Issue: Volume 4, Number 3
Date: March 2004
From: Mark J. Anderson, Stat-Ease, Inc. (www.statease.com)

Dear Experimenter,

Here's another set of frequently asked questions (FAQs) about doing design of experiments (DOE), plus alerts to timely information and free software updates. If you missed previous DOE FAQ Alerts, please click on the links at the bottom of this page. If you have a question that needs answering, click the Search tab and enter the key words. This finds not only answers from previous Alerts, but also other documents posted to the Stat-Ease web site.

Feel free to forward this newsletter to your colleagues. They can subscribe by going to http://www.statease.com/doealertreg.html. If this newsletter prompts you ask to your own questions about DOE, please address them to stathelp@statease.com.

"Houston, we have lift off!"  No, this has nothing to do with the costume malfunction at the recent Super Bowl. These are the words that thrill those who follow the adventures of NASA rocket ships, the latest of which involves the planet Mars. To see the latest pictures from the two Mars rovers, Spirit and Opportunity, click on http://www.marsquestonline.org/mer/mer_viewer.php?location=gusev. My brother Paul works for the company, Wind River Systems, who wrote the operating systems for the rovers.  Therefore I kept tuned in when Spirit went off-line for several days. The problem is one that I seem to be fighting constantly with my home PC--the household users keep loading programs and large audiovisual files until the hard-drive fills up and the operating system screeches to halt. For the whole story from Wind River, copy www.computerworld.com/newsletter/0,4902,89829,00.html?nlid=WK to your browser, or click on http://makeashorterlink.com/?B2B762D77. Enjoy how the spokesman insists that the problem was not a "bug."

Here's what I cover in the body text of this DOE FAQ Alert (topics that delve into statistical detail are designated "Expert"):
 
1. FAQ: What if a second outlier crops up after already ignoring one that came up in the first statistical analysis?  
2. FAQ: Numerical optimization--can it be done on factorial designs?  
3. Events alert: Link to a schedule of appearances by Stat-Ease, including Six Sigma training at Ohio State University  
4. Workshop alert: See when and where to learn about DOE--A class is coming to Philadelphia

PS. Quote for the month--A description of the NASA Rover from the view-point of Martian creatures?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1. FAQ: What if a second outlier crops up after already ignoring one that came up in the first statistical analysis?

-----Original Question-----
From:
California

"After reviewing the outlier-t plot and removing outliers, the next iteration of the outlier-t plot shows new outliers--points that weren't outlying the first time, but are now (due to the tighter distribution). What is the validity of eliminating the second set of outliers? I can understand the dangers of doing this too much, but I wonder if there is a statistician's rule of thumb for the maximum number of iterations to perform your outlier elimination and analysis."

Answer (from Stat-Ease Consultant Shari Kraber):
"Your question about outliers is an excellent one.  I haven't seen any rule-of-thumb for removing outliers that occur iteratively. My experience says that if multiple outliers occur, then the initial model should be questioned. We have a saying, "If you eliminate all the data that doesn't fit the model, the model will fit the remaining data."  Eliminating data should only be done when a specific cause is identified that makes you believe the data point is not representative of the true process.  If no identifiable cause can be found, then the data should be left in the model.  It is possible that outliers can appear randomly and they are part of the true process.

In some cases, a transformation of the data would help. The Box-Cox plot helps to make this decision, but again, it's just a guide.

For additional information, you might enjoy reading Mark's article "How to Use Graphs to Diagnose and Deal with Bad Experimental Data" that is on our web site. Here's the link to the article:  http://www.statease.com/pubs/baddata.pdf.

(Learn more about outliers and transformations by attending the three-day computer-intensive workshop "Experiment Design Made Easy." See http://www.statease.com/clas_edme.html for a course description.  Link from this page to the course outline and schedule. Then, if you like, enroll online.)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

2. FAQ: Numerical optimization--can it be done on factorial designs?

-----Original Question-----
From:
Philippines

"Thanks as always for the interesting topics! I learned that optimization is only possible via response surface designs which require at least one continuous factor. However, if I look at the software, there is a provision for optimization even when the design is just factorial. Why is this so?"

