Table 3. Regression model results for the associations between normalized rates of change in improved water and sanitation coverage and socioeconomic indicators.
| Model type, coverage typea | Independent variable | Regression type | Inclusion of countries with armed conflict? | n | Coefficient | SE (95% CI) |
|---|---|---|---|---|---|---|
| Univariate | ||||||
| Water | Povertyb | Linear | Yes | 63 | 0.004 | 0.0018 (0.0004 to 0.0077) |
| Water | Gini coefficient | Linear | No | 27 | 0.015 | 0.0068 (0.0010 to 0.0291) |
| Multivariate | ||||||
| Sanitation | Component 2c | Linear | Yes | 50 | –0.0903 | 0.0449 (–0.1801 to –0.00004) |
| Water | Component 2c | Linear | No | 23 | 0.124 | 0.0573 (0.0048 to 0.2433) |
CI: confidence interval; SE: standard error.
a Only the results for regressions that gave P-values of no greater than 0.05 are shown.
b Proportion of the population with daily income below 1.25 United States dollars.
c Second component obtained from principal components analysis (Table 2).