In this paper, the probability properties are investigated for a stochastic SIS epidemic model. Transition probabilities of the susceptible process are obtained by using Laplace transform and perturbation variables. According to two cases: basic reproduction number [Formula: see text] and [Formula: see text], the dynamical behaviors in probability of the process are analyzed. It is shown that when [Formula: see text] the disease-free equilibrium is globally asymptotically stable with probability one, and when [Formula: see text] and [Formula: see text] is a positive integer, the endemic equilibrium is globally asymptotically stable with probability one. These results coincide with the corresponding deterministic SIS epidemic model. However, when [Formula: see text] and [Formula: see text] is not a positive integer, there are different properties between the deterministic and stochastic models. Numerical simulations are also performed to validate these results.