Answer:
Strictly speaking, I'd agree with your assessment that response surface methods (RSM) are the tools for optimization, whereas factorial designs screen for vital factors and (hopefully) point out breakthrough combinations. Via its optimization (numerical) tool, Design-Expert® software searches for desirable combinations of factors, regardless of the design choice (factorial OK) or nature of factors (categorical accepted).  I suppose "optimization" may be a misnomer for what the program does. How about "satisfization"? What we hope to provide is a satisfactory compromise for trading off multiple responses that is most desirable based on what the user enters for criteria. (Note: Design-Expert accepts goals for input factors also--for example to minimize an expensive material, or maximize throughput.)

(Learn more about optimization by attending the three-day computer-intensive workshop "Response Surface Methods for Process Optimization." See http://www.statease.com/clas_rsm.html for a complete description. Link from this page to the course outline and schedule. Then, if you like, enroll online.)


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

3. Events alert: Link to a schedule of appearances by Stat-Ease

Click on http://www.statease.com/events.html for a list of appearances by Stat-Ease professionals.  We hope to see you sometime in the near future!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

4. Workshop alert: See when and where to learn about DOE--A class is coming to Philadelphia

"Experiment Design Made Easy" (EDME) will be coming to Philadelphia March 30 through April 1 (no fooling!).
See http://www.statease.com/clas_pub.html for schedule and site information on all Stat-Ease workshops open to the public. To enroll, click on the "register online" link on our web site or call Stat-Ease at 1.612.378.9449.  If spots remain available, bring along several colleagues and take advantage of quantity discounts in tuition, or consider bringing in an expert from Stat-Ease to teach a private class at your site.  Call us to get a quote.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I hope you learned something from this issue. Address your general questions and comments to me at: mark@statease.com

Sincerely,

Mark

Mark J. Anderson, PE, CQE
Principal, Stat-Ease, Inc. (http://www.statease.com)
Minneapolis, Minnesota USA

PS. Quote for the month--A description of the NASA Rover from the view-point of Martian creatures?

"Seen nearer, the Thing was incredibly strange, with a ringing metallic pace, and long, flexible, glittering tentacles swinging and rattling about its strange body."

-- H. G. Wells in Chapter 1 of "The War of the Worlds" (1898) describing a machine sent by Martians to invade the earth.

Trademarks: Design-Ease, Design-Expert and Stat-Ease are registered trademarks of Stat-Ease, Inc.

Acknowledgements to contributors:

—Students of Stat-Ease training and users of Stat-Ease software
—Fellow Stat-Ease consultants Pat Whitcomb and Shari Kraber (see http://www.statease.com/consult.html for resumes)
—Statistical advisor to Stat-Ease: Dr. Gary Oehlert (http://www.statease.com/garyoehl.html)
—Stat-Ease programmers, especially Tryg Helseth (http://www.statease.com/pgmstaff.html)
—Heidi Hansel, Stat-Ease marketing director, and all the remaining staff

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Interested in previous FAQ DOE Alert e-mail newsletters?
To view a past issue, choose it below.

#1 Mar 01
, #2 Apr 01, #3 May 01, #4 Jun 01, #5 Jul 01 , #6 Aug 01, #7 Sep 01, #8 Oct 01, #9 Nov 01, #10 Dec 01, #2-1 Jan 02, #2-2 Feb 02, #2-3 Mar 02, #2-4 Apr 02, #2-5 May 02, #2-6 Jun 02, #2-7 Jul 02, #2-8 Aug 02, #2-9 Sep 02, #2-10 Oct 02, #2-11 Nov 02, #2-12 Dec 02, #3-1 Jan 03, #3-2 Feb 03, #3-3 Mar 03, #3-4 Apr 03, #3-5 May 03, #3-6 Jun 03
, #3-7 Jul 03, #3-8 Aug 03, #3-9 Sep 03 #3-10 Oct 03, #3-11 Nov 03, #3-12 Dec 03, #4-1 Jan 04, #4-2 Feb 04, #4-3 Mar 04 (see above)

Click here to add your name to the FAQ DOE Alert newsletter list server.

Statistics Made Easy™

DOE FAQ Alert ©2004 Stat-Ease, Inc.
All rights reserved.

 


Software      Training      Consulting      Publications      Order Online      Contact Us       Search

Stat-Ease, Inc.
2021 E. Hennepin Avenue, Ste 480
Minneapolis, MN 55413-2726
e-mail: info@statease.com
p: 612.378.9449, f: 612.378.2152