Stat-Ease
If you are having trouble viewing this email view it online.
 
Vol: 12 | No: 6 | Nov/Dec'12
Stat-Ease
The DOE FAQ Alert
     
 

Heads-up (below!)
The key to science provided by one of the top physicists of all time

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 previous DOE FAQ Alerts click here.

TIP: Get immediate answers to questions about DOE via the Search feature on the main menu of the Stat-Ease® web site. This pores not only 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 this registration page.

Also, Stat-Ease offers an interactive web site—The Support Forum for Experiment Design. Anyone (after gaining approval for registration) can post questions and answers to the Forum, which is open for all to see (with moderation). Furthermore the Forum provides program help for Design-Ease® and Design-Expert® software. Check it out and search for answers. If you come up empty, do not be shy: Ask your question! Also, this being a forum, we encourage you to weigh in with answers! The following Support Forum topic provides a sampling of threads that developed since my last Alert:

  • Area: Forum Rules, Topic: “State of the Forum”, Question: “We here at Stat-Ease would like to thank those of you who have participated on the forum, especially those who have taken the time to answer questions. Given a little more time this forum has to the potential to grow into something beneficial not only for Stat-Ease customers, but statistics in general. Keep up the good work. Together we'll make STATistics as EASy as possible.” --Wayne

To open yet another avenue of communications with fellow DOE aficionados, sign up for The Stat-Ease Professional Network on Linked In and start or participate in discussions with other software users. A recent thread features nearly two dozen answers spelling out why DOEs fail such as this list posted by a Director of Technology & Quality for a major oil company:

“Here are the top 3 reasons I have seen in the past 25 years for "failure" (meaning not achieving the experimental goals):

  • “Not knowing enough about the system to pick the appropriate design and/or levels (having been too impatient to do screening experiments or other pre-work)
  • Performing a doomed experiment because of unaddressed noise then leaving DOE with a bad reputation that causes others (like managers) not to believe in its power
  • Not choosing DOE in the first place because knowing the number of runs up front makes an accurate estimate of cost possible and can never compete with the unrealistic expectations of trial and error.”
 
Stats Made Easy Blog

StatsMadeEasy offers wry comments weekly from an engineer with a bent for experimentation and statistics. Simply enter your e-mail in the forwarding field at www.StatsMadeEasy.net and get new StatsMadeEasy entries delivered directly to your inbox. Or, click this link to:

Subscribe with Feedburner

“Your StatsMadeEasy blogs brighten up a dreary work day...”
—Applied Statistician, Florida :-)

Topics discussed since the last issue of the DOE FAQ Alert (latest one first):

Please do not be shy about adding your take about any news or views you see in StatsMadeEasy.  Thanks for paying attention.

 

 
 


If this newsletter prompts you to ask your own questions about DOE, please address them via e-mail to: [email protected].


-

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

1:  FAQ: Confidence interval versus prediction interval
2:  FAQ: How to interpret big day-by-day differences blocked out in the experiment design
3:  Webinar alert: Learn some tricks of the trade from Real-life DOE—Encore for India and AustralAsia
4:  Events alert: Look forward to the 2013 schedule of Stat-Ease talks and exhibits
5:  Workshop alert: Get a Stat-Ease workshop on your 2013 training budget: Tool up!
 
 


PS. Quote for the month: The key to science provided by one of the top physicists of all time.
(Page down to the end of this e-zine to enjoy the actual quote.)



- Back to top -


1: FAQ: Confidence interval versus prediction interval

Original Question:

From a Manufacturing Engineer:
“I have a question about confidence intervals (CI) versus prediction intervals (PI).  I have looked at the book and the online forum.  I'm still confused.  I had thought that the CI was used for the averages of any future observations, so the mean of future observations would land within the CI.  I was also thinking that only a single, individual future observation would fall within the PI, so single observations would land within the PI.  Is this accurate?  The book makes it seem as though only the PI is used for future observations.”

Answer:

From Stat-Ease Consultant Wayne Adams:

“You are not alone in your troubles understanding what a CI versus PI really mean.

  • The confidence interval is about the average, the TRUE average.  The 95% part means that if an infinite number experiments were conducted and 95% confidence intervals calculated for each one at the same factor settings, then 95% of the intervals will contain the true mean at that factor setting. 
  • The prediction interval is about a future sample’s average.  When the sample size is one, it is the individual observation.  The 95% part is the same as above, just for a future sample’s average.  Future is usually taken to mean next sample.”

P.S. from Mark: For further details, see this two-part series by Stephen N. Luko and Dean V. Neubauer for ASTM Standardization News on confidence intervals versus prediction intervals.  Also, learn how Design-Expert® software generates appropriate prediction intervals by following this link to FAQ 4 in the Sep/Oct'11 DOE FAQ Alert, which tells users “what to make of multiple confirmation runs.”


