Issue: Volume 8, Number 5
Date: May 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):

— Baseball batting averages throw some curves at statisticians
— Musings on matrices
— Could a butterfly in Brazil cause a twister in Texas? (One comment on this entry, so far...)
— The action bias drives one to go left or right
— Catapulting into the world of Second Life

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

1. FAQ:Two-level factorial effects falling above the line on a half-normal plot — should they be selected for the model?
2. FAQ: Can general factorial designs be used for process optimization?
3. Info Alert: A bonanza of case-study articles that provide inspiration to apply DOE throughout a wide variety of industries
4. Webinar Alert: The difference between repeats and replicates
5. Reader Response: Microwave popcorn goes off half-popped?
6. Events alert: Quality & Productivity Research Conference
7. Workshop alert:Experiment Design Made Easy sells out consecutive classes, so enroll early if you need DOE basics

PS. Quote for the month: Prayer for an experimenter? (Page through to the end of this e-mail to enjoy the actual quote.)


1. FAQ: Two-level factorial effects falling above the line on a half-normal plot — should they be selected for the model?

-----Original Question-----
A food scientist working on process development
"What should I do about effect points that are above the error line on the half-normal plot? For many of my responses, when I select these effects, the analysis of variance (ANOVA) comes out very insignificant (p > 0.2) for the overall model and none of the individual terms achieve significance."

Points above the line on the half-normal plot of effects do not convey any useful information — the thing to watch for are points falling far off to the right. It’s best to focus on the origin of this graph (zero-zero) and look for a lineup of points (these may be a bit 'snaky'!) emanating from there. These are the trivial many effects (with error triangles mixed in if available to estimate from replicates in the design).

In your case, some (if not all) responses exhibit effects that all line out, that is, none stand out as being abnormally large.

As a chemical engineer working on process development, I'd often run sensitivity studies where I'd take the chemists recommended operating conditions for temperature, etc., and try going a fair amount (but not dangerously!) above and below these set points. The first thing I’d do is look over the response column (Montgomery calls this the ‘intraocular test,’ but I’d just say that I "eyeballed the data"). Often it would range well within the normal span of a lined-out process — giving the normal variability of the machinery, the reaction itself, the sampling and the test results. Then I’d see a lined out half-normal plot of effects — implying nothing significant statistically. For a sensitivity test, that would be a desirable outcome, albeit not very exciting.

Of course, before concluding that the tested factors make little difference to the process, you must assess the potential size of effects and consider the power of your experiment, which is determined by the number of runs in your chosen design. That leads to many other questions that I'd rather not get into!*

*More advanced issues like this, which get into the area of robust design, are covered in the "DOE for DFSS: Variation by Design" (DDFSS) workshop, which is detailed at this web page:

(Learn more about the basics of design and analysis of two-level factorial screening designs by attending the three-day computer-intensive workshop "Experiment Design Made Easy." For a description, see Link from this page to the course outline and schedule. Then, if you like, enroll online.)


2. FAQ: Can general factorial designs be used for process optimization?

-----Original Question-----
From: A bioprocess engineer
"I am using a trial version of your Design-Ease® software, which does not include the response surface method (RSM) design tab. Can I get comparable designs and analysis using the General Factorial design option. I know the vital few factors in my process, but I need to prove this DOE methodology to my
management before I will get approval to purchase the software.

PS. Do you recommend the RSM for inexperienced DOEers? I was trying to get the practical experience at this level to determine if I need to continue training in the RSM course. I want to do the analysis right the first time."

Answer (from Stat-Ease Consultant Shari Kraber):
"The General Factorial will treat all factors as if they are categoric, which may not be what you want. It would be better for you to download the trial version of our top-of-the-line program Design-Expert®. It will look almost the same, but there are more tabs with other design options. This software will be fully- functional for 45 days. If you need more time to gain approval for purchase, send an e-mail to and they can extend your time limit for another period.

In regard to your PS, I will tell you about my experience. When I started my industrial engineering career, I mostly used response surface designs, and rarely used factorial designs. This was because I was working in an area where previous engineers had already determined the factors that controlled the system — they
had good subject matter knowledge. My job was to keep the process 'optimized' as things varied over time.

Factorial designs are used to explore more factors and learn about which ones are the controlling factors, whereas RSM designs are used to optimize once you think you know the significant factors. So, it is certainly okay for an inexperienced 'DOE’er' to use an RSM design — it’s most important to pick the design that fits the problem."

(Learn more about RSM by attending the three-day computer-intensive workshop "Response Methods for Process Optimization."

