The flow stress of many commercial alloys depends on applied strain rate, and this dependence contributes to their resistance to strain localization (necking). Measurement of strain-rate sensitivity for these alloys is complicated by possible dependence on strain, strain rate, testing method, and method of analysis. Polycrystalline zinc sheet was isothermally tested continuously, by multiple rate jumps, and by single rate jumps using standard sheet tensile specimens. A range of strains was examined (6 pct to 66 pct engineering strain) as well as a range of strain rates (10-6-10-1). Comparison of results shows that strain-rate sensitivity indexes obtained by the various methods can differ significantly and that the major cause of these differences is a strain-based stress transient following rapid rate changes. When sufficient care is taken to extrapolate post-transient curves, the data appear to agree well. All of the methods employed show a strong dependence of strain-rate sensitivity on strain rate.