The hydroxyl radical (OH) is one of the most important oxidants in the atmosphere, as it is involved in many reactions that affect regional air quality and global climate change. Because of its high reactivity, measurements of OH radical concentrations in the atmosphere are difficult, and often require careful calibrations that rely on the production of a known concentration of OH at atmospheric pressure. The Indiana University OH instrument, based on the Fluorescence Assay by Gas Expansion technique (FAGE), has been calibrated in the laboratory using two different approaches: the production of OH from the UV-photolysis of water-vapor, and the steady-state production of OH from the reaction of ozone with alkenes. The former technique relies on two different actinometric methods to measure the product of the lamp flux at 184.9 nm and the photolysis time. This quantity derived from N₂O actinometry was found to be 1.5 times higher than that derived from O₂ actinometry. The water photolysis and ozone-alkene techniques are shown to agree within their experimental uncertainties (respectively 17% and 44%), although the sensitivities derived from the ozone-alkene technique were systematically lower by 40% than those derived from the water-vapor UV- photolysis technique using O₂ actinometry. The agreement between the two different methods improves the confidence of the water-vapor photolysis method as an accurate calibration technique for HO_x instruments. Because several aspects of the mechanism of the gas phase ozonolysis of alkenes are still uncertain, this technique should be used with caution to calibrate OH instruments.