Home / Tourist Destination

Evaluating the Fit of Theoretical Models Using Structural Equation Modeling (Sem)

Table 4.22 Estimated values ​​of the model's relationship in confirmatory factor analysis (CFA) of independent factor scales.


Relationship



Not standardized



Standardized



Weight

SE

CR

P-value

Weight

HDTN1

<---

HDTN

1




0.891

HDTN2

<---

HDTN

0.894

0.043

20,696

***

0.811

HDTN3

<---

HDTN

0.834

0.041

20,191

***

0.797

HDTN4

<---

HDTN

0.858

0.043

20,048

***

0.861

HDKDDL1

<---

HDKDDL

1




0.965

HDKDDL2

<---

HDKDDL

0.996

0.02

48,811

***

0.97

HDKDDL3

<---

HDKDDL

0.663

0.041

16,273

***

0.626

HDKDDL4

<---

HDKDDL

0.811

0.031

26,533

***

0.808

MTDD1

<---

MTDD

1




0.915

MTDD2

<---

MTDD

0.912

0.037

24,896

***

0.835

MTDD3

<---

MTDD

0.881

0.036

24,771

***

0.833

MTDD4

<---

MTDD

0.961

0.037

26,197

***

0.857

MTDD5

<---

MTDD

0.851

0.038

22,227

***

0.792

PTSP1

<---

PTSP

1




0.707

PTSP2

<---

PTSP

1,283

0.072

17,777

***

0.901

PTSP3

<---

PTSP

1,232

0.074

16,578

***

0.831

PTSP4

<---

PTSP

1,141

0.066

17,163

***

0.863

PTSP5

<---

PTSP

0.885

0.067

13,223

***

0.661

THKDL1

<---

THKDL

1




0.855

THKDL2

<---

THKDL

0.996

0.042

23,832

***

0.875

THKDL3

<---

THKDL

0.99

0.044

22,321

***

0.845

THKDL4

<---

THKDL

0.945

0.043

21,884

***

0.813

THKDL5

<---

THKDL

0.798

0.04

20,174

***

0.774

SKDD1

<---

SKDD

1




0.897

SKDD2

<---

SKDD

0.843

0.043

19,479

***

0.766

SKDD3

<---

SKDD

0.966

0.042

23,048

***

0.86

SKDD4

<---

SKDD

0.738

0.049

14,971

***

0.717

XDTH1

<---

XDTH

1




0.711

XDTH2

<---

XDTH

1,019

0.062

16,424

***

0.713

XDTH3

<---

XDTH

1,326

0.074

17.86

***

0.947

Maybe you are interested!

Relationship



Not standardized



Standardized




Weight

SE

CR

P-value

Weight

XDTH4

<---

XDTH

1.14

0.068

16,758

***

0.825

XDTH5

<---

XDTH

0.932

0.063

14,901

***

0.734

NCKDL2

<---

NCKDL

1




0.575

NCKDL3

<---

NCKDL

1,552

0.114

13,609

***

0.852

NCKDL4

<---

NCKDL

1,785

0.125

14,233

***

0.995

NCKDL5

<---

NCKDL

1,317

0.089

14,774

***

0.804

HDLS1

<---

HDLS

1




0.826

HDLS2

<---

HDLS

1,165

0.048

24,085

***

0.988

HDLS3

<---

HDLS

0.563

0.044

12.74

***

0.506

HDLS4

<---

HDLS

0.904

0.045

20,113

***

0.786

QLDD1

<---

QLDD

1




0.872

QLDD2

<---

QLDD

0.941

0.049

19,304

***

0.774

QLDD3

<---

QLDD

0.957

0.046

20,903

***

0.817

QLDD4

<---

QLDD

0.868

0.045

19,175

***

0.77

NLCT2

<---

NLCT

1




0.904

NLCT3

<---

NLCT

0.982

0.036

27,535

***

0.882

NLCT4

<---

NLCT

0.963

0.035

27,638

***

0.883

NLCT5

<---

NLCT

0.942

0.038

24,508

***

0.834

NNL1

<---

NNL

1




0.643

NNL5

<---

NNL

1,239

0.131

9.48

***

0.647

NNL6

<---

NNL

1,166

0.122

9,563

***

0.662

KCHT1

<---

KCHT

1




0.609

KCHT2

<---

KCHT

1,451

0.116

12,531

***

0.783

KCHT3

<---

KCHT

1,208

0.089

13,613

***

0.7

KCHT4

<---

KCHT

1,455

0.112

12,978

***

0.838

NNL4

<---

NNL

0.871

0.099

8,762

***

0.564


Source: Calculation results from a direct survey of 450 tourists in Bac Lieu in 2018.

Note: ***: Significance level 1%..

4.4.4 Structural Equation Modeling (SEM)

4.4.4.1 Assessment of the suitability of the theoretical model using structural equation modeling (SEM)

After CFA analysis, the study used SEM structural model to determine the influencing factors and the level of influence of the component factors on the aspects of enhancing destination competitiveness. SEM analysis was conducted starting from the initially proposed research model, then the model was adjusted to obtain a better model. Using linear structural model (SEM) to test the proposed research model.

In testing hypotheses and research models, the SEM linear structural model has many advantages over traditional multivariate analysis methods such as multiple regression and multivariate regression because it can calculate measurement errors and estimate research hypotheses as well as complex research models.

