Reliable estimates of population parameters derived from survey methods are essential for decision making in management of endangered species. We evaluated whether point-count surveys used in conjunction with occupancy and binomial mixture models (BMMs) constituted a reliable approach for monitoring the federally endangered Golden-cheeked Warbler (Setophaga chrysoparia) on a preserve in central Texas. Occupancy and abundance were estimated using point-count surveys conducted on each of five 113-ha detection grids in 2008 and seven grids in 2009. Single-season occupancy models and BMMs were used to estimate occupancy and abundance, respectively. Occupancy estimates per grid ranged from 0.48 to 1.0 in 2008 and from 0.52 to 1.0 in 2009. Estimates of abundance were compared with territory densities independently estimated using spot mapping, the standard by which all other avian survey methods are often compared. Abundance estimates produced by BMMs were significantly higher than territory density estimates at all but one site in 2008 and two sites in 2009. While estimation techniques incorporating detection probabilities should be considered in monitoring programs, our results suggest that BMMs deserve careful scrutiny before being used to estimate abundance or to monitor population trends.