Skip to main content

Table 2 Traditional, two level and three level empty random components models for MCU

From: Application of variance components to the identification of determinants of modern contraceptive use in the Tanzania demographic and health survey data

Parameter

Model 0 (Traditional)

Model 1 (PSU)

Model 2 (Region)

Model 3 (PSU and Region)

\({\updelta }_{\mathrm{u}0}^{2}\)(Variance \({\mathrm{u}}_{0\mathrm{j}}\))

-

0.74 (0.308; 0.566)

 

0.47 (0.308; 0.566)

\({\updelta }_{\mathrm{u}1}^{2}\)(Variance \({\mathrm{u}}_{0\mathrm{k}}\))

-

 

0.35 (0.22; 564)

0.39 (0.267; 0.849)

ICC(Region)

-

 

0.10 (0.063; 0.146)

0.09 (0.058;147)

ICC(PSU)

-

0.18 (0.149; 0.224)

 

0.21 (0.160; 0.261)

AIC

7330.46

6885.68

6961.31

6793.76

BIC

7337.011

6898.78

6974.41

6813.41

Loglikelihood

-3664.23

-3440.84

-3478.66

-3393.88