# Dixon's q-test in IR

Question asked by isaac on Nov 2, 2018
Latest reply on Nov 8, 2018 by rhonsberg

Hello ,

I'm using openlab chemstation edition software version C 01.07.SR2 and i'd like to create a sequence summary report which report the dixon's Q-test. My sequence have 5 level of calibration with 3 run for level.Q test, is used for identification and rejection of outliers and it's apllied on peak_area.

Application of the Dixon Q-Test: .1. Arrange the number of values “N” in ascending order. x1 < x2 < . . . < xN

2. Calculate the experimental Q-value (Qexp). This is a ratio of the difference of the suspect value from the next nearest value divided by the range of the values. For testing x1 or xN (as the only possible outliers) we use the following

Qexp values: Qexp = ( x2 - x1) / ( xN - x1) or Qexp = (xN - xN-1) / ( xN - x1)

3. The obtained Qexp value is compared to a critical Q-value (Qcrit) found in the following table. This critical value should correspond to the confidence level (CL) we have decided to apply in the test (at minimum: CL=95%).

If Qexp > Qcrit, then the suspect value can be characterized as an outlier and it can be rejected, if not, the suspect value must be retained and used in all subsequent calculations.

For 3 value  Qcrit.(at CL=95%). is 0.970.

1 I Create a table with group repeat criteria (calibration level) with a descending order sorting option peak_area

for the formula Qexp = ( x2 - x1) / ( xN - x1) in my case xN is max(peak_area) x1 is min(peak_area) the problem is X2 how i can put this value in a variabile ?

thank's a lot.