Interferometric methods are known to be very sensitive, allowing metrology with resolutions better than the wavelength of the light used for illumination. On the other hand, those methods are susceptible to environmental and mechanical noise, usually. We use Shearography, a version of Speckle interferometry, which, in contrast, is a robust method, resistant to noise and vibrations and compatible to industrial applications. We survey thermally-induced Shearography in order to detect defects in plastic materials, especially in carbon fiber reinforced plastics. We show that by analyzing out-of-plane-deformations, it is possible to evaluate those data quantitatively, enabling the determination of the size as well as the depth of defects. The method of depth determination is based on a gray-scale evaluation with respect to the deformations induced. It has been applied for defects localized in depths up to 10 mm, so far. The method of size determination is based on modeling the dependence of the apparent defect size as a function of the amount of shearing. Simulations of the out-of-plane-deformations for specific defects in different materials have been performed as well. The simulations' results help to understand how to interpret the experimental data. In addition they suggest that the sensitivity for the detection of defects is essentially higher in thermally-induced shearography- than in thermography-experiments.