Census counts are subject to different types of nonsampling errors. One of these main errors is coverage error. Undercount and overcount are two types of coverage error. Undercount usually occurs more than the other, thus net undercount estimation is important. There are various methods for estimating the coverage error in censuses. One of these methods is dual system (DS) that usually uses data from the census and a post-enumeration survey (PES). In this paper, the coverage error and necessity of its evaluation, PES design and DS method are explained. Then PES associated approaches and their effects on DS estimation are illustrated and these approaches are compared. Finally, we explain the Statistical Center of Iran method of estimating net undercount in Iran 2006 population and dwelling census and a suggestion will be given for improving net undercount estimation in population and dwelling censuses of Iran.