A study has been made of birefringence in oriented glassy polystyrene. The aim was to develop a methodology for the prediction of birefringence in glassy polymers with frozen-in molecular orientation. The predictive approach employed was to use a recently proposed constitutive model, consisting of a coupling between the Likhtman-Graham Rolie-Poly model for polymer melt rheology and a previously established model for polymer glasses. The methodology was validated with an experimental study of oriented glassy polystyrenes with known melt-stretching histories. It was found that, for orientation processes sufficiently far above T-g (i.e., not including subentanglement stretch), birefringence could be predicted accurately in monodisperse materials.