1. Use the assigned Template for 12-Trial P-B Test Matrix

2. Write the average response of the system of interest in the "Y1" column. Note: The original trial order is not the recommended run order - randomize!

3. For each Xi, total the values of all the responses associated with a "+" symbol. Write the total on the line labeled "Sum +".

4. For each Xi, total the values of all the responses associated with a "-" symbol. Write the total on the line labeled "Sum -".

5. Add the "Sum +" and "Sum -" figures and enter this as "Sum Check".

6. Subtract "Sum -" from "Sum +" and enter the signed value into "Difference".

7. Divide Difference by the number of "+"'s in that column and enter this into "Effect".

8. Combine the unassigned effects to obtain an estimate of the standard error of a factor effect, s_fe .

9. Compute the "+" and "-" part of the confidence interval, t s_fe using the student "t" distribution and the appropriate confidence level and degrees of freedom.

10. Subtract the t s_fe from each effect to find an estimate of the lower control limit and enter this number into "LCL".

11. Add the t s_fe from each effect to find an estimate of the upper control limit and enter this number into "UCL".

12. Mark the factor effects for which the control limits do not include zero. These are the factors identified for further investigation.