Issue: Volume 9, Number 5
Date: May 2009
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, see below.

==> Tip: Get immediate answers to questions about DOE via the Search feature on the main menu of the Stat-Ease® web site. This not only pores over previous alerts, but also the wealth of technical publications posted throughout the 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):

—Awesome demonstration of design of experiments
—Technology facilitates building a stronger database on blood pressure and other medical measurements
—TV detectives stumble over odds of matching birthdays
—The science of "guesstimation"

* Need a feed from StatsMadeEasy to Microsoft's Outlook? See
** (See the comment to this blog and follow its link to an amazing collection of stats and graphs on record temps across the USA)

Also, Stat-Ease offers an interactive web site — its Support Forum for Experiment Design at Whereas this monthly ezine — the DOE FAQ Alert — shares one-on-one communications with Stat-Ease StatHelp, anyone (after gaining approval for registration) can post questions and answers to the Forum, which is open for all to see (with moderation). Check it
out and weigh in!

Here's a synopsis the most recent question posted (found in the DOE "Analysis" section):
>From: A Registered Member
"I'm analyzing a 2^3 full-factorial design with 2 blocks... What are the practical consequences of accepting the non-hierarchical model?"

See the answer (from Stat-Ease consultant Wayne Adams) and feel free to contribute to this discussion thread by getting registered as a user of the Support Forum for Experiment Design (if you're not already).

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

1. FAQ: How do predictive equations differ in terms of coded versus actual factors?
2. Expert FAQ: Why do some minimum-run resolution V (MR5) designs include one additional run beyond that needed to fit the two-factor interaction model?
3. Info Alert: Tablet formulators succeed via mixture design
4. Reader Response: Modeling mixture behavior as polynomials
5. Reader Contribution: Frustration with scientist objecting to the use of statistically designed experiments
6. Webinar alert (3rd): "DOE — What's In It for Me" — an executive summary on the power of matrix-based multifactor testing
7. Events Alert: Several quality talks on tap
8. Workshop Alert: Designed Experiments for Life Sciences (DELS) coming to Cambridge, Mass.

P.S. Quote for the month: What Gerry Hahn sees as major challenges to statisticians working in industry today.


1. FAQ: How do predictive equations differ in terms of coded versus actual factors?

-----Original Question-----
Senior Research Associate and Applications Engineer
"We are working on DOE, and want to use the equation in the ANOVA in a spreadsheet. Two appear at the bottom of the sheet, 'Final equation in terms of coded factors,' and, 'Final equation in terms of actual factors.' There was no transform used, but the coefficients are different between the two equations (attached). Why is this, and which should we use for our spreadsheet exercise? The goal is to run confirmation tests, and the chemist would like to incorporate the solution equation into the spreadsheet and his notes."

Answer (from Stat-Ease Consultant Shari Kraber):
"Thank you for your question. Both the coded and the actual equations make the same predictions. To use the coded equation, you enter factor values such as -1 and +1 to represent the low and high levels of your design. To use the actual equation, you can enter factor levels in the original units of measure. The chemist would likely prefer to use the actual equation because it is in the natural units.

The coefficients in the coded equation for the main effects tell you how much the response is changing (and which direction) as you move one coded unit (-1 to 0, or 0 to +1) of the factor level. Because all factors are coded to the same range (-1 to +1) the regression is not overly weighted by a factor that might have a large range versus one that has a small range. You can look at the coded coefficients and see which factor has a larger effect.

However, the presence of interactions does start confusing any simple interpretations. The coefficients in the actual equation have to do double duty. They not only represent how much the response is changing, but also have to account for varying factor ranges. Typically, a factor that has a large range will end up with a small coefficient, while a factor with a small range will have a large coefficient.

So, to summarize: both equations will make the same predictions, but as an engineer or chemist, I prefer to use the Actual equation in practice. However, if someone plans to interpret the coefficients to understand which factor has a greater or lesser effect on the response, only the coded equation will provide the correct answer."

(Learn more about factorial models by attending the three-day computer-intensive workshop "Experiment Design Made Easy." See for a description of this class and link from this page to the course outline and schedule. Then, if you like, enroll online.)


2. Expert FAQ: Why do some minimum-run resolution V (MR5) designs include one additional run beyond that needed to fit the two-factor interaction model?

