Using the Rock Canyon Creek carbonate-hosted rare earth element (REE)–F–Ba deposit as an example, we demonstrate the need for verifying inherited geochemical data prior to reinterpretation. Inherited La, Ce, Nd and Sm data obtained by pressed pellet X-ray fluorescence (XRF), and La and Y data obtained by aqua regia digestion inductively coupled plasma atomic emission spectroscopy (ICP-AES) for more than 300 drill-core samples were analysed in 2009 and were subsequently compared to sample subsets re-analysed using lithium metaborate-tetraborate (LMB) fusion ICP mass spectroscopy (ICP-MS), Na2O2 fusion ICP-MS, and LMB fusion-XRF. We determine that LMB ICP-MS and Na2O2 ICP-MS accurately determined REE concentrations in control reference materials (CRM) SY-2 and SY-4, and provided precision of about 10%. Fusion-XRF was precise for La, Ce and Nd at concentrations greater than ten times the lower detection limit; however, accuracy of this method was not established because REE concentrations in SY-4 were below the lower detection limit. Analysis of the sample subset revealed substantial discrepancies for Ce concentrations determined by pressed pellet XRF in comparison to those determined by other methods due to Ba spectral interference. Samarium, present in lower concentrations than other REE that were determined, was consistently underestimated by XRF methods relative to ICP-MS methods. This may be the result of Sm concentrations approaching the lower detection limits of XRF methods, elemental interference or inadequate background corrections. Aqua regia dissolution results, reporting only for La and Y, are underestimated relative to the other methods. We highlight the importance of selecting the most appropriate analytical method and reference materials for determining the REE content of mineralized rock which may be several orders of magnitude higher than that of typical host rock.