Soil ingestion by children is an important pathway in assessing public health risks associated with exposure to arsenic-contaminated soils. Soil chemical extraction methods and spectroscopic methods are able to characterize various pools and species of arsenic in soil. Use of these methods to estimate bioavailable As is desirable but their ability to measure oral bioavailability of arsenic is unknown. In this work, arsenic extracted by five commonly used soil extractants and characterized by Extended X-Ray Absorption Fine Structure was compared with relative bioavailable arsenic (RBA As) measured in vivo by immature swine (Sus scrofa) dosing trials. Fifteen contaminated soils that contained 233 to 17 500 mg kg^-1 As (V) were studied. Soil extractants were selected to dissolve adsorbed and/or readily soluble arsenic and As in Fe and Mn oxide minerals. The strongest relationship between As determined by soil chemical extraction and in vivo bioavailable arsenic was found for hydroxylamine hydrochloride extractant. Although water soluble As, sodium acetate, and phosphate extractable As all had significant correlations with RBA As, these relationships were nonlinear. Speciation by EXAFS identified scorodite, sorbed As, elemental As, and löllingite with scorodite as the primary As solid phase. A strong inverse relationship was found between scorodite content and RBA As. A strong inverse relationship was found between percent As_ox, As determined by acid ammonium oxalate and expressed as a percentage of Fe + Al content, and RBA As. These results suggest the % As_ox and scorodite fractions represent the non-bioavailable portion of As in these soils. Arsenic (V) speciation by EXAFS and select chemical extraction methods can estimate bioavailable As in these smelter contaminated soils. However, further studies are needed to determine the ability of speciation methods to characterize As availability for other As sources and soils.