- Back to top -


2: FAQ: How to interpret big day-by-day differences blocked out in the experiment design

Original Question:

From a Six Sigma Master Black Belt:
“Someone brought results from their DOE to me to analyze.  The lab equipment requires set up each day, so they blocked the runs accordingly.  It turned out that day-to-day differences were big, which leads to these two questions:

  1. Is it statistically sound to calculate an F ratio for the blocks?  In this case, I would get a large value suggesting that the blocks effect is very significant.
  2. The block coefficients turn out to be on the same order of magnitude as the coded model coefficients.  Is it valid to assume then that the block effect is just as important as the variables that were varied intentionally?”

Answer:

From Stat-Ease Consultant Brooks Henderson:

“Those are exactly the things we teach in our class to determine whether the blocking was ‘significant' or not.  You are right in making your comparisons.  We don’t print an F-value because one of the ANOVA assumptions is that the residuals are randomly distributed.  By restricting the randomization in blocking, that assumption is invalid.  However, it doesn’t stop us from using it as a rule of thumb.  We usually say that if you divide the Block mean square by the residual mean square and the ratio is 4 to 1 or higher, then the block effect was sizeable.”

The other thing you mention in part B is the another benchmark.  If the block coefficient is as big or bigger than your coded coefficients (which span the whole factor range), then the block effects are quite substantial.”

(Learn more about blocking by attending the two-day computer-intensive workshop Experiment Design Made Easy.  Click on the title for a description of this class and link from this page to the course outline and schedule.  Then, if you like, enroll online.)


- Back to top -


3: Webinar alert: Learn some tricks of the trade from Real-life DOE—Encore for India and AustralAsia

On Monday, December 3 at 9 PM USA-CST* I will reveal some tricks of our trade for Real-Life DOE to our friends on the other side of the globe.  If you are just beginning with factorial screening and characterization experiments, this webinar is for you!  More advanced practitioners might also glean an “aha!” as well, and/or follow up afterward with suggestions that I can share via the DOE FAQ Alert.

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 Dulski, via [email protected].  If you can be accommodated, she will provide immediate confirmation and, in timely fashion, the link with instructions from our web-conferencing vendor GotoWebinar.

*(To determine the time in your zone of the world, try using this link.  We are based in Minneapolis, which appears on the city list that you must manipulate to calculate the time correctly.)


- Back to top -


4: Events alert: Look forward to the 2013 schedule of Stat-Ease talks and exhibits

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

  1. Life Science Alley 2012, Minneapolis, MN, December 5, 2012
    Booth 703
  2. ASA Conference on Statistical Practice 2013, New Orleans, LA, February 21-23, 2013
  3. MN ASQ Professional Development Summit, Brooklyn Park, MN, February 26-27, 2013
    Talk by Brooks Henderson, "Quality by Design (QbD) for Everyone"
  4. ASQ WCQI/ICQI, Indianapolis, IN, May 6-8, 2013
    Talk by Mark Anderson, "Quality by Design (QbD) for Pharmaceuticals and Beyond"
    Booth 608
  5. Quality & Productivity Research Conference (QPRC), Schenectady, NY, June 5-7, 2013
  6. 59th World Statistics Congress (WSC), Hong Kong, China, August 25-30, 2013
    Talk by Mark Anderson, "Practical Considerations of Algorithmic Design"
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 reimbursement for travel expenses.  In any case, it never hurts to ask Stat-Ease for a speaker on this topic.


- Back to top -


5: Workshop alert: Get a Stat-Ease workshop on your 2013 training budget: Tool up!

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!  Also, take advantage of a $395 discount when you take two complementary workshops that are offered on consecutive days.

All classes listed below will be held at the Stat-Ease training center in Minneapolis unless otherwise noted.

* Take both EDME and MIX to earn $395 off the combined tuition!

** Take both EDME and RSM to earn $395 off the combined tuition!

*** Take both MIX and MIX2 to earn $395 off the combined tuition!

See this web page 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 [email protected].


- Back to top -


Please do not send me requests to subscribe or unsubscribe—follow the instructions at the very end of this message.
I hope you learned something from this issue. Address your general questions and comments to me at: [email protected].

Sincerely,

Mark

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


PS. Quote for the month—The key to science provided by one of the top physicists of all time:


"
If it disagrees with experiment, it’s wrong. And that simple statement is the key to science. It doesn’t make a difference how beautiful your guess is, it doesn’t matter how smart you are, who made the guess, or what his name is. If it disagrees with experiment, it’s wrong. That’s all there is to it."

Richard Feynman

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, Wayne Adams and Brooks Henderson
—Statistical advisor to Stat-Ease: Dr. Gary Oehlert
Stat-Ease programmers led by Neal Vaughn
—Heidi Hansel Wolfe, Stat-Ease marketing director, Karen Dulski, and all the remaining staff that provide such supreme support!

Twitter-SmileyFor breaking news from Stat-Ease go to this Twitter site.

DOE FAQ Alert ©2012 Stat-Ease, Inc.
Circulation: Over 6200 worldwide
All rights reserved.

 
  Subscribe