See for a complete description of this class. Link from this page to
the course outline and schedule. Then, if you like, enroll online.)


3. Info Alert: A bonanza of case-study articles that provide inspiration to apply DOE throughout a wide variety of industries

— "Design Challenge" (published in Adhesives and Sealants Industry):

—"Design of Experiments Helps Solve Adhesive Tape Production Problem" (published by Paper & Packaging):

—"Design of Experiments Helps Reduce Welding Defects from 24 to 0 per Month" (a research study commissioned by a major automotive supplier at Middle Tennessee State University):

—"Pharmaceutical Manufacturer Increases Yield with Response Surface Methods" (published by Pharmaceutical Processing):

—"Implementation of high throughput systems for media and process development" (a poster session presented by scientists from Invitrogen):

—"Designed Experiment Optimizes Method for Removing Endocrine Disrupters" (by researchers at University of Sherbrooke):

—"Statistical Design of Experiments on Fabrication of Starch Nanoparticles — A Case Study for Application of Response Surface Methods (RSM)" (research done at the School of Pharmacy International Medical University in Kuala Lumpur, Malaysia):


4. Webinar alert: The difference between repeats and replicates

You are invited to attend a free web conference by Stat-Ease on "The difference between repeats and replicates in DOE" at 8 AM in the Central USA Time Zone on Tuesday, May 13. It will be offered again at 12 PM (noon) on Wednesday, May 14. The presenter, Stat-Ease Consultant Wayne Adams, will explain the distinction between repeats and replicates ,and he will discuss their relative costs versus benefits as design tools. Wayne promises to keep the presentation practical, but do not be surprised if he spices things up with a few formulas and explanatory stats. Thus we rate this webinar as basic to intermediate in relative level of technicality.

When developing these educational sessions, our presenters often draw valuable material from Stat-Ease DOE workshops. Attendance may be limited for one or both of these two one-hour webinar sessions. Contact our Communications Specialist, Karen, 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. Reader Response: Microwave popcorn goes off half-popped?

-----Original Comment-----
Mark Martin, Biostatistician, Siemens Healthcare Diagnostics

"Dear Mark, I found one possibly confusing statement in the Popcorn Experiment article (republished in the March 2008 Stat-Teaser After showing the interaction plot (Figure 1) for time x setting and their effect on taste, one sentence says"... cooking microwave popcorn at a high setting and for a shorter rather than a longer time probably produces a tastier

My own interpretation of Figure 1 is that low power gives a more robust taste result. That is, taste is very sensitive to changes in time when on high power, whereas taste is consistently good regardless of time when low power is used.

Of course, an important piece of information (not explicitly reported) is the effect of power level on unpopped kernels (bullets). If low power significantly increased bullets, then I would understand and agree with the above quoted conclusion. Otherwise, I would recommend low power so that a long enough time
interval could be used to minimize bullets and maintain good taste.

Thanks for an enjoyable and informative article."

Yes, good point. Since publishing that article many years ago, I've always told the rest of the story (as radio essayist Paul Harvey likes to say) — the fact that at low power, one will never get the yield of popped corn garnered at the low time with the microwave going full blast. Therefore this particular setup
provides a ‘sweet spot’ for trading off yield for taste.

The beauty of this example is that it is one for which instructors can say "DO try this at home!" However, I always add this caveat: "But do not burn your house down in the process!"


6. Events alert: Quality & Productivity Research Conference

Consultant Shari Kraber will talk about "Applications of DOE during Verification Stage" for the 2008 Quality & Productivity Research Conference in Madison, WI, on June 4-6, 2008. For details, see 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: Experiment Design Made Easy sells out consecutive classes, so enroll early if you need DOE basics

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 3-5 (Minneapolis, MN) -> Sold Out: Contact Elicia to be added to the waiting list.)
> July 8-10 (San Francisco, CA)
> August 19-21 (Minneapolis)

—> Mixture Design for Optimal Formulations (MIX) (
> June 10-12 (Minneapolis)

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

—> DOE for DFSS: Variation by Design (DDFSS) (
> November 11-12 (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 — prayer for an experimenter (?):

"Fear and trembling have taken hold of me,
and darkness has descended upon me;
Have mercy upon me, O Lord, have mercy,
...let me never be confounded."

—From Francis Poulenc's song "Timor et Tremor," which my daughter sang along with her high school choir at a concert I attended in April. My guess is that Poulenc did not anticipate how someone like me might interpret the term "confounded"! All I could think of was clients trying to unravel models from ill-advised resolution III two-level factorial designs.

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, #8-4 Apr 08, #8-5 May 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