SEARCH

SEARCH BY CITATION

Keywords:

  • epidemiology;
  • humanitarian emergency;
  • war

ABSTRACT Obtaining quality data in a timely manner from humanitarian emergencies is inherently difficult. Conditions of war, famine, population displacement, and other humanitarian disasters, cause limitations in the ability to widely survey. These limitations hold the potential to introduce fatal biases into study results. The cluster sample method is the most frequently used technique to draw a representative sample in these types of scenarios. A recent study utilizing the cluster sample method to estimate the number of excess deaths due to the invasion of Iraq has generated much controversy and confusion about this sampling technique. Although subject to certain intrinsic limitations, cluster sampling allows researchers to utilize statistical methods to draw inferences regarding entire populations when data gathering would otherwise be impossible.