Nanocomposites with poly(methyl methacrylate) as the matrix and three different types of fillers, namely, nanoclay, nanosilica, and multiwalled carbon nanotubes, were prepared using various methods for the dispersion of the particular fillers. The morphology was analyzed using light and transmission electron microscopy. The composites were investigated by creep-recovery experiments, in addition to conventional dynamic-mechanical measurements. While the changes of the storage modulus G(') are not very distinctive for low filler concentrations, the differences in the recoverable compliance measured by creep-recovery experiments become significant. The better the quality of dispersion, the higher is the recoverable compliance. This finding is explained by an interaction of macromolecules with the particle surface which is more pronounced for well dispersed particles as they offer a larger surface area. Therefore, creep-recovery measurements can very sensitively be used for the determination of the quality of dispersion. This rheological method integrates over the state of the whole sample, whereas transmission electron microscopy analyses only a tiny part of it.