# DOE for attribute data

Six Sigma – iSixSigma Forums Old Forums General DOE for attribute data

Viewing 7 posts - 1 through 7 (of 7 total)
• Author
Posts
• #27492

Schuette
Participant

I’d like to do a DOE using 5 factors (a half-fraction of a 2^5). But the response variable is pass-fail attribute data. I’ve heard that such DOE’s are possible but don’t recall where I saw that information. Does anyone have an easy to read reference on this, or other advice they can share?

0
#67468

Bob Sutter
Participant

You can conduct the DOE as you would conduct it for variables data. The difference is the analytical technique used to analyze the data. Instead of ANOVA or linear regression you will need to use logistic regression. In order to do so you will also need to create a variable that identifies distinct treatment conditions.

0
#67481

Neil Polhemus
Participant

If you can collect a lot of samples at each combination of the factors, then you can use the proportion of failures as the response. You could then do a logistic regression. However, you may be able to use a standard ANOVA approach with an arc sine square root transformation (Box, Hunter and Hunter discuss this approach on p. 134). If you are getting only one response at each combination of the factors, then you may have to resort to simply plotting the outcome.

0
#67484

Schuette
Participant

Bob,
Thanks for the tip on logistic regression.  I found some information and examples so far in the Minitab Help files.  I’ll also check my Regression textbook (Draper & Smith). I believe there may be a chapter on logistic regression there but I’m not sure.  At any rate, I’ll check it all out AFTER vacation, which starts at 4:30 today.
Jim

0
#67485

Schuette
Participant

Thanks Neil.  I think I’m going to try the logistic regression approach rather than the Box, Hunter, & Hunter approach.
Jim

0
#67493

Murray
Participant

Just remember to have an adequate sample size.

0
#67585

Bob Sutter
Participant

Jim:
You may want to consider the textbook “Applied Logisitc Regression” by David Hosmer and Stanley Lemeshow.

Bob

0
Viewing 7 posts - 1 through 7 (of 7 total)

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