In many pronghorn populations there exists uncertainty about which factors most affect survival rates. Also, although pronghorn within the same population might exhibit both migratory and non-migratory behaviors, few studies have assessed whether migration status affects survival rates. We determined cause-specific mortality for 134 adult, radio-tagged pronghorn in southwestern North Dakota, 2004–2008, estimated survival rates, and examined support for 11 a priori models containing combinations of year, age, sex, season, and migration status. We documented 102 deaths (52 females, 50 males) of radio-collared pronghorn. Over half of the observed mortality was due to hunter harvest (58%). Migration status was not an important determinant of pronghorn survival when compared to other factors. A model containing a season × sex interaction was most supported. Seasonal survival rates were >0.90 for males and females except in fall when female survival rates were reduced to 0.8, and male survival was less than half (0.42), which was due to hunting mortality. Survival rates were highest in winter, which might have been facilitated by mild winter weather during our study. Given the important role of harvest on pronghorn survival, biologists can best manipulate pronghorn abundance, and reduce landowner conflicts, through hunting license allocation.