Using a generic coarse-grained bead-spring model, Hoy and Robbins reproduced important experimental observations on strain hardening, specifically the generally observed Gaussian strain hardening response and its dependence on network density and temperature. Moreover, their simulation results showed that the strain hardening response at different strain rates collapses to a single curve when scaled to the value of the flow stress, a phenomenon that has not yet been verified experimentally. In the present study, the proposed scaling law is experimentally investigated on a variety of polymer glasses: poly(methyl methacrylate), poly(phenylene ether), polycarbonate, polystyrene, and poly(ethylene terephthalate)-glycol. For these polymers, true stress-strain curves in uniaxial compression were collected over a range of strain rates and temperatures and scaled to the flow stress. It was found that, generally, the curves do not collapse on a mastercurve. In all cases, the strain hardening modulus is observed to increase linearly, but not proportionally to the flow stress. The experimental data, therefore, unambiguously demonstrate that the proposed scaling law does not apply within the range of temperature and strain rate covered in this study. (c) 2008 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 46: 2475-2481, 2008