Demography plays a central role in ecology, and accurate estimation of demographic parameters is essential to testing basic theories of life-history evolution, population regulation, and species coexistence, as well as applying ecological theory to a broad range of practical issues including species conservation, predicting shifts in species distribution due to climate change, and understanding the emergence and spread of new diseases. Our goals in this chapter are twofold: first, to demonstrate a problem with estimating demographic parameters that stems from ignoring dispersal, and second, to propose a solution to this problem. We illustrate the problem with a simulation model that shows how ignoring dispersal may lead to the misclassification of sources and sinks; we attempt to solve the problem by using generalized linear models to differentiate population change due to fecundity from population change due to dispersal, thereby helping to improve source and sink classification. We believe that the particular example discussed is only one of a number of parameter-estimation problems limiting our ability to test and apply ecological theories in field systems and that an approach to parameter estimation similar to that used here may be broadly useful in narrowing the gap between ecological theory and field testing of that theory.