The validity of the approximations of neglecting proton attenuation due to nuclear collisions and all reaction products from nuclear collisions in space vehicle shielding studies is tested. Comparisons of absorbed-dose and dose-equivalent results obtained with the production and transport of nuclear-reaction products included with similar results obtained using the approximations are presented. The use of the approximations is found to lead to an overestimation ( a factor of two) of the absorbed dose and dose equivalent for a wide range of incident Van Allen belt and solar-flare proton spectra and for a variety of shield thicknesses and shield materials.