The characterization of bone/scaffold composite mechanical properties is essential for translation to the clinic, but in vivo studies require resources and personnel not available to many investigators. Therefore, the ability to predict composite properties could facilitate scaffold evaluation and reduce the number of in vivo studies required. To date, there have been no studies that have used experimental data to formulate a model of bone morphology or that have examined morphology as a variable in composite properties. In this study, a simple model was developed to predict the effective elastic properties of hydroxyapatite (HA) scaffold/bone composites using representative volume elements (RVE) and finite element analysis. While the RVE for the scaffold is clear, the choice of RVE for bone is not. Two bone geometries were generated for the RVE based on data from an in vivo study: a uniform coating and bridges in pores. Three scaffolds were evaluated in order to consider the effects of scaffold material modulus and porosity. Results showed that the bone geometry had little influence on composite elastic properties when compared to experimental error from the in vivo study. The implication is that such properties can be estimated by measuring the volume fraction of bone using a non-destructive method like microcomputerized tomography and the simple RVE model.