The "at. %" in the xps result just means the relative atom number of corresponding elements, which equals to their molar ratio. You can find the "at. %" in the last sheet of your xps data.

You can find the appropriate information here: http://www.cem.msu.edu/~cem924sg/XPSASFs.html

Regards,

Dmitry

Simple - Take the integrated peak area and divide by its elemental sensitivity factor.

Remember to use sensitivity factors supplied by the instrument manufacturer where possible so that angular effects for non s-orbitals  are accounted for

Thank you all.

To add to David Morgan's answer, the instrument manufacturer should also include the transmission function of the analyser for a particular peak, which is dependent on the kinetic energy of the photoemitted electron. 

to Andrew G Thomas

It is depending mostly on energy analyzer rather than on the XPS system.

Why: because the type (cylinder mirror - CMA or hemispherical - HA) and most important specifications of energy analyzer (acceptance angle of photoelectrons, curvature radius, pass energy "window", retarding potential, magnetic shielding, etc.) are STRONGLY affecting the transmission function as well as the signal-to-noise ratio and effective resolution of the XPS system.

The analyzers are prodused by VG, PHI, ThermoFisher and Shimadzu (Kratos).

So it is not so hard to detect what type of an analyzer is installed and then find some data on transmission function.

Most linear charachteristics have VG analyzers (if I am not mistaken, VG Scienta MK III).

Regards,

Dmitry 

In addition to the above answers on the transmission function, if the data is in Vamas format and you are using software that takes this in to account (e.g. CasaXPS), then all transmission correction is done automatically.