# P-value formula in DOE, Help

Six Sigma – iSixSigma Forums Old Forums General P-value formula in DOE, Help

Viewing 6 posts - 1 through 6 (of 6 total)
• Author
Posts
• #37406

Clowers
Participant

I have a pareto of effects graph from a doe, 3 Factor 2 levels
I understand the P-Value .05 significance level.
I want to place a formula on the graph to explain how the P-Value is calculated.  Could someone help me on what the formula would be??
Thanks,
Kathy

0
#110095

Clowers
Participant

a

0
#110096

Clowers
Participant

Is there a equation for the p-value?

0
#110097

Ken Feldman
Participant

Which graph are you referring to, the probability plot or the Pareto chart?  The value for the p value is the area beyond some calculated value, either Z, t, F or other statistic and depends on the distribution in question.  Minitab gives you the value or if you are adventurous, you can look it up in the appropriate table for the statistic you are using.  For your DOE, check the values above in the Session Window, assuming you are using Minitab.  See if you can’t find some relationship between them and the graphs.

0
#110113

P3
Participant

Kathy,
What Darth’s saying is right, you have to use the corresponding statistic for which P-value you’re looking for (even though the P-values are the same for T and F, which they should be).

To get the P-value for the F-statistic:P = Fdist(F, df1, df2)  where the df1 is the degrees of freedom for the factor of interest, and df2 is the degrees of freedom for the Error term. These df-values are shown in the Session Window in MINITAB.
If you don’t have MINITAB then you can calculate the P-value using Excel exactly as I wrote in the formula above:  = FDIST(F, df1, df2)
What you get is the area (in an F-distribution) beyond the point which is stated by the F-value itself.

To get the P-value for the T-statistic:The mathematical relationship between F and T is simply: F equals T^2 (using the single Terms only)  therefore there’s no difference between their P-values.

I believe you’re using the “Source” as describing your Pareto Chart, so the F-distribution is your choice (whatever difference that make).
/P3

0
#110114

Clowers
Participant

Thanks for the help guys.  I understand it better now.
Kathy

0
Viewing 6 posts - 1 through 6 (of 6 total)

The forum ‘General’ is closed to new topics and replies.