γ-ray spectrometry is extensively used for 2 3 8 U determination because of its simplicity in terms of sample preparation. Nevertheless, processing of measurement data presents several challenges, such as evaluation of overlapping photopeaks and calibration of detector efficiencies for different matrices and photon emissions. Besides, the emission probability data for 2 3 8 U and its daughters show significant discrepancies among different studies. In this study, we assess the suitability of four methods, using the photopeaks at 63.3, 92.6, 766.4 and 1001.0keV, respectively, to determine 2 3 8 U in environmental samples. With that aim we analyse several reference and environmental samples, detailing the entire procedure involved in the calculation of the 2 3 8 U activities and highlighting possible error sources. Based on the results obtained, we propose new yields for the 2 3 4 Th photons and compare these to corresponding values from the literature.