Model Joint Business Group Based Knowledge for Fishermen Community Empowerment Strategies: Results and Discussion
After analyzing measurement model through confirmatory factor analysis and see that each indicator can be used to define a latent construct, then a full SEM models can be analyzed. Analysis of the data processing stage full SEM models was done by testing the suitability and the test statistic. The results of processing data for a full analysis of SEM models show in
Figure 2. Suitability indices as the model used in the confirmatory factor analysis. SEM model testing was intended to look at the suitability of the model. The results of treatment were presented in Table 1.
The results of Table 1. showed chi square a large (965.292> 381.873) and a probability value below 0.05 was equal to 0.000. In addition, the value of GFI (0.744 <0.9), AGFI (0.693 <0.9), CMIN/DF (2.856> 2.00) and RMSEA (0.102> 0.08) did not meet the requirements. However, TLI (0.832 <0.95) and CFI (0.849 <0.95) were included in the marginal fit. Thus, it can be concluded that the model developed did not meet the eligibility requirements necessary to modify the model so that the model in Figure 3. This is in accordance with the opinion of Ferdinand (2005), that the confirmatory factor analysis. Structural Equation Model testing on modeling studies were also performed with two kinds of test, the suitability model and the significance test of causality through regression coefficient test. Suitability indices as the model used in the confirmatory factor analysis. SEM model testing was intended to look at the suitability of the model.
Figure 2. Full analysis model before modification
Figure 3. Full analysis model after modification
Table 1. The result of feasibility testing model on full model
|Goodness of Fit Index||Cut off Value||Result||EvaluationModel|
|Chi-Square||< 381,873(x2tabel with db: 338 and p:5%)||965.292||Not Good|
|Probability||> 0.05||0.000||Not Good|
|RMSEA||< 0.08||0.102||Not Good|
|GFI||> 0.90||0.744||Not Good|
|AGFI||> 0.90||0.693||Not Good|
|CMIN/DF||< 2.00||2.856||Not Good|