The Solar Ultraviolet Spectral Irradiance Monitor on the Upper Atmospheric Research Satellite has been measuring the solar spectral irradiance between 115 nm and 410 nm since October 11, 1991. The original goal of the experiment was to maintain an absolute accuracy of 6 - 10% (a wavelength dependent, 2-sigma value) and a precision of 2% over a lifetime of the experiment, perhaps as long as an 11 year solar cycle. To do this, one must provide a time dependent accounting of the degradation of the optical elements' responsivity. The instrument was designed to maintain its calibration through the use of two identically-designed spectrometers, redundant optical elements, and four deuterium calibration lamps. Calibration measurements, which include solar as well as lamp spectral scans, are performed cyclically at intervals of approximately 1-month, 3-months, 6-months, and 12-months. We discuss the technique of determining the wavelength dependent degradation of the routinely used optical paths from the calibration measurements of sun and lamps. In so doing, we demonstrate the multiplicity of calibration data required to sequentially refine determination of the working channels' degradation. Estimated responsivity changes are also shown for the four years plus of instrument observations.