Chemical Forums

Chemistry Forums for Students => Analytical Chemistry Forum => Topic started by: chlorinecrown on October 03, 2017, 10:19:57 AM

Title: Are correction factors kosher? When, why, why not?
Post by: chlorinecrown on October 03, 2017, 10:19:57 AM

Hey,

I'm validating a method for quantifying Pt on ICP-OES this week. I took 7 replicates at around what I expected the reporting limit to be at 4 different wavelengths. One of them I can throw out, the other three read just about 10% higher than they ought to, with standard deviations at just about the level I wanted.

I want to just add specification to the method we actually use to multiply by x, where x gets my average to 100% recovery for the MDL determination rounds, and would be a required step before evaluating all future QC, etc.

If you were auditing me, would you accept my justification? (Our validation rounds show systemic bias, so we added a correction factor) What would be the physical significance of this factor? How would I justify multiplying instead of adding, other than "multiplying leads to good agreement with expectations, subtracting makes my blanks looks crazy"

The EPA MDL reference (https://www.epa.gov/sites/production/files/2016-12/documents/mdl-procedure_rev2_12-13-2016.pdf) (December 2016) doesn't mention this at all.