Dear Experimenter,

Here’s a fresh set of answers to frequently asked questions (FAQs) about design of experiments (DOE); plus, timely alerts for events, publications, and software updates. Check it out!

On behalf of our Stat-Ease team, I thank you for your ongoing support of our mission to make the most of every experiment. With gratitude and warm wishes for a joyous holiday season and the coming new year,

Mark J. Anderson, PE, CQE
Engineering Consultant, Stat-Ease, Inc.

P.S. Stay safe!

P.P.S. Quote for the month:

Timely for the trend to overcome extremely weak data with overly strong assertions.

(Page down to the end of this e-zine to enjoy the actual quote.)

Vol. 20, No. 6 - Nov/Dec 2020

Why exclude known factors from initial screening?

Info Alert
Valuable quick-start resources for DOE beginners

Webinar Alert
Robust Design and Tolerance Analysis

Workshop Alert
Sharpen up on DOE - Enroll before winter classes fill

StatsMadeEasy Blog
My wry look at all things statistical and/or scientific with an engineering perspective.
Also, see the Stat-Ease blog for tips on making DOE easy. For example, a recent posting provides “Breaking beyond A/B splits for better business experiments”. Take a look!
Why Exclude Known Factors from Initial Screening?

Original question from an R&D Engineer:
“Based on the strategy of experimentation you laid out in your ‘Know the SCOR’ presentation for the 2020 Stat-Ease DOE Summit*, I should omit known main effects in the screening experiments performed. This assumes that filtered-out factors—the ‘trivial many’—do not interact significantly with the known factors temporarily set aside. Is that true?”
*View this talk and others at this conference site. -Mark

Great question! Holding out known factors facilitates screening more of the unknown ones within the experimental budget. Furthermore, the screening does not get overwhelmed by big main effects from the factors already identified as major ones. However, this strategy of experimentation bets that only the vital few factors that survive the screening will interact significantly with the strong factors excluded from this first phase of process development. It relies on Phase 2, characterization, to resolve any two-factor interactions by requiring at least a resolution V design (or full factorial).

It is a bit chancy to discard the trivial many factors based on a resolution IV design that excludes known factors and detects main effects only. If that worries you (and runs come relatively cheap), consider bypassing the screening phase with one of Design-Expert’s exclusive minimum-run resolution V (MR5) characterization designs. For a given number of runs, these MR5s handle far more factors than standard (2^k-p) templates.
- Mark

PS I strongly advise against the temptation to run Plackett-Burman or other resolution III designs for screening due to them aliasing main effects with two-factor interactions. These designs are better deployed for the final ruggedness testing—the “R” in the SCOR strategy.
(Learn more about screening and characterization designs by attending the next distance-learning presentation of Modern DOE for Process Optimization.)
Valuable Quick-Start Resources for Beginners
Revamped with a new case study by me that illustrates the magic of multifactor testing, this treasure trove of educational resources provides a quick start for DOE ‘newbies’. If you know anyone who remains mired in non-statistical and/or one-factor-at-time (OFAT) experiments, please share this Alert with them.
“Overview of Robust Design, Propagation of Error and Tolerance Analysis”
Stat-Ease Consultant Pat Whitcomb lays out tools for robust design and tolerance analysis in a free webinar presented on Tuesday, December 15 at 10 am CST. He will demonstrate how Design-Expert® software deploys propagation of error (POE) to account for variation transmitted from deviations in factor levels. Using POE, you can find the flats—high plateaus or broad valleys of response, whichever direction one wants to go—maximum or minimum; respectively. Pat will also show how to apply tolerance analysis for drilling down to the variation of individual units, thus facilitating improvement of process capability. Sign up now for this highly informative free webinar.

PS Do you need a speaker on DOE for a learning session within your company or technical society at regional, national, or international levels? If so, please get back to me. – Mark
Sharpen up on DOE—Enroll before winter classes fill

You can do no better for quickly advancing your DOE skills than attending a Stat-Ease workshop. Our expert instructors provide you with a lively and extremely informative series of lectures interspersed by valuable hands-on exercises. Enroll early to ensure your spot! See this web page for the complete schedule of upcoming Stat-Ease distance-learning courses. To enroll in the workshop that suits you best, click Register, or email our Lead Client Specialist Rachel Poleke at
PS If you lead a group of 6 or more colleagues, save money and customize content via a private workshop. For a quote, email
“The weaker the data available upon which to base one's conclusion, the greater the precision which should be quoted in order to give the data authenticity.”
—Law XXXV from Norman R. Augustine, aeronautical engineer, former CEO of Lockheed Martin, published in Augustine's Laws, Sixth Edition 6th Edition, 1997.
Stat-Ease, Design-Expert and Statistics Made Easy are registered trademarks of Stat-Ease, Inc.

Circulation: Over 4000 worldwide
Copyright © *|CURRENT_YEAR|* *|LIST:COMPANY|*, All rights reserved.

Our mailing address is:

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.