates.

Chen, C. C. (2001). Research on the learning engagement of the taekwondo members in the training hall in Yunlin county. Unpublished master thesis, National Taiwan Sport University, Taichang campus, Taiwan.

Chen, W. C. (2005). A study of indoor swimming pool relationships among consumer involvement, satisfaction and consumer loyalty: Using the Kaohsiung and Pingtung regions as examples. Journal of Physical Education Fu Jen Catholic University, 4, 81-98.

Chen, W. C., Chang, C. M., & Li, Y. M. (2005). A study of consumer loyalty in indoor heated swimming pools. NCKU Sport, 38(3), 67-81.

Cheng, H. C. & Yuang, H. F. (2007). Research on badminton building consumer participant behavior in south Taiwan. Journal of Leisure and Tourism Industry Research, 2(1), 118-132.

Chien, T. M. (2007). A case of study of operative strategies in a taekwondo training hall. Unpublished master thesis, National Taiwan Normal University, Taiwan.

Chou, C. P., & Bentler, P. M. (1995). Estimates and tests in structural equation modeling.In R. H. Hoyle(Ed.), Structural Equation Modeling: Concepts, Issues, and Applications (pp. 37-55). Sage CA: Thousands Oaks.

Gahwiler, P., & Havitz, M. (1995). Toward a relational understanding of leisure social worlds, involvement, psychological commitment, and behavioral loyalty. Journal of Park and Recreation Administration, 13(2), 1-23.

Hair, J.F. Jr., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5th ed.). UK: Prentice Hall International.

Hsu, J. Y. (2003). The study of elite taekwondo athletes’ perceptions on coaching leadership behavior and satisfaction. Chinese Culture University, Taiwan.

Huang, M. C. (2005). A study on the participative motives of the members of a taekwondo club in elementary school. Unpublished master thesis, Asia University, Taiwan.

Huang, Y. H. (2006). A study on participating motivation, customer satisfaction, and brand loyalty of health and fitness club members in Taichung City. Sports Industries Management Conference, 5, 75-87.

Hung, J. M. (2007).The structural equation model of quality of service, members’ satisfaction and loyalty of fitness center in Tainan. Taipei Physical Education College Press, 15, 193-203.

Jöreskog, K. G., & Sörbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language.Chicago: Scientific Software International.

Kline, R. B. (1998). Principles and practice of structural equation modeling. New York: The Guilford Press.

Park, S. H., & Kim, Y. M. (2000). Conceptualizing and measuring the attitudinal loyalty construct in recreational sport contexts. Journal of Sport Management, 14, 197-207.

Sung, M. H. (2005). A study of participative behavior, participation motivation and satisfaction of taekwondo training hall members. Unpublished master thesis, NationalPingtung University of Education, Taiwan.

Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York: HarperCollins College Publishers.

Teng, T. Y. (2005). A study on the satisfaction levels of taekwondo club members – using Taipei County as an example. Fu-Jen Catholic University Graduate Institute of Physical Education.

Yang, C. H. (2006). Research on the participating motivation and satisfaction of the taekwondo members in the training hall in Chang-Hua County. Unpublished master thesis, Dayeh University, Taiwan.

Yang, S. H. (2005). A study on the participation motivation of the taekwondo club members at high schools in Nanou county. Unpublished master thesis, Asia University, Taiwan.

Yang, W. C. (2005). A study of the participation motivation and training satisfaction of taekwondo athletes in elementary school in Taichung Country. Unpublished master thesis, Taipei physical Education College, Taiwan.

Yeh, C. H. (1999). The study of taekwondo’s course of development in Taiwan. Unpublished master thesis, Taiwan Normal University, Taiwan.

Print Friendly, PDF & Email