Absolute values of work functions can be determined in ultraviolet photoemission spectroscopy (UPS) by measuring the minimum kinetic energy of secondary electrons generated by a known photon energy. However, some samples can produce spectra that are difficult to interpret due to additional intensity below the true secondary electron cutoff. Disordered absorbates add small intensity below the onset for the transition metal surfaces studied, but for WO3-x films multiple onsets appear with comparable intensity. Lower energy onsets are attributed to energy losses after photoelectrons are generated, for WO3-x these may be excitations of surface plasmon resonances. False onsets (in the context of work function measurements) can be minimized by optimizing experimental detection parameters including limiting analyzer acceptance angles and pass energy. True work functions can be identified by examining the onsets as the sample bias is varied.