Measurements of elastic electron backscattering probabilities may provide the electron inelastic mean free paths (IMFP). This technique is known as elastic peak electron spectroscopy (EPES). The IMFPs in Ti canbe determined from optical data by Tanuma et al., Gries or the TPP-2M predictive formulae. Studies on Si and Ni show a weak influence of surface roughness, but a stronger influence of atomic composition, density, texture, average grain size and surface excitations on the IMFPs. In the present work, the elastic electron backscattering probability and IMFPs are determined for Ti exposed to sputtering and annealing. The atomic composition of Ti is determined by XPS and texture and grain size is determined by x-ray diffraction (XRD). Ti samples exhibit no texture and grain size from 1 µm to 100 µm. For annealed samples the titanium hydride phase can be observed from the binding energy shift of Ti 2p XPS spectra. Scatter in the resulting IMFPs due to density and composition change (O and C contamination, and H from vacuum background) is negligible in comparison to scatter due to the grain size and standard used (Al and Cu). The IMFPs in Ti of grain size 1-10 µm are smaller than IMFPs in Ti of grain size 100 µm. These differences are explained by effects not accounted for in the calculations, i.e. the inelastic losses on the grain boundaries which are more pronounced in Ti of small grain size, and surface excitations which differ for Ti, Al and Cu.