In assessing the surface quality of low-scatter laser mirrors, it is important to understand the advantages and limitations of various measurement methods. For example, values for the root mean square (rms) surface roughness may depend critically on the lateral resolution of the profile measuring instrument and on the surface spatial wavelength range sampled if the correlation lengths of the surface microirregularities are less than or equal to the instrumental resolution, or if appreciable long range surface waviness is present. If the surface roughness is deduced from a measurement of the total integrated scattering, the measurement wavelength, collection angles, and spatial wavelengths of the surface microroughness all have an effect. In the present study, several methods were used to measure the rms roughness of a group of samples which included low-scatter laser mirror substrates and additional samples of varying roughness, all partly coated with high reflectance dielectric multilayers and aluminum overcoatings. Measurement techniques and instruments included a diamond stylus surface profiling instrument (Talystep), Mireau heterodyne interferometer (Wyko), optical heterodyne profilometer (Lawrence Livermore National Laboratory), and total integrated scattering (Naval Weapons Center and Balzers). The roughness values thus obtained are compared and conclusions are drawn concerning the effect of the surface spatial wavelength range sampled and the instrumental lateral resolution.