We present a comparison of in situ stratospheric water vapor measurements with selected water vapor measurements from the stratospheric aerosol and gas II satellite instrument. The criteria for a valid intercomparison typically involve both measurements falling within specified space and time windows. While such an approach is easy to implement, it may be in some cases too restrictive, or even too permissive. We examine a variety of methods for choosing appropriate data sets for comparison. One approach is to use calculated isentropic back trajectories to determine which satellite measurement or measurements are best compared to specific balloon in situ soundings. This selection of intercomparison events does improve the agreement between satellite and in situ measurements. Another approach takes advantage of the long lifetime of water in the midlatitude stratosphere and looks at narrowly constrained zonal mean profiles constructed from the satellite measurements. Both of these approaches show general agreement between the data sets, but the uncertainty in the water vapor measurements largely precludes detailed comparison and examination of observed differences. ¿ American Geophysical Union 1996