Table 1.
Regression coefficients and 95% credible intervals for explanatory variables of net use conditional on access in betabinomial regression models with spatial random effects
| Effect | 2014 | 2016 | ||
|---|---|---|---|---|
| Exponentiated coefficient | 95% credible interval | Exponentiated coefficient | 95% credible interval | |
| Wealth (ref: poorest) | ||||
| Poorer | 0.99 | (0.82–1.20) | 0.94 | (0.74–1.20) |
| Middle | 0.77* | (0.62–0.96)* | 0.69* | (0.52–0.91)* |
| Richer | 0.55* | (0.43–0.71)* | 0.60* | (0.45–0.81)* |
| Richest | 0.48* | (0.36–0.64)* | 0.44* | (0.32–0.61)* |
| Number of nets per person | 1.02 | (0.89–1.16) | 1.07 | (0.90–1.27) |
| Humidex | 1.00 | (0.99–1.00) | 1.00 | (0.99–1.01) |
| Messaging | 2.53* | (1.52–4.22)* | 1.16 | (0.75–1.83) |
| Setting (ref: urban) | 2.54* | (1.51–4.31)* | 1.59* | (1.08–2.33)* |
| IRS in last 12 months | 0.76 | (0.36–1.59) | 0.82 | (0.45–1.50) |
| Electricity in household | 0.84 | (0.52–1.36) | 0.76 | (0.53–1.08) |
| Messaging × rural | 0.68 | (0.42–1.10) | 1.19 | (0.87–1.65) |
| Messaging × IRS | 0.76 | (0.42–1.36) | 1.12 | (0.72–1.75) |
| Messaging × electricity | 0.94 | (0.60–1.47) | 1.03 | (0.68–1.57) |
| Rural × IRS | 1.15 | (0.76–1.74) | 1.09 | (0.67–1.76) |
| Rural × electricity | 0.79 | (0.58–1.07) | 0.85 | (0.57–1.26) |
| IRS × electricity | 1.06 | (0.74–1.52) | 0.77 | (0.47–1.25) |
× Indicates interaction term
*Statistically significant (p ≤ 0.05)