Table 4. Pooled multivariable regression estimates of the association between wealth and EHA access, across different definitions of the wealth index; for countries with multiple rounds (Panel A) and latest data from all countries (Panel B).
| VARIABLES | Piped water | Improved water | Improved sanitation | Improved fuel | Electricity | Bed net | Mobile phone |
|---|---|---|---|---|---|---|---|
| PANEL A: 29 countries with more than 1 round since 2000 | |||||||
| Country-specific measures | |||||||
| Wealth quintile—all assets from | 0.035*** | 0.033*** | 0.076*** | 0.044*** | 0.085*** | 0.035*** | 0.091*** |
| latest survey round | (0.0043) | (0.0050) | (0.0092) | (0.0076) | (0.0099) | (0.0065) | (0.0080) |
| Wealth quintile—all assets from | 0.034*** | 0.034*** | 0.073*** | 0.042*** | 0.084*** | 0.033*** | 0.088*** |
| original survey round | (0.0043) | (0.0050) | (0.0087) | (0.0074) | (0.0097) | (0.0059) | (0.0079) |
| Global measure | |||||||
| Wealth quintile–only assets | 0.041*** | 0.041*** | 0.070*** | 0.040*** | 0.090*** | 0.031*** | 0.091*** |
| included for all countries | (0.0051) | (0.0047) | (0.0073) | (0.0087) | (0.0063) | (0.0045) | (0.0060) |
| PANEL B: All 41 countries with DHS between 2008–2013 | |||||||
| Country-specific measures | |||||||
| Wealth quintile—all assets | 0.036*** | 0.037** | 0.066*** | 0.049*** | 0.086*** | 0.023*** | 0.10*** |
| (0.0042) | (0.0040) | (0.0074) | (0.0086) | (0.0085) | (0.0060) | (0.0049) | |
| Wealth quintile–only assets | 0.032*** | 0.036** | 0.061*** | 0.043*** | 0.080*** | 0.019*** | 0.090*** |
| included for 75% of countries | (0.0039) | (0.0041) | (0.0074) | (0.0085) | (0.0073) | (0.0054) | (0.0050) |
| Wealth quintile–only assets | 0.033*** | 0.037** | 0.059*** | 0.040*** | 0.082*** | 0.024*** | 0.092*** |
| included for all countries | (0.0038) | (0.0030) | (0.0069) | (0.0076) | (0.0060) | (0.0050) | (0.0042) |
| Global measure | |||||||
| Wealth quintile–only assets | 0.041*** | 0.060*** | 0.062*** | 0.038*** | 0.11*** | 0.015*** | 0.10*** |
| included for all countries | (0.0062) | (0.0047) | (0.0065) | (0.0069) | (0.0082) | (0.0045) | (0.0048) |
Standard errors clustered at the country level are shown in parentheses, models include year of survey and region fixed effects (Panel A) or year of survey and country fixed effects (Panel B). The wealth indices were constructed using the first principal component obtained using PCA over different subsets of asset variables, only excluding the outcome variables, as described in the text. Significance of the coefficients is indicated as follows
*** p<0.01
** p<0.05
* p<0.1.