Rangelands in the southwestern United States represent a current stronghold for northern bobwhite (Colinus virginianus); however, bobwhite populations in rangelands exhibit extreme inter‐annual variability in abundance in relation to fluctuating weather patterns. Recent declining bobwhite population trends within this region have led to the supposition that landscape‐scale processes, such as habitat loss and fragmentation and predation from increased raptor abundance, may be acting in conjunction with weather to reduce bobwhite populations. Our objective was to determine the relative effects of these factors on bobwhite populations in the rangeland environments of Texas and Oklahoma, USA. We obtained publicly available datasets for bobwhite counts (Breeding Bird Survey, state‐agency roadside counts), weather (PRISM), land cover (National Land Cover Database), and raptors (Christmas Bird Counts) for 3 5‐year periods (1990–1994, 1999–2003, 2009–2013). Data were collected at route and landscape scales based on routes within the Rio Grande Plains region of Texas and the Central Mixed Grass Prairie region of Texas and Oklahoma. We used generalized linear mixed models with a backward selection approach to determine top models for each dataset based on scale and ecoregion. Covariate relationships with bobwhite abundance followed expected patterns, with positive relationships with habitat, precipitation, and minimum temperatures and negative relationships with maximum temperatures and raptor abundance. Weather variables were the factors most consistently selected within both regions, while minimum winter temperature was overall the top variable. These relationships occurred within a landscape still containing relatively vast amounts of unfragmented bobwhite habitat (>60% rangeland; >15 million ha). Management within these regions should be focused on retaining habitat at a broad scale, while managing for suitable cover at a local scale to help mitigate weather effects.