Results of testing the research model using SEM for the first time:

The results of the first SEM research model testing show that:

- The model has 1420 degrees of freedom (df = 1420).

- The model is suitable for market data, because:

+ GFI = 0.849; TLI = 0.943 (> 0.9); and CFI = 0.948 (> 0.9);

+ RMSEA = 0.038 (< 0.08);

+ CMIN/df = 1,646 (< 2).

Table 4.23 Estimated values ​​of the model relationships in the first SEM analysis.


Relationship

Not standardized

SECR P-value weight


MTDD

<---

NCKDL

0.302

0.064

4,752

***

MTDD

<---

PTSP

0.463

0.069

6,749

***

MTDD

<---

XDTH

0.193

0.058

3,356

***

THKDL

<---

HDTN

0.344

0.046

7,435

***

THKDL

<---

HDLS

-0.043

0.04

-1,087

0.277

THKDL

<---

SKDD

0.344

0.049

7.08

***

QLDD

<---

HDKDDL

0.157

0.036

4,329

***

QLDD

<---

KCHT

0.844

0.098

8,576

***

QLDD

<---

NNL

0.056

0.063

0.891

0.373

NLCT

<---

QLDD

0.089

0.063

1,396

0.163

NLCT

<---

THKDL

0.178

0.052

3,404

***

NLCT

<---

MTDD

0.248

0.05

4,972

***

Source: Calculation results from a direct survey of 450 tourists in Bac Lieu in 2018.

Note: ***: 1% significance level.


Source: Calculation results from a direct survey of 450 tourists in Bac Lieu in 2018.

Figure 4.10 Model testing results.


153

- The unstandardized weights are all statistically significant (p-value) at 1%, except for the weights of the NNL factor and the HDLS factor which are not statistically significant at p-value = 1%. Therefore, we will remove these factors one by one to run the SEM test again for the second time. Specifically, when running the SEM test for the second time, we will remove the NNL factor.

Results of testing the research model using SEM for the second time:

The results of the second SEM test of the research model show that:

- The model has 1219 degrees of freedom (df = 1219).

- The model is suitable for market data, because:

+ GFI = 0.855; TLI = 0.946 (> 0.9); and CFI = 0.950 (> 0.9);

+ RMSEA = 0.039 (< 0.08);

+ CMIN/df = 1,699 (< 2).

Table 4.24 Estimated values ​​of the model relationships in the second SEM analysis.


Relationship

Not standardized

SECR P-value weight


MTDD

<---

NCKDL

0.302

0.064

4,751

***

MTDD

<---

PTSP

0.463

0.069

6.75

***

MTDD

<---

XDTH

0.193

0.058

3,357

***

THKDL

<---

HDTN

0.343

0.046

7,424

***

THKDL

<---

HDLS

-0.043

0.04

-1,086

0.277

THKDL

<---

SKDD

0.344

0.049

7,077

***

QLDD

<---

HDKDDL

0.156

0.036

4,316

***

QLDD

<---

KCHT

0.848

0.099

8,593

***

NLCT

<---

QLDD

0.089

0.063

1.4

0.162

NLCT

<---

THKDL

0.178

0.052

3,403

***

NLCT

<---

MTDD

0.248

0.05

4.97

***

Source: Calculation results from a direct survey of 450 tourists in Bac Lieu in 2018.

Note: ***: 1% significance level.

- The unstandardized weights are all statistically significant (p-value) at 1%, except for the weight of the HDLS factor which is not statistically significant at p-value = 1%. Therefore, when running the third SEM test, the HDLS factor will be eliminated.


154



Source: Calculation results from a direct survey of 450 tourists in Bac Lieu in 2018.

Figure 4.11 Results of testing the research model using SEM (standardized) for the second time.



155


see:

Results of the third (final) SEM research model testing:

The results of the third (final) SEM test of the research model show that


- The model has 1034 degrees of freedom (df = 1034).

- The model is suitable for market data, because:

+ GFI = 0.868; TLI = 0.954 (> 0.9); and CFI = 0.958 (> 0.9);

+ RMSEA = 0.038 (< 0.08);

+ CMIN/df = 1,654 (< 2).

Table 4.25 Estimated values ​​of the model relationships in the third (final) SEM analysis.



Not standardized


Standardized

Relationship




Weight

number

SE

CR

P-

value

Weight

MTDD

<---

NCKDL

0.302

0.064

4,746

***

0.228

MTDD

<---

PTSP

0.46

0.069

6,717

***

0.337

MTDD

<---

XDTH

0.193

0.058

3,349

***

0.155

THKDL

<---

HDTN

0.339

0.046

7.36

***

0.358

THKDL

<---

SKDD

0.341

0.049

7,023

***

0.34

QLDD

<---

HDKDDL

0.185

0.042

4,432

***

0.211

QLDD

<---

KCHT

1,021

0.112

9,118

***

0.593

NLCT

<---

QLDD

0.173

0.057

3,052

***

0.171

NLCT

<---

THKDL

0.173

0.051

3,373

***

0.167

NLCT

<---

MTDD

0.198

0.049

4,065

***

0.196

Source: Calculation results from a direct survey of 450 tourists in Bac Lieu in 2018.

Note: ***: 1% significance level.


156

Comment


Agree Privacy Policy *