Constructing global upper-air temperature datasets for climate monitoring is a time-consuming activity that involves making many decisions about data selection, quality-control, adjustment, and aggregation. Until recently, only one or two groups have endeavored to assemble climate-quality data products using either satellite or radiosonde data. Now, however, several more data products are either available or under construction for release in the near future. We will compare global and hemispheric tropospheric and stratospheric temperature anomalies from four radiosonde data products and two Microwave Sounding Unit products. Statistical measures of variability, correlation, and trends will be compared.