Please can you elaborate your question, I do not fully understand what you mean. Normalising a ratio does not seem to make sense. Can you show us what results you have and say what you are trying to achieve with your analysis.

Hello, if you asked about the basics, you'll find them everywhere. However, if this is a specific question, it is well hidden! Question please then in more detail.

http://emu.uct.ac.za/training/sem-school/basic-theory-of-edx-quantitative-analysis-correction/

I have attached the format of the result. The link provided by Volker is helpful. Thanks.

No extension, what type of file is it? 

You need to attach it in different format. 

Okay, all is clear for you, but for use is not clear the history of your file!? ;-)

epmalab.uoregon.edu/reports/kratio.doc

Vladimir - it is a text file

here

--------

Other

Element Normalized K-Ratio rounded to first decimal

Na 31.9

Al 3.0

Si 7.1

S 13.3

Cl 0.6

K 6.8

Ca 9.1

Ti 0.0

Cr 0.6

Mn 0.0

Fe 0.0

Ni 0.5

Cu 0.5

Zn 2.6

Br 0.0

Sr 9.5

Zr 0.0

Ag 0.0

Sn 4.0

Sb 5.5

Ba 3.2

La 0.1

Ce 1.8

W 0.0

Au 0.0

Hg 0.0

Pb 0.0

Bi 0.0

--------

So I presume the numbers have been produced using the K-ratio _method_  rather than these being the K-ratios themselves. This means that no matrix corrections have been applied. They are first approximations, and only valid if the standards are very close in composition to the sample. 

It is usual to quote EDX results in either el wt% or at. % after matrix correction and with some report of the error of the measurement in a sigma value. Whether you use normalised values or not depends on whether or not you have done a full quantitative calibration. 

What type of material is this, how was it prepared, what type of analysis did you do. What are you hoping to achieve?

 ... as Quick Start Guide, you can use our new paper by Florian:

Article Effect of the stoichiometry of niobium oxide on the resistiv...

Further to Ian Slipper's comment, I think the word "normalized" in this case indicates that the "K-ratio values" have been normalized to sum up to 100% Total. These K-ratio values are indeed first approximation Element percent values obtained by comparison with either a physical standard or in the standardless mode.  

I appreciate you all for your contributions.

I think romano and ian are right. I just want to add one important remark for you. I would not even call the k ratio a guess for the elemental composition as there are many effects that can alter the ratio of the k fluorescence lines in a sample.

Especially in a sample like yours where light and heavy elements are present you should expect attenuation, secondary excitation and more... so be very careful with the k ratios if you want to derive quantitative informations.

Maybe this helps also:

http://link.springer.com/content/pdf/10.1007/978-3-540-36722-2_5.pdf