Populations of giant Canada geese (Branta canadensis maxima) have been established in most of the lower 48 United States. However, establishment and spread of these populations has led to an increasing number of human-goose conflicts. Knowing the pattern of dispersal of these populations may be useful to wildlife managers interested in minimizing nuisance problems. Consequently, we analyzed band recovery data from six Bird Conservation Regions (BCRs) of four midwestern states to determine if there was a common pattern of dispersal among these populations. We used negative binomial regression to test a series of models that included age at recovery, gender, number of years after initial population established, recovery year, and banding location (BCR) to explain dispersal distance. Mean dispersal distances were <100 km for all BCRs. We did not detect a consistent pattern of dispersal followed by giant Canada geese from different BCRs. However, dispersal distances decreased for birds recovered many years (>~12) after banding. The Central Mixed-grass Prairie (CMP) had considerably shorter dispersal distances than the Central Hardwoods (CH) BCR. The interaction of Recovery Year and Region (RYR*Region) model indicated reductions in dispersal distance during RYR 2 in the Eastern Tallgrass Prairie (ETP), RYR 7 in the ETP, and RYR 16 in the Mississippi Alluvial Valley (MAV). Otherwise, factors affecting dispersal distances varied little among BCRs and most geese were recovered at or near banding locations. Based on our results, giant Canada goose populations in one region or state are not dispersing to nearby regions or states. Key words: band recovery, Bird Conservation Regions, Branta canadensis maxima, dispersal, giant Canada geese, human-wildlife conflict, regression,
