Fig 1.
District-wise distribution of sample size (left hand side plot) and sample count (right hand side plot) in 2014 BDHS data.
Table 1.
Summary of sample size and sample count in 2014 BDHS data.
Table 2.
Model parameters for the generalised linear models for diarrhoeal prevalence.
Fig 2.
District-wise survey weighted vs unweighted direct estimates of diarrhoea prevalence (%) in Bangladesh.
Fig 3.
District-wise effective sample size vs observed sample size in BDHS 2014 data.
Fig 4.
District-wise effective sample count vs observed sample count in BDHS 2014 data.
Fig 5.
Histograms (left plot), normal q-q plots (centre plot) and distributions of the district-level residuals (right plot).
Fig 6.
Bias diagnostic plot with y = x line (dotted) and regression line (solid) for diarrhoea prevalence (%) in Bangladesh.
Fig 7.
District-wise percentage coefficients of variation (CV, %) for the model-based small area estimate generated by EP (solid line, ●) and the direct estimate (dash line, ○) for the diarrhoea prevalence in Bangladesh.
Districts are arranged in increasing order of CV of direct estimates.
Table 3.
Direct (DIR) and model-based (EP) estimates along with 95% confidence interval (95% CI) and percentage coefficient of variation (CV,%) of the diarrhoea incidence (%) by District in Bangladesh.
Fig 8.
District-wise 95 percentage confidence interval (%95 CI) for the model-based small area estimate generated by EP (solid line, ●) and the direct estimate (dash line, ○) for the diarrhoea prevelance in Bangladesh.
Districts are arranged in increasing order of sample size.
Table 4.
Aggregated level estimates of diarrhoea prevalence (%) generated by direct (DIR) and model-based (EP) method.
Fig 9.
District-wise map showing the spatial distribution of diarrhoea prevalence (%) generated by model-based (EP) method in Bangladesh.