-----Original Question-----
R&D Fellow and Tech Lead DFSS
"One of our folks was setting up a min-run res V design for 7 factors. He asked me why DX7 calls for 30 runs, since 29 should be enough. The same +1 run is seen for the 8factor DoE, but not for
6, 9 or 10-factor designs. The D-optimal design appears to call for the 29 runs for the 7-factor model and 37 for the 8-factor model as one would expect, and does not have any aliased 2-factor interactions. Why is there one additional run for the 7 and 8 factor DoE's?"

Answer (from Stat-Ease Consultant Shari Kraber):
"If you look at the long name on the screen, it is 'Minimum Run Equireplicated Resolution V' design. Having an even number of runs provides a balance to the design that improves many of the statistical features of the design. So, all the MR5 designs have an even number of runs. If the minimum number happens to already be even, then it is left alone. If the minimum number is odd, we add one run to make it 'equally replicated.' Here's a related article that discusses the MR5 designs, then augments them into the Central Composite design:"


3. Info Alert: Tablet formulators succeed via mixture design

The February/March 2009 issue of Pharmaceutical Formulation & Quality (PFQ) features an article showing how a stats-savvy R&D scientist helped Schiff Nutrition International "Build a More Effective Tablet Press Process." The production staff had varied press parameters without success. Then they ran a mixture design that verified the standard formulation. For details, see

(Learn more about these statistical tools 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. Reader Response: Modeling mixture behavior as polynomials

-----Original Comment-----
William R. Wilcox, Professor of Chemical and Biomolecular Engineering, Clarkson University
"I have a major problem with blindly modeling mixture behavior as polynomials. There's an implicit assumption that there are no discontinuities in behavior. Such discontinuities are typical in solid mixtures and sometimes occur in liquid mixtures as well. (The gold-copper melting point behavior in your example* is actually very unusual, because they are completely soluble in the solid — at least until they are cooled.)... For this reason, in dealing with mixtures one must take great care that such discontinuities don't occur — this may require lots of data and sophisticated methods of measurement, such as electron microscopy."

Response (from Stat-Ease Consultant Pat Whitcomb):
"You raise a very valid point and one that needs continual emphasis in all DOE applications not just mixtures. The saving grace is that most designs are used for optimization within a relatively restricted region of the design space, for example, varying the iron, nickel and chromium plus and minus around a
particular stainless steel of interest. In our training we stress using first principles when designing a set of experiments and focusing your region of interest to avoid discontinuities and restricting factor ranges so runs do not 'fall off the table.' In supplemental training (usually as part of consulting with particular clients) we also show how to use mechanistic models, which can be used to fit a broader design space and/or develop an equation that can be used to predict beyond the experiment space."

*In free "Primer on Mixture Design for Optimal Formulation": announced in April DOE FAQ Alert.


5. Reader Contribution: Frustration with scientist objecting to the use of statistically designed experiments

-----Original Comment-----
Quality Systems Manager
"I attended a meeting last night to look at some data generated by one of our developers. This fellow is very intelligent and generally very open minded. He completely avoided the use of DOE or any other statistics in drawing conclusions from the data; all conclusions were made based on graphical representations. Several times, I suggested he use DOE in an attempt to understand cause-effect relationships, understanding whether things like run-order were affecting the data, etc., and those suggestions were rejected. This morning I received a lengthy e-mail from him defending his position on DOE. Here’s a little of what he said:
'We are relying on the statistical results from the likes of a process DOE as the end-all. There are flaws in this.
> Often the critical factors are not identified.
> Because the results are statistics based, they are taken verbatim.
> In part, it feels like the main application of a statistical approach is to comply, not to find the truth.
> DOE time is not balanced with the time to find critical factors.
Your thoughts would be greatly appreciated."

That’s too bad — a top-notch technical expert falling into the trap of making this (science vs DOE) an either/or proposition. First off, I share his concerns about allowing DOE and stats to become the end rather than the means for developing profound knowledge of a system. For that reason, we at Stat-Ease go all out to put subject-matter knowledge first and foremost as the foundation for technical success. Unfortunately it seems that high-powered scientists are put off entirely by the empirical methods of DOE and, on the other hand, those with a more practical and less academic bent take to these statistical methods. The challenge is getting the two camps together — a classic setup for a synergistic interaction that pays off in far more solid and productive experiments.


