The detection of scattered radiation is recognized as one of the major sources of error in single photon emission computed tomography (SPECT). In this work three scatter correction techniques have been assessed and compared. Scatter coefficients and parameters characteristic of each technique have been calculated through Monte Carlo simulations and experimentally measured for various source geometries. Their dependence on the source/matter distribution and their spatial non-stationarity have been described. Each of the three scatter correction methods has then been tested on several SPECT phantom studies. The three methods provided comparable results. Following scatter compensation, both image quality and quantitative accuracy improved. In particular a slight improvement in spatial resolution and a statistically significant increase in cold lesion contrast, hot lesion recovery coefficient, and signal/noise ratio have been demonstrated with all methods.