The global (50¿N--50¿S) distribution of stratospheric column ozone (SCO) is derived using solar backscattered ultraviolet (SBUV) profiles and compared with SCO amounts derived from Stratospheric Aerosol and Gas Experiment (SAGE) and ground-based measurements. An evaluation of archived SBUV (version 6) ozone profiles with ozonesonde profiles shows that the low resolution of the SBUV instrument in the troposphere and lower stratosphere leads to a low bias in the SBUV profile in the troposphere and a high bias in the lower stratosphere in regions where anthropogenic tropospheric ozone production influences the climatology. An empirical correction applied to the SBUV profile prior to separating the stratosphere from the troposphere reduces the bias in the lower stratosphere and results in a SCO distribution in good agreement with SCO derived from SAGE ozone profiles. Because the empirical correction is most pronounced at northern middle latitudes, we compare these resultant SCO values with those measured at two northern middle latitude sites (Wallops Island and Hohenpeissenberg) using concurrent measurements from Dobson spectrophotometers and ozonesondes. Our analysis shows that the empirically corrected SCO at these sites captures the seasonal cycle of SCO as well as the seasonal cycle derived from SAGE stratospheric ozone profiles. These results have important implications for the derivation of tropospheric ozone from SBUV ozone profiles in conjunction with Total Ozone Mapping Spectrometer (TOMS) total ozone measurements using the tropospheric ozone residual (TOR) methodology.