In the initial phase of a complex emergency, an immediate population size assessment method, based on area sampling, is vital to provide relief workers with a rapid population estimate in refugee camps. In the past decade, the method has been progressively improved, six examples are presented in this paper and questions raised about its statistical validity as well as important issues for further research. There are two stages. The first is to map the camp by registering all of its coordinates. In the second stage, the total camp population is estimated by counting the population living in a limited number of square blocks of known surface area, and by extrapolating average population calculated per block to the total camp surface. In six camps selected in Asia and Africa, between 1992 and 1994, population figures were estimated within one to two days. After measuring all external limits, surfaces were calculated and ranged between 121,300 and 2, 770, 000 square metres. ln five camps, the mean average population per square was obtained using blocks 25 by 25 metres (625m(2)) and for another camp with blocks 100 by 100m(2). In three camps, different population density zones were defined Total camp populations obtained were 16,800 to 113,600. Although this method is a valuable public health tool in emergency situations, it has several limitations. Issues related to population density and number and size of blocks to be selected require further research for the method to be better validated.
