Plasma potentials computed from Langmuir probe data rely on a method to account for secondary electron emission (SEE) from the electrodes. However, significant variations exist among published models for SEE and the reported experimental parameters used to evaluate them. As a means to critically assess SEE computation methods, two of four tungsten electrodes on a Langmuir–Mach probe head were replaced with molybdenum and exposed to Alcator C-Mod boundary plasmas where electron temperatures exceed 50 eV and SEE becomes significant. In this situation, plasma potentials computed for either material should be identical—the SEE evaluation method should properly account for the differences in SEE yields. Of the six methods used to compute SEE, two are found to produce consistent results (Sternglass model with Bronstein experimental parameters and Young–Dekker model with Bronstein experimental parameters). In contrast, the method previously used for C-Mod data analysis (Sternglass model with Kollath parameters) was found to be inconsistent. We have since adopted Young–Dekker–Bronstein as the preferred method.