The validity of ozone profile data from the satellite sensor SAGE was tested in a series of correlative experiments conducted at five fixed sites between 6¿S and 65¿N during 1979--1980. The intercomparisons included data taken with electrochemical ozone (ECC) balloonsondes and chemiluminescent and optical rocketsonds. The average mean difference for 17 separate comparisons between the SAGE and ECC balloonsonde observations over the altitudes 18--28 km was 9.3% with a standard deviation of 2.8%. Excluding comparisons separated by greater than 500 km reduces the average mean difference to 8.9% and the standad deviation to 2.1%. The average mean difference between SAGE and three optical rocketsond observations over the altitudes 25--50 km was 11% and between SAGE and two chemiluminescent rocketsonds over the altitudes 20--60 km it was 13.5%. Considering the differences in vertical resolution, experimental errors, and ozone time and space gradients, the agreement between SAGE-derived ozone profiles and these correlative measurements is considered very good. In addition, isopleths of ozone mixing ratio versus latitude and altitude are in good agreement with previously published results.