Multilevel binary logistic regression analysis of predictors towards accessibility of sanitation facilities in Ethiopia, EDHS 2011
Variables | Model 0 (null model) | Model 1 AOR (95% CI) | Model 2 AOR (95% CI) | Model 3 AOR (95% CI) |
Individual-level factors | ||||
Sex of HHH | ||||
Male | 0.97 (0.82, 1.04) | 1.01 (0.84,1.11) | ||
Female | 1 | 1 | 1 | |
Age of HHH | ||||
<30 | 0.97 (0.94,1.01) | 0.96 (0.20, 2.34) | ||
30–40 | 0.57 (0.35, 2.02) | 1.49 (0.13, 2.41) | ||
41–54 | 0.43 (0.22, 0.61) | 1.17 (0.86, 1.59) | ||
>54 | 1 | 1 | 1 | |
Educational status | ||||
No education | 1 | 1 | ||
Primary | 3.23 (2.63, 3.96)** | 2.43 (2.00, 2.95)** | ||
Secondary | 6.36 (5.26, 7.67)** | 2.02 (1.61, 2.54)** | ||
Higher | 8.11 (7.16, 9.19)* | 4.12 (3.35, 7.54)** | ||
Wealth index | ||||
Poor | 1 | 1 | ||
Middle | 2.96 (2.34, 4.24)** | 1.49 (1.21, 1.83)** | ||
Rich | 5.48 (3.45, 5.89)** | 3.15 (2.55, 3.89)** | ||
Having television | ||||
No | 1 | 1 | ||
Yes | 4.81 (4.16, 5.56)** | 5.49 (4.37, 6.89)** | ||
Community-level factors | ||||
Community-level education | ||||
Higher | 6.50 (5.82, 7.27)** | 3.90 (3.15, 4.82)** | ||
Lower | 1 | 1 | ||
Community-level media exposure | ||||
Exposed | 6.07 (5.42, 6.81)** | 5.61 (3.84, 10.09)** | ||
Unexposed | 1 | 1 | ||
Type of residence | ||||
Urban | 16.74 (11.85, 23.65)** | 9.08 (6.69, 12.33)** | ||
Rural | 1 | 1 | ||
Region | ||||
Tigray | 0.50 (0.41, 0.62)** | 0.46 (0.37, 0.57)** | ||
Afar | 0.86 (0.67,1.09) | 0.67 (0.51, 0.86)* | ||
Amhara | 0.43 (0.35, 0.53)** | 0.40 (0.32,0.49)** | ||
Oromia | 0.89 (0.72, 1.09) | 0.90 (0.73, 1.12) | ||
SNNRP | 0.25 (0.20, 0.31)* | 0.28 (0.22, 0.34)** | ||
Somali | 0.33 (0.26, 0.41)** | 0.23 (0.18, 0.29)** | ||
Benishangul Gumuz | 0.74 (0.59, 0.94)* | 0.63 (0.49, 0.80)** | ||
Gambella | 0.51 (0.41, 0.64)** | 0.45 (0.35, 0.57)** | ||
Harari | 0.64 (0.52, 0.80)** | |||
Addis Ababa | 1 | 1 | ||
Dire Dawa | 0.25 (0.21, 0.31)** | 0.28 (0.22, 0.35)** | ||
VIF | 2.35 | 2.27 | 2.05 |
1= reference.
**P value < 0.001(Adjusted OR), *P value < 0.05(Adjusted OR).
Model 1 is adjusted for individuallevel variables. Model 2 is adjusted for community-level variables; Model 3 is the final model adjusted for both individual- and community-level predictors.
AOR, adjusted OR; EDHS, Ethiopian Demographic and Health Survey; HHH, household head; VIF, variance inflation factor.