The intrinsic compression strength of ceramics can be difficult to measure. The specimen geometry and test fixture, if not properly designed, can result in the generation of undesirable tensile stresses that can lead to misleadingly low strength values. Since compression strength is a parameter in numerous modeling and simulation packages, properly measuring it is critical. We have previously shown a dumbbell-shaped specimen is ideal for measuring the compression strength of ceramics at both quasi-static and dynamic strain rates. This specimen was designed to increase the likelihood of fracture initiating in the gage section and a measurement is considered valid if it does, via a high speed camera. However, the observed damage on the surface may not be indicative of interior damage in the gage section. Thus, we performed Kolsky bar experiments of these specimens with in-situ x-ray phase contrast imaging (XPCI) at the Dynamic Compression Sector at the Advanced Photon Source at Argonne National Lab. Samples were loaded for ~75 µs and XPCI images were captured at nano- and micro-second time scales to observe interior fracture initiation and evolution in the gage section. We will compare the observations from in-situ XPCI and visible light high-speed imaging as a function of the applied stress in the specimen gage section for different ceramic materials.