AOAC ISPAM Stakeholder Panel Meeting Book (August 26, 2018)
Action Item #1-A
WG 2 proposed a 2 level approach: 1.Work with standard deviation pooled for all levels (as performed in ISO 19036) 2.Add second criteria if the method is rejected during the first evaluation with AL = 0.5
In order to get smaller T- values because of increased number of degrees of freedom
2nd part ensures the method is not falsely rejected due to high variation between both methods
Action Item #1-A
1.0
0.5
bias lower TI upper TI
0.0
0.00
1.00
2.00
3.00
4.00
5.00
6.00
lower accept upper accept
-0.5
alt accept alt accept
-1.0
Comparison
If large s for both the alternative and reference, indicates no bias .
y = 0.9933x - 0.01
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Initial AP Æ reject Current AP Æ accept
Alternative
0.0
1.0
2.0
3.0
4.0
5.0
6.0
Reference
Made with FlippingBook HTML5