Significance Test for Observed Variables

86 order to convince that the observed variables have significant contribution to the land use change happen in Siak District, the procedure of likelihood ratio tests have been done in order to check the contribution of each effect to the land use change model. Table 20. Likelihood Ratio Tests for Observed Variables Likelihood Ratio Tests 43614.131 1788.275 25 .000 49206.686 7380.831 25 .000 42221.697 395.841 25 .000 42168.217 342.362 25 .000 42353.022 527.167 25 .000 42027.977 202.121 25 .000 Effect Intercept CROPDIST02 ROADDIST KWSID RTRWPID CONCESSID -2 Log Likelihood of Reduced Model Model Fitting Criteria Chi-Square df Sig. Likelihood Ratio Tests The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model. The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0. The result of likelihood ratio tests for each observed variable done in MLR model analysis shows that the significance levels of the observed variables tested are less than 0.05, and in other word all observed variables may be considered as significant variables of land use change in Siak District and would be included into the final model. The list of significant variables would be included into the model is shown in Table 20.

4.4.3.3 Significance Test for Final Model 2

nd Scenario There were two tests which have been done in order to determine whether the final model were adequate to explain the land use change happen in Siak District or not. The tests done were Likelihood Ratio Test of the final model and Pseudo R-Square statistics. Based on the likelihood ratio test of the final model, the significance level produced is less than 0.05 Sig.0.05 which may be concluded that the final model which has been produced was outperforming the Null. In other words, the land use change model of Siak District that has been developed by considering the observed variables could be concluded as a good fit model. 87 Table 21. Likelihood Ratio Test of the Final Model 2 nd Scenario Model Fitting Information 53960.379 41825.856 12134.524 125 .000 Model Intercept Only Final -2 Log Likelihood Model Fitting Criteria Chi-Square df Sig. Likelihood Ratio Tests Table 22. Pseudo R-Square Statistics 2 nd Scenario Pseudo R-Square .703 .706 .225 Cox and Snell Nagelkerke McFadden Furthermore, the pseudo r-squared statistics, either Cox and Snell or Nagelkerke, indicate that the variability of land use change in Siak District can be explained more than 70 by the observed variables, whereas according to McFadden, it is only 22.5. However, in general the pseudo r-squared statistics that have been produced shows that most of variability in land use change happen in Siak District could be explained by the final model which was developed by using the observed variables in the research site. The two tests for final model which have been conducted above indicate that the final model of land use change in Siak District which has been developed by considering the observed variables in MLR model is a good model that can explain most of the variability of land use change in the research site. The observed variables that have been determined can be concluded as driving factors of land use transitions in Siak District.

4.4.3.4 Model Validation 2

nd Scenario The coefficient of the observed variables Appendix 6 and spatial data layersparameters 2005 x i have been simulated on MLR model in order to produce the conditional probability maps of land use transitions during 2005 – 2008. The computation of coefficient of the observed variables β i and spatial data layersparameters 2005 x i on MLR model equation produced 26 conditional