Unequal probability sampling improves national death surveillance site selection.
The researchers compared cluster and unequal probability sampling methods for selecting death surveillance sites. Unequal probability sampling was found to be more suitable as it can cover a wider population and provide more accurate mortality estimates for the entire province or municipality. Unequal probability stratified cluster sampling had slightly lower design effects compared to equal probability stratified cluster sampling. This method can help in obtaining more monitoring data and better represent the population for death surveillance.