Multicomponent composite materials comprised of a dispersed phase suspended in a matrix material are important in a wide variety of scientific and engineering applications including electronic encapsulation, functionally graded materials, and fiber-reinforced structural components among others. Modelling of this class of composites is typically performed using an effective property approach. This approach presumes that the characteristic dimension of the dispersed phase elements is small in comparison to the characteristic length scale of the physical problem under consideration. However, it is not possible to predict a third effective elastic property based on two independent effective elastic properties as it is for homogeneous elastic isotropic materials. Therefore, a macroscale simulation based on an effective Young's modulus and Poisson ratio may yield poor results for a material subjected to shear loading since there is a potentially incorrect presumed effective shear modulus for the simulation. In the current research, boundary element simulations are performed for mesoscopic samples of composite materials to determine effective bulk moduli, shear moduli, Young's moduli, and Poisson ratios. From these analyses, limitations in the effective property approach can be examined.