Since the universe is endowed with small irregularities at all kinds of different length scales, they constitute a portion of energy density and pressure of the universe on average, and thereby can affect the overall expansion rate of the universe, though these are expected to be rather small. In this thesis we are to show how to calculate the spatial averages of small irregularities in quantities like energy density and pressure up to the second order within the framework of the general-relativistic perturbation theory. We apply the results to some models of cosmology to consider the effects of irregularities on the expansion rate of the universe. Particularly we focus on Cardassian model and Chaplygin model which purport to use unusual equations of state. We discuss the constraints on the parameters of these models in regard to the growth of irregularities as the universe expands.