A new method is proposed to correct for attenuation effects by adventitious organic contamination in quantitative XPS. The corrected intensity ratio I o X =I o Y involving two different substrate peaks is determined as the slope of the plot of I X =I C1s as a function of I Y =I C1s for samples covered with organic overlayers of varying thickness. The thickness of deposited organic overlayers does not need to be perfectly uniform provided that it does not vary by more than ยฑ40%. The sample surface can either be flat or present a random roughness (e.g. powders). Shadowing effects accompanying overlayer non-uniformity are taken into account by a simple model involving effective thickness and fractional coverage, which is shown to enhance greatly the accuracy of computed intensity ratios. Constraints of the method and possible implementations were examined with an NaCl powder covered with adventitious contamination, polystyrene (PS) and polydimethylsiloxane (PDMS). The most straightforward implementation of the method would be to record repeatedly the peaks of interest while decreasing the adventitious contamination overlayer thickness by thermal desorption. The error on I o Na2s =I o Na1s is Y10%, with the apparent overlayer thicknesses varying from 1.1 to 0.3 nm. The method is applied to check the reliability of the spectrometer transmission functions reported previously by Weng et al.