A dry bias in daytime launches of Vaisala RS90 and RS92 radiosondes is well-characterized. This dry bias results in reduced values of observed instability due to an underestimation of moisture and latent heating present in the environment. One method of correcting this bias is to scale radiosonde observations with colocated microwave radiometer observations of total precipitable water. Over four years of mid-day radiosonde launches from the Atmospheric Radiation Measurement (ARM) southern Great Plains (SGP) site in Lamont, Oklahoma, have been scaled in this manner. Calculations of several convective indices, including Convective Available Potential Energy (CAPE) and Lifted Index (LI) have been performed and compared to the non-scaled versions. Statistics on these correlations are presented.