In many medical ultrasound imaging systems, clinical data are often presented in the form of tissue displacement and tissue velocity maps. In this paper, we determine the accuracy of tissue displacement estimates obtained using a well established crosscorrelation, speckle tracking technique. We evaluate the accuracy of the cross-correlation technique as a function of 3 parameters: 1) the magnitude of the tissue displacement, 2) the size of the region being tracked and 3) the host tissue type. Three different sample types: porcine liver, porcine muscle and woolen sea sponge samples were used to study the effect of different scattering media on displacement estimates. For identically sized target regions and displacements, liver samples produced estimates with the largest uncertainties. For all 3 sample types, the accuracy and precision of displacement estimates deteriorated with increasing sample displacements and decreasing target dimensions.