[1] There is limited information about the relative magnitude of the spectral variations in the ionizing component of solar irradiance on solar cycle timescales. We found that the Thermosphere, Ionosphere, Mesosphere, Energetics, and Dynamics (TIMED)/Solar Extreme Ultraviolet Experiment (SEE) Version 9 irradiance values predict relatively more ionospheric heating at solar minimum than those from Version 8. These changes have direct impacts on solar cycle timescale variations in ionospheric and thermospheric energy inputs derived from them. Photoelectron observations from the Fast Auroral Snapshot (FAST) satellite obtained from 2002 to 2008 are used with solar irradiance data, photoelectron flux models, and models of solar irradiance to examine the solar cycle variations of irradiance in the 4-27 nm range derived from the XPS sensor in the TIMED/SEE instrument suite. Good (±50%) agreement is found between daily photoelectron observations and model predictions. The largest differences between observed and modeled fluxes are in the 4-10 nm range, where the Fast Auroral Snapshot data show that the SEE Version 9 irradiances are systematically low. Our analysis suggests that variation on solar cycle timescales in the TIMED/SEE Version 9 and Flare Irradiance Spectral Model irradiance derived from them are systematically low in the 18-27 nm region. Because of uncertainties in the absolute value of the observed photoelectron fluxes and solar irradiances, differences between observed and modeled photoelectron fluxes are not sufficient to determine more exactly the magnitude of variation on solar cycle timescales of solar irradiance in the 4-27 nm region. These suggestions can be confirmed by higher spectral resolution observations that will be made on the Solar Dynamics Observatory mission.