Issue: Volume 8, Number 3
Date: March 2008
From: Mark J. Anderson, Stat-Ease, Inc., Statistics Made Easy® Blog

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 the previous DOE FAQ Alert, 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 If this newsletter prompts you to ask your own questions about DOE, please address them via mail

For an assortment of appetizers to get this Alert off to a good start, see these new blogs at (beginning with the most recent one):
—Not always right, but never in doubt
—Models snowed by vagaries of winter weather
—Counterintuitive finding: Sugar substitute correlated to weight GAIN
—Smoker vindicated: Saves $90,000 in health care by not quitting.

Also check out the thoughtful feedback to this prior blog:
—The smoking statistician (1 comment, so far, plus a private correspondence from a fellow I needled for defending his habit. He finally quit the cancer sticks—thank goodness).

Topics in the body text of this DOE FAQ Alert are headlined below (the "Expert" ones, if any, delve into statistical details).

1. Released: Version 7.1.4 of Design-Ease® software
2. FAQ: How do I pick the factors for my experiment?
3. Expert-FAQ: How to combine components in a mixture design and when it may be appropriate to do so
4. Webinar Alert (2nd one): Learn how to accomplish multiple response optimization with Design-Expert
5. Webinar Q&A (follow up from February's talk on "10 Ways to Mess Up an Experiment & 8 Ways to Clean It Up"):
—Question: "Any thoughts on blocking as a pitfall?"
—Comment: "Four common and painful mistakes."
6. Events alert: European DOE User Meeting, Biomedical Focus in Minnesota
7. Workshop alert: Full slate, including DOE for DFSS

PS. Quote for the month: Inventor of DOE suggests alternative to one question at a time for prying secrets loose from nature. (Page through to the end of this e-mail to enjoy the actual quote.)


1. Released: Version 7.1.4 of Design-Ease software

A free, fully-functional Design-Ease ("DE") software version 7.1.4 download is now posted at for evaluation.* This web site also provides free patches to update older, licensed, versions of 7.1. The new release primarily addresses maintenance issues. View the ReadMe file for installation tips,** known 'bugs,' change history, and FAQs.

Design-Expert® ("DX") software achieved release of V7.1.4 earlier, as announced in last month's DOE FAQ Alert. According to our user database, many V6 licensees have not taken advantage of favorable pricing for the tremendous upgrades developed by Stat-Ease over the last 8 years (DX6 came out around 2000). See for a detailing on most of the new features (too many to list exhaustively!).

*If you cannot download the software due to organizational security, consider taking the "online tour" to see how easy Stat-Ease software makes the design and analysis of experiments, specifically two-level factorials for discovery or verification.

**Network administrators are advised of some important changes that must be understood before updating to V7.1.4.


2. FAQ: How do I pick the factors for my experiment?

-----Original Message-----
From: A University Student
"I was reading through your article on the "Keys to successful designed experiments" posted by iSixSigma. [They link to the publication posted at] First of all, I have to say this is a well-written article that makes this topic easy to understand. However, while thinking over your Key #1 (Set Good Objectives), I wondered: How do I pick the factors for my experiment?"

I recommend you start by trying to think of as many variables as possible that affect the response of interest. This typically generates dozens of potential factors, far more than can be accommodated in any single experiment. Thus it is best to simply hold fixed a number of known variables at first and screen others that potentially could be important. For example, you could look for significant main effects from 8 factors run at two levels each in 16 runs—a standard template one finds in a program like the general statistics program at your school's computer lab, or better yet (for DOE in particular)—Design-Ease or Design-Expert software from Stat-Ease (pardon my bias!). Then combine the vital few from this screening experiment with others that had been set aside temporarily and do an in-depth follow up.

This is the beginning of a proven strategy for iterative experimentation that is detailed at:

(Learn more about the basics of DOE by studying "DOE Simplified: Practical Tools for Effective Experimentation, 2nd Edition" by Mark J. Anderson & Patrick J. Whitcomb. See its description at Intended as a do-it-yourself primer for industrial experimenters, this soft-cover book has become an integral part of many Six Sigma short courses and found a home at a number of colleges, particularly for engineering. It comes with a fully-functional, educational version of Design-Ease version 7.1 software that operates for 180 days on any given personal computer. A network version can be purchased for computer classrooms. For a quote, contact


3. Expert-FAQ: How to combine components in a mixture design and when it may be appropriate to do so

-----Original Message-----
From: An industrial statistician
"When analyzing a mixture design, if I fit a linear model and see that the coefficients of two components are very similar in size, can I combine them within Design-Expert and re-analyze the result? Is this legitimate?"

Yes, by right-clicking one of the component headers in the design a menu pops up with the option to combine them. Our programmers made this as easy as pouring milk into tea.* To keep things straight, I advise you re-name the newly combined component column to indicate this manipulation. Furthermore, as I advised in the November 2004 DOE FAQ Alert (#2), like real mixtures it's easy to combine components with Design-Expert but difficult to separate them again, so before doing this, save a backup of your file with a "before combo" suffix or the like and then, after performing this input-data manipulation, save it with your original file name.

