Abstract: In the global extinction crisis currently underway, threats to biodiversity are not simply limited to species with particular risks in particular cases. Biodiversity will be increasingly affected by the wholesale decline of entire clades. In the face of this crisis, with amphibians ranking at the top of all vertebrates in the rate of extinctions, there is a great need for macroecological studies addressing three key areas of species declines. The processes that put species at great risk for extinction may be associated with 1) environmental factors, 2) spatially structured, or geographic effects, and 3) evolutionary predispositions to those processes (phylogenetic structure in vulnerability). I present a collection of work to address primarily the phylogenetic and geographic components of species vulnerability. First, I construct a theoretical foundation for using phylogenetic comparative methods for conservation assessments, emphasizing the importance of evolutionarily specific parameters and trees. I prescribe a greater conservation focus on understanding the severity of clade-level threats and potential data deficient species vulnerability, and identify evolutionary scenarios with the greatest return on resource investments. Second, I identify autocorrelated threats in the amphibian tree of life representing potential evolutionary predispositions to enigmatic rapid declines and Redlist threatened status in the superfamily Hyloidea. Third, I focus in on a family in Hyloidea, Hylidae, and use multiple phylogenetic comparative methods to identify phylogenetic signal in processes that selectively threaten lineages in the tree. I find phylogenetic signal (a predisposition to vulnerability) in pollution, habitat loss, species with multiple threatening processes, Redlist threatened status, and enigmatic rapid declines, concentrated in the clade Hylini. Among the comparative methods employed is a new application for conservation of a more flexible measure of phylogenetic signal accounting for selection using an Ornstein-Uhlenbeck model. Fourth, I use a landscape perspective to discover what spatial and environmental factors predict threats to amphibians in Venezuela. I find that traditional measures of human impact (population density and ecological footprint) effectively predict higher numbers of threatened amphibians, but indigenous peoples population density does not predict threats. Accounting for spatial dependence in the landscape reveals that cultural stewardship, i.e., parks on indigenous versus nonindigenous land, cannot predict threatened species distributions, failing to validate typical conservation concerns over indigenous population impacts to parks and biodiversity. Using a local spatial autocorrelation metric, I also find that the northwest region of Venezuela is a hotspot of geographic irreplaceability, for spatially autocorrelated threatened species, endemics and data deficient species. Overall, my collection of work addresses key themes in the amphibian extinction crisis using a macro-analytic approach: evolutionary predisposition to threats, anthropogenic threatening processes and spatial autocorrelation (or clumping) of threatened species. My work supports the emerging consensus that the extinction crisis is widespread, in terms of impact to phylogenetic diversity and geographic regions, but my findings also point to advantages for conservation policy and management gained by prioritizing vulnerable clades and geographic regions that stand to lose the most diversity and hold the greatest potential management payoffs.