Issue: Volume 8, Number 11
Date: November 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, 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):

—Candy is dandy, but for melting the ice, coffee has the hot hand
—A sign I never saw on a beach in Minnesota
—A gladiator for snaky adders? Answer: the "Addiator"
* Need a feed from StatsMadeEasy to Microsoft's Outlook? See

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

Here's the most recent post (perhaps still unanswered) to be found in the DOE "Analysis" section:
From: Chris F. (Registered Member)
>When fitting RSM models typically alpha in is 0.05. So there's a 1 in 20 probability that a factor included in a model is not actually significant, right?! Is it legitimate to make alpha much smaller, say 0.001, to "ensure" factors introduced into the model are "very" significant?<
Feel free to contribute to this discussion thread.

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

1. Peer-recognition alert: Stat-Ease statisticians honored by Shewell Award for best presentation at Fall Technical conference
2. Expert-FAQs: Three questions on the central composite design (CCD) for response surface methods (RSM):
A. Does it make sense to put axial runs all in the second block?
B. Why are runs and center points not balanced between blocks?
C. How far out should the axial points go?
3. Book giveaway: Winners announced
4. Expert-FAQ: Choosing a model via backward selection
5. Webinar alert (1st): An Introduction to Mixture Design for Optimal Formulations
6. Workshop alert: Most popular Stat-Ease class, Experiment Design Made Easy, coming to Dallas and San Diego!

P.S. Quote for the month: How United Way makes sure that donations achieve maximum good — the answer involves data gathering.


1. Peer-recognition alert: Stat-Ease statisticians honored by Shewell Award for best presentation at Fall Technical Conference

Congratulations to Stat-Ease founder Pat Whitcomb, presenter of a talk on "Graphical Selection of Effects in General Factorials" co-authored by Professor Gary W. Oehlert of the University of Minnesota. For the second time, Pat won the Shewell Award for best presentation at a Fall Technical Conference. Attendees, primarily industrial statisticians, rate speakers on the basis of their talk and handouts, thus this represents purely peer recognition of an outstanding presentation. Aside from the gratification for a job well done by Pat, these awards also indicate an approval of ideas by him and our advisors (Gary and his predecessor Kinley Larntz) for making statistics easier — the mission of Stat-Ease.


2. Expert-FAQs: Three questions on the central composite design (CCD) for response surface methods (RSM)

A. Does it make sense to put axial runs all in the second block?
-----Original Question-----
Life scientist
"I took your Designed Experiments for Life Sciences course* back in June and I've designed my first DOE. It’s a RSM CCD with 8 numeric factors and one categorical factor using a Min-Run Res V core to reduce runs.

Setting up the DOE returns a total of ~120 runs. I work in 96 well plates, so I decided to run the experiment in 2 blocks (days).

In looking at the axial point runs, they are all in block 2. Does this make sense? I ask because the center point runs seem divided between the blocks."


Answer (from Stat-Ease Statistician Wayne Adams):
"Yes, that is the best way for various reasons to split the runs — fractional factorials with center points in one block, and axial runs with center points in another.

The most common reason for splitting the runs this way is so the analysis can be conducted on the factorial blocks. If the curvature statistic proves to be insignificant then there is less reason to run the axial block. The axial block is all about modeling curvature that, based on an insignificant curvature test, doesn't exist. Put short and simple, it is very possible that you will not need the axial block at all.

B. Why are runs and center points not balanced between blocks?
-----Original Question-----
Clinical Program Director for a diagnostics company
"It has taken me some weeks to dive back in since I became the proud owner of a copy of Design-Expert! My first design has me a little puzzled. I did a CCD, 3-factor 2-day design, and am wondering why the number of runs and center points per day is not balanced, and why all the factorial runs occur on Day 1."

This is by design of the response surface method (RSM) you selected. The central composite design is comprised of a core two-level factorial with a number of center points selected for statistical reasons. Their replication provides power to estimate the ‘bulls-eye’ of the experimental region — a cube in this case of three factors. The second block contains the axial points, which Design-Expert® software makes easy to identify by right-clicking the vertical Select button at the upper left corner of your design layout and turning on display of Point Type. The axial runs fall outside the ‘box’ that provide leverage for estimating effects. A few more center points are mixed into this second block (day 2) as a link to the majority of center points that are designed into the first block (day 1).

That’s it in a nutshell. For far more detail, see "RSM Simplified"* — published in 2005 by Productivity Press (New York, NY).

C. How far out should the axial points go?
-----Original Question-----
French semiconductor engineer
"I like to use a central composite design. But I never know if it is preferable to use a face-centered central composite design or to use an alpha coefficient greater than 1."

I too like the CCD for process optimization via RSM — either the face-centered design (FCD) option or, if operating constraints permit a wider range, with the new "practical" alpha values developed by my colleague Pat Whitcomb. For an easy-to-follow detailing (via case study) of the CCD with practical alpha levels, see This is a new posting by Stat-Ease.