However, being a chemical engineer by profession, I naturally fear that combining components only on the basis of their similar effects might be an oversimplification. The one time I did take advantage of the combine-components feature of our software was when experimenting on two flours that showed little impact, relatively speaking, on taste on a pound-cake recipe I experimented on at home. For presentation purposes, it made things easier in the end to combine these two components so I could focus
on flour as a whole, versus the other ingredients (butter, sugar and eggs). See for my detailing of this mixture experiment.

In conclusion, if I were you, I would check back with the subject-matter expert for this experiment (presuming you are providing stat help and not actually doing the research). Do they say these two components are physically alike? If so, I’d go ahead and combine them for simplicity in analysis and modeling.

*This pays homage to the inventor of DOE, Sir Ronald Fisher, who illustrated basic principals in his story about a Lady tasting tea who claimed the ability to tell whether milk was added to tea, or tea poured into milk. See my alert #4 in the DOE FAQ Alert from July 2004 posted at for the outcome of Fisher's verification study on this remarkable claim of the tea-loving Lady.

(Learn more about mixture design by attending the three-day computer-intensive workshop "Mixture Design for Optimal Formulations." For a complete description of this class, see Link from this page to the course outline and schedule. Then, if you like, enroll online.)


4. Webinar alert (2nd one): Learn how to accomplish multiple response optimization with Design-Expert

You are invited to attend a free web conference by Stat-Ease on "Multiple response optimization with Design-Expert" at 8 AM in the Central USA Time Zone on Tuesday, April 1. It will be presented again at 12 PM (noon) on Wednesday, April 2. Here is the abstract from the talk's author, Consultant Shari Kraber:

"The optimization module in Design-Expert searches for combinations of factor levels that simultaneously satisfy the requirements placed on each of the responses and factors. Discover how to get the most out of the optimization module in order to find the "sweet spot" for your product or an operating window for your process. Learn how to fine-tune your search by adding weights and importance settings to your basic criteria. A case study will be used to illustrate all the features of Design-Expert's optimization module."

Attendance may be limited for one or both of these two one-hour webinar sessions. Contact our Communications Specialist, Karen Dulski, via to sign up. If you can be accommodated, she will send you the link for the WebConnect and dial-in for ConferenceNow voice via telephone. Toll-free access extends worldwide, but not to all countries.


5. Webinar Q&A (follow up from February's talk on "10 Ways to Mess Up an Experiment & 8 Ways to Clean It Up")*:
—Question: "Any thoughts on blocking as a pitfall?"
—Comment: "Four common and painful mistakes."

-----Original Question-----
Attendee connecting from Indiana
"Any thoughts on blocking as a pitfall?"

At some point any design becomes segmented so badly by blocking that it degrades the resolution of effects, especially so for fractional two-level factorials. For example, if you divide an 8-run 2^3 factorial into 4 blocks, all three two-factor interactions (2FI) will be lost. In that case, see if what you are attempting to block out, such as lot-to-lot differences, can be fixed or controlled enough that it will not likely bias results, and (of course!) randomize the experimental run order as insurance.

I encourage blocking if the design is stout enough to absorb the price in degrees of freedom. It often removes variations so large that they would otherwise overwhelm important effects.

-----Original Comment-----
From: A Master Black Belt from Michigan
"Mark, I've got four mistakes that I've found common and painful (I think most of these were covered in your webinar, if not in text, than in your discussion). In no particular order these are:

1. Not enough material set aside for mistakes and re-running of runs. In other words, we plan sixteen lab runs and get just enough material to run the sixteen runs. Something bad happens in a run and you clearly want to abort it, now you can't re-do that run. (Solution --> plan for oops's, they happen. Also can use to
re-run outlier runs to validate them).
2. You assume that the design will be run the way you designed it. (Solution --> good communication and actively following the experiment while it's happening).
3. You let the cost of doing it right outweigh the costs of not looking hard enough. Most novice DOE practitioners will get sticker shock on how many runs we really should do. (Solution --> use power to illustrate the risk, do replicates in blocks so that if the signal is so big that you can see effects without repeats you can end the experiment and cut it in half).
4. You don't set the factors at appropriate levels and either find no effect or work in a design space with no practical value. (Solution --> Use your subject matter experts to find the "Goldilocks" factor settings. Not too tight, not too far apart, but just right. Test the corners of the design because even the SME's can be wrong.)

All good, especially the point about potentially replicating in blocks, which would only be feasible if you have enough material set aside! Thank you for this contribution to the cause. —Mark

* The Powerpoint slides from prior webinars are posted at Over 100 attendees enrolled for the most recent presentation on "10 Ways to Mess Up an Experiment & 8 Ways to Clean It Up." This basic-level presentation provided practical advice for actual experimenters. The prior presentations by Consultant Pat Whitcomb provided more advanced statistical details on sizing designs for process screening and mixture optimization. The upcoming webinar by Shari Kraber will be aimed at an intermediate level of expertise on DOE.


