A general approach to the application of neutron count moment analysis to passive assay is presented. The higher moments of the neutron count distribution are derived with the aid of the probability generating function and are used to formulate an analytic relation between the measurement uncertainty and the assay system parameters. The measurement uncertainty, expressed by the relative resolving power function, for the reduced variance method is developed and analyzed in detail. The study suggests an iterative approach for data processing where the interpretational models are chosen to yield the lowest possible resolving power.