SINEs are short interspersed repeated DNA elements which are considered to spread throughout genomes via RNA intermediates. Polymorphisms with regard to the presence or absence of SINE are occasionally observed in a specific location of a genome. We modeled the evolution of SINEs with regard to this type of polymorphism. Because SINEs are rarely deleted, multiplication of elements is confined to a certain period, and a few master copies are considered to be responsible for their multiplication, the usual population genetic models of transposable elements assuming the equilibrium state are not applicable to describe the evolution of SINEs. Taking into account these properties and assuming selective neutrality, we computed conditional probabilities of finding a SINE at a specific site given that this site is first found because it is occupied by a SINE in an original sample. Using these probabilities, we investigated ways to estimate the multiplication period and infer relationships among populations. The latter inference procedures are shown to be strongly dependent on the multiplication period.