Back-analysis of a landslide is performed to estimate the mobilized shear strength of the problematic layer. Back-analysis can be an effective procedure for estimating the mobilized shear strength because it avoids the 

problems associated with laboratory and in-situ testing of the problematic layer. The mobilized strength can be used to assess landslide causation and design remedial measures. Even though the factor of safety is known (unity), backanalysis can still involve uncertainties, e.g. failure surface, slope geometry, pore-water pressures, use of tension cracks, shear strength and unit weight of other materials, and accuracy of the stability method. This paper investigates the influence of some of these factors on the back-analysis of landslides using case histories. The results of this investigation are used to develop guidelines for the back-analysis of landslides.