(Learn more about RSM designs, including CCD, by attending the three-day computer-intensive workshop "Response Surface Methods for Process Optimization." For a description of this class, see Link from this page to the course outline and schedule. Then, if you like, enroll online.)


3. Book giveaway: Winners announced

These lucky readers were drawn at random from approximately 50 entrants to a drawing* for several copies of "Design & Analysis of Experiments," 6th Edition by Douglas C. Montgomery:

-> Dennis Urban, Principal Specialist, Lancaster Laboratories, Lancaster, Pennsylvania
-> Carolyn Cooney, Research Assistant, MatTek Corp, Milford, Massachusetts
-> John Krech, Scientist, US Ink, Carlstadt, New Jersey

Congratulations to these three winners and condolences to the others who entered into this drawing. Keep watching for more great books to be given away in the future.

*(Sorry, due to the high cost of shipping, this offer applied only to residents of the United States and Canada.)


4. Reader response: Why should 5% be the critical "p" value?

-----Original Comment-----
From: Eric Kvaalen, Consultant in Engineering and Statistics.
Paris, France
"Dear Mark, A couple comments on your latest DOE FAQ Alert #4 on "Choosing a model via backward selection":

First of all, ...there's no law that says "Anything with a p- value less than 5% is significant and anything with a p-value greater than 5% is not significant"! That's just an arbitrary convention that ..was apparently based on an illustration by Fisher in which he said that if a bet on something has odds greater than a pound to a shilling (that is, 20 to 1), then we consider it pretty unlikely. In my opinion, we should use our own judgment about what we consider significant, and we should report p-values so others can decide what they think, rather than just saying something is significant or not.

My second comment is that if you play around looking for the best model, then the "p-values" are not valid. That's why one uses the Bonferroni correction — because otherwise you could play around till you found a model with a particular term which by chance has an uncorrected p-value less than the magic number (0.05), and then claim that this is significant.

This is a controversy that I'd rather not stir up. However, being a chemical engineer by profession who earned my experiment designer stripes doing process development under constant pressure from product management, I tend to be more pragmatic about p. Thus, if forced to choose between Sir Ronald Fisher's more rigid view versus that of Master Brewer W. S. Gosset (see below), I would go for the latter.

"The level of significance fulfills the conditions of the rational grounds for the disbelief it engenders."

"Results are only valuable when the amount by which they probably differ from the truth is so small as to be insignificant for the purposes of the experiment. What the odds should be depends —
1. On the degree of accuracy for which the nature of the experiment allows, and
2. On the importance of the issues at stake."


5. Webinar alert (1st): An Introduction to Mixture Design for Optimal Formulations

You are invited to attend a free web conference by me on "An Introduction to Mixture Design for Optimal Formulations." This free conference, which I will keep at a beginner level statistically, will be broadcast on Wednesday, January 21 at 2 PM USA Central Time* (CT). I will repeat my webinar on Thursday, January 22 at 8 AM. It is aimed at product formulators who at best may be using standard factorial designs, or worse yet, the one-variable-at-a-time method. Keeping it simple and making it fun, I will introduce tools of multicomponent mixture design, modeling and statistic analysis. My hope is to generate interest in these powerful DOE methods for quickly converging on the sweet spot — where all desired product attributes are achieved.

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 as soon as you see your way clear by contacting our Communications Specialist, Karen, via 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).

*(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 on international communications is even more complicated than statistics! Good luck!)


6. Workshop alert: Most popular Stat-Ease class — Experiment Design Made Easy — coming to Dallas and San Diego!

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!

—> Designed Experiments for Life Sciences (DELS)
> November 18-20 (Minneapolis)

—> Experiment Design Made Easy (EDME)
(Detailed at
> December 9-11(Dallas, TX)
> January 27-29, 2009 (San Diego, CA)
> February 24-26 (Minneapolis, MN)

—> Response Surface Methods for Process Optimization (RSM)
> December 9-11(Minneapolis)

—> Mixture Design for Optimal Formulations (MIX)
> February 3-5 (Minneapolis)

—> DOE for DFSS: Variation by Design (DDFSS)
> May 5-6 (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 — How United Way makes sure that donations achieve maximum good — the answer involves data gathering:

"What gets measured gets done."
—Greater Twin Cities United Way 2008 Campaign Video "Agenda for Lasting Change" (I researched the provenance for this adage which is the title of a 2006 white paper posted at The author, Robert M. Williamson, attributes this saying to Peter Drucker, Tom Peters, Edwards Deming or perhaps Lord Kelvin. Check out the additions provided by leadership trainer John Jones.
— Mark)

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 Wolfe, Stat-Ease sales and 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, #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 (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