6. Webinar alert (3rd): "DOE — What's In It for Me" — an executive summary on the power of matrix-based multifactor testing

You are invited to attend a free web conference by Stat-Ease Consultant Wayne Adams on "DOE — What's In It for Me." This free conference, which Wayne will keep at a managerial level statistically, will be broadcast on Wednesday, May 27 at 2 PM USA Central Time* (CT). He will repeat his webinar on Thursday, May 28 at 8 AM. It is aimed at those who need convincing on how design of experiments harnesses the power of matrix-based multifactor testing. Ideally, after seeing what DOE can do, it will be favored over the old-fashioned one-factor-at-a-time (OFAT) method.

Stat-Ease webinars vary somewhat in length depending on the presenter and the particular session -- mainly due to breaks for questions: Plan for 45 minutes to 1.5 hours, with 1 hour being the target median. When developing these one-hour educational sessions, our presenters often draw valuable material from Stat-Ease DOE workshops. Attendance may be limited, so sign up soon by contacting our Communications Specialist, Karen, via . If you can be accommodated, she will send you the link to WebConnect and the voice dial-in telephone number for ConferenceNow (toll-free access extends worldwide, but not to all countries).

*(To determine the time in your zone of the world, try using this link: Note that we are based in Minneapolis, which appears on the city list that you must manipulate to calculate the time correctly. It seems that figuring out the clock for international communications is even more complicated than statistics! Good luck!


7. Events alert: Several quality talks on tap

Stat-Ease Consultant Shari Kraber will present "Best Practices to Plan/Analyze a Verification DOE" on Tuesday, May 19 in a session sponsored by the Statistics Division of the American Society of Quality (ASQ) for the World Conference on Quality Improvement (WCQI) in Minneapolis. For details on WCQI, see

As an added bonus for those of you who attend WCQI, Stat-Ease Consultant Pat Whitcomb will speak at an Open Council Meeting of ASQ's Chemical & Process Industries Division (CPID) on "Sizing Mixture Designs" at 5 PM on Monday, May 18. At the 2009 Quality & Productivity Research Conference (QPRC) in Yorktown, NY on June 3-5, I will present "Dual Response Surface Methods (RSM) to Make Processes More Robust." This is one of three talks in the "New Developments in Design of Experiments" invited session organized by Douglas C. Montgomery of Arizona State University. For all the details on QPRC, see


8. Workshop alert: Designed Experiments for Life Sciences (DELS) coming to Cambridge, Mass.

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
> June 9-11 (Minneapolis)
> June 10-12 (Marlborough, MA)

—> Mixture Design for Optimal Formulations (MIX)
> June 23-25 (Edison, NJ)

—> Response Surface Methods for Process Optimization (RSM)
> July 7-9 (Minneapolis)

—> Designed Experiments for Life Sciences (DELS)
> June 8-9 (Cambridge, MA)
> July 28-29 (Minneapolis)

—> DOE for DFSS: Variation by Design (DDFSS)
> November 10-11 (Minneapolis)

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 —What Gerry Hahn sees as major challenges to statisticians working in industry today:

"The use of statistics...especially design of experiments...and the so-called democratization of statistics — providing increased accessibility to statistical software and on-line advice on how to use it... Today, indeed is a golden age for statistics in business and industry! But it is not a golden age for statisticians... This should be of concern, not only to the profession but also to management, because it can result in ineffective use of statistics and, more importantly, incorrect conclusions."

—Gerry Hahn, excerpted from p140 of "A Conversation with Gerry Hahn," Quality Engineering, V21, #2, 2009, April-June.

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, led by Neal Vaughn and Tryg Helseth (
—Heidi Hansel Wolfe, Stat-Ease sales and marketing director, and all the remaining staff that provide such supreme support!


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, #8-4 Apr 08, #8-5 May 08, #8-6 June 08, #8-7 July 08, #8-8 Aug 08, #8-9 Sep 08, #8-10 Oct 08, #8-11 Nov 08, #8-12 Dec 08, #9-01 Jan 09, #9-02 Feb 09, #9-03 Mar 09, #9-04 Apr 09, #9-05 May 09 (see above)

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

Statistics Made Easy®

DOE FAQ Alert ©2009 Stat-Ease, Inc.
All rights reserved.

Software      Training      Consulting      Publications      Order Online      Support      Contact Us       Search

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