Table 4.
Random effects (ordered) probit estimates for the effect of household cooking fuels on women’s health in rural areas: CFPS 2010–2012.
| Variable | Random Effects Probit (1) Chronic/Acute Disease | Random Effects Ordered Probit (2) Self-Reported Health (Excellent) |
|---|---|---|
| Coal | −0.006 | 0.023 ** |
| (0.020) | (0.010) | |
| 95% CI | [−0.046, 0.034] | [0.003, 0.044] |
| LPG | −0.052 *** | 0.020 *** |
| (0.015) | (0.007) | |
| 95% CI | [−0.081, −0.024] | [0.006, 0.035] |
| N | 9770 | 10,000 |
Notes: The dependent variable is either a dummy for whether the respondent has suffered from a chronic or acute illness (1 = yes, 0 = no) or a 5-point measure of self-reported health (SRH, from 1 = poor to 5 = excellent). The controls are domestic cooking fuel (0 = wood/straw, 1 = coal, 2 = liquefied petroleum gas (LPG), with wood/straw as the referent), age, age squared, household income and household size. Also reported are marginal effects, which for SRH indicate the probability of excellent health. Standard errors are in parentheses and 95% confidence intervals (CI) are in brackets. * p < 0.1, ** p < 0.05, *** p < 0.01.