6. Events alert: European DOE User Meeting, Biomedical Focus in Minnesota

The Second European DOE User Meeting will be held March 10-12 in Berlin, Germany. Come to increase your understanding of design of experiments (DOE) techniques, learn of successful real-life applications of DOE, and also attend presentations specific to Stat-Ease software and its features. For more information, see

See Stat-Ease exhibiting at the 2008 Biomedical Focus Conference in Brooklyn Center, Minnesota on March 17. On the following day, the 18th of March, Stat-Ease Consultant Pat Whitcomb will present a workshop on "DOE Case Studies in the Biomedical Industry." See for details.

Click for a list of upcoming appearances by Stat-Ease professionals. We hope to see you sometime in the near future!

PS. Do you need a speaker on DOE for a learning session within your company or technical society at regional, national, or even international levels? If so, contact me. It may not cost you anything if Stat-Ease has a consultant close by, or if a web conference will be suitable. However, for presentations involving travel, we appreciate reimbursements for airfare, hotel and meals—expenses only. In any case, it never hurts to ask Stat-Ease for a speaker on this topic. Contact if you have an event coming up with an open slot for a presentation.


7. Workshop alert: Full slate, including DOE for DFSS

Seats are filling fast for the following DOE classes. If possible, enroll at least 4 weeks prior to the date so your place can be assured. However, do not hesitate to ask whether seats remain on classes that are fast approaching!

--> Experiment Design Made Easy (EDME)
(Detailed at
> April 29 through May 1 (Minneapolis, MN)

--> Mixture Design for Optimal Formulations (MIX)
> April 8-10, 2008 (Minneapolis)

--> Response Surface Methods for Process Optimization (RSM)
> September 23-25 (Minneapolis)

--> DOE for DFSS: Variation by Design (DDFSS)
> March 11-12 (Minneapolis, MN)

See for complete schedule and site information on all Stat-Ease workshops open to the public. To enroll, click the "register online" link on our web site or call Elicia at 612.746.2038. 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.*

*Once you achieve a critical mass of about 6 students, it becomes very economical to sponsor a private workshop, which is most convenient and effective for your staff. For a quote, e-mail


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



Mark J. Anderson, PE, CQE
Principal, Stat-Ease, Inc. (
2021 East Hennepin Avenue, Suite 480
Minneapolis, Minnesota 55413 USA

PS. Quote for the month—Inventor of DOE suggests alternative to one question at a time for prying secrets loose from nature:

"No aphorism is more frequently repeated in connection with field trials, than that we must ask Nature few questions, or, ideally, one question, at a time. The writer is convinced that this view is wholly mistaken. Nature, he suggests, will best respond to a logical and carefully thought out questionnaire."
-- R. A. Fisher

PPS. The old-fashioned alternative to DOE is often referred to as OFAT, or one factor at a time. However, I recently came across another acronym for this: COST, or changing only a single thing.

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

Acknowledgements to contributors:
—Students of Stat-Ease training and users of Stat-Ease software
—Stat-Ease consultants Pat Whitcomb, Shari Kraber and Wayne Adams (see for resumes)
—Statistical advisor to Stat-Ease: Dr. Gary Oehlert (
—Stat-Ease programmers, especially Tryg Helseth and Neal Vaughn (
—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, #4-4 Apr 04, #4-5 May 04, #4-6 Jun 04, #4-7 Jul 04, #4-8 Aug 04, #4-9 Sep 04, #4-10 Oct 04, #4-11 Nov 04, #4-12 Dec 04, #5-1 Jan 05, #5-2 Feb 05, #5-3 Mar 05, #5-4 Apr 05, #5-5 May 05, #5-6 Jun 05, #5-7 Jul 05, #5-8 Aug 05, #5-9 Sep 05, #5-10 Oct 05, #5-11 Nov 05, #5-12 Dec 05, #6-01 Jan 06, #6-02 Feb 06, #6-03 Mar 06, #6-4 Apr 06, #6-5 May 06, #6-6 Jun 06, #6-7 Jul 06, #6-8 Aug 06, #6-9 Sep 06, #6-10 Oct 06, #6-11 Nov 06, #6-12 Dec 06, #7-1 Jan 07, #7-2 Feb 07, #7-3 Mar 07, #7-4 Apr 07, #7-5 May 07, #7-6 Jun 07, #7-7 Jul 07, #7-8 Aug 07, #7-9 Sep 07, #7-10 Oct 07, #7-11 Nov 07, #7-12 Dec 07, #8-1 Jan 08, #8-2 Feb 08, #8-3 Mar 08 (see above)

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

Statistics Made Easy®

DOE FAQ Alert ©2008 Stat-Ease, Inc.
All rights reserved.


Software      Training      Consulting      Publications      Order Online      Contact Us       Search

Stat-Ease, Inc.
2021 E. Hennepin Avenue, Suite 480
Minneapolis, MN 55413-2726
p: 612.378.9449, f: 612.746.2069