Reversing decades of fire exclusion by hardwood midstory reduction is now used to recover populations of the federally endangered Red-cockaded Woodpecker (Picoides borealis) in longleaf pine (Pinus palustris) forest ecosystems. The effects of Red-cockaded Woodpecker management on winter birds in longleaf pine sandhill forests are largely unknown. Examining habitat use of winter migrants, some of which are declining, may influence the selection of habitat management techniques used for Red-cockaded Woodpeckers to benefit overwintering migrants. During the winters (December–February) of 1997–1998 and 1998–1999, we tested experimentally the effects of hardwood reduction treatments applied in 1995 on winter birds at Eglin Air Force Base in fire-excluded northwest Florida longleaf pine sandhills. Treatments were (1) prescribed spring burning, (2) herbicide application, (3) mechanical felling and girdling, and (4) a control where decades of fire exclusion was maintained. We also sampled winter bird flocks in frequently burned, nonexperimental reference plots to measure management success. Hardwood reduction techniques had no effect on flock species richness, which averaged 7.9 and 7.2, respectively, during 1997–1998 and 1998–1999. Larger flocks in felling and girdling and in herbicide plots were primarily due to significantly higher numbers of overwintering Chipping Sparrows (Spizella passerina), as well as resident Red-cockaded Woodpeckers and an influx of temperate migrant Pine Warblers (Dendroica pinus). In contrast, flocks in control plots were smaller (flock size and species composition in spring burn plots were intermediate) and composed of hardwood-associated species, such as Tufted Titmouse (Baeolophus bicolor) and Carolina Chickadee (Poecile carolinensis). The relative uses of longleaf pines and hardwoods by Red-cockaded Woodpeckers, Pine Warblers, and Brown-headed Nuthatches (Sitta pusilla) during both winters best explained that winter birds present in herbicide, felling and girdling, and reference plots were more likely to forage on the same tree species and substrates than birds in spring-burned plots, and least likely to forage on the same species and substrates as birds in the control plots. Those differences corresponded to the following increasing order of hardwood stem mortality among treatments: control, spring burn (41%), felling and girdling (62%), and herbicide (92%). Repeated burning is recommended to restore the reference foraging condition because it was eight times less expensive than other techniques, which favored mostly Chipping Sparrows.