In this paper, the results of the modeling calculations carried out for predicting the interference effects expected in the sum frequency generation ͑SFG͒ spectra of a specific thin-layer system, described in the accompanying paper, are tested by comparing them with the experimental spectra obtained for a real thin-layer film comprising an organic monolayer/variable thickness dielectric layer/gold substrate. In this system, two contributions to the SFG spectra arise, a resonant contribution from the organic film and a nonresonant contribution from the gold substrate. The modeling calculations are in excellent agreement with the experimental spectra over a wide range of thicknesses and for different polarization combinations. The introduction of another resonant monolayer adjacent to the gold substrate and with the molecules having a reverse orientation has a significant affect on the spectral shapes which is predicted. If a dielectric substrate such as CaF 2 is used instead of a gold substrate, only the spectral intensities vary with the film thickness but not the spectral shapes. The counterpropagating beam geometry will change both the thickness dependent spectral shapes and the intensity of different vibrational modes in comparison with a copropagating geometry. The influences of these experimental factors, i.e., the molecular orientational structure in the thin film, the nature of the substrate, and the selected incident beam geometry, on the experimental SFG spectra are quantitatively predicted by the calculations. The thickness effects on the signals from a SFG active monolayer contained in a thin liquid-layer cell of the type frequently used for in situ electrochemical measurements is also discussed. The modeling calculation is also valid for application to other thin-film systems comprising more than two resonant SFG active interfaces by appropriate choice of optical geometries and relevant optical properties.