Issue:
Volume 3, Number 7
Date: July
2003
From:
Mark J. Anderson, Stat-Ease,
Inc. (www.statease.com)
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, please click on the links at
the bottom of this page.
Feel free to forward this newsletter to your colleagues. They can
subscribe by going to http://www.statease.com/doealertreg.html.
If this newsletter prompts you ask to your own questions about DOE,
please address them to stathelp@statease.com.
Here are some good-old summertime appetizers to
get this Alert off to a good start:
1. Click http://makeashorterlink.com/?N1AE53405
to see an unusual way to make ice cream suggested by Theodore Gray,
a co-founder of Wolfram Research, who publishes Mathematica software.
2. If you decide to buy a cone instead and end up juggling change
while trying to keep the ice cream from melting, consider how different
denominations could save on coins:
18 cents for USA
83 cents for Canada
1.33 or 1.37 Euros for Europe
Believe it or not, these would be optimal according to an article
at http://www.sciencenews.org/20030510/mathtrek.asp.
3. Flash! See http://www.msnbc.com/news/931938.asp#BODY
(page no longer available) for startling news on lightning jets!
Here's what I cover in the body text of this DOE
FAQ Alert (topics that delve into statistical detail are designated
"Expert"):
1. FAQ: Graphing two responses
to view correlation
2. Expert-FAQ: Why do replicated
center-points in a central composite design (CCD) for response surface
methods (RSM)?
3. Heads-up: Scary e-mails with
a scientific flavor -- Read all about a typical one on the perils
of plastic, but do not pass it on!
4. Events alert: Stat-Ease is
exhibiting and hosting a roundtable discussion at the Joint Statistical
Meetings in San Francisco
5. Workshop alert: We are coming
to San Jose later next month, Philadelphia in September, plus much
more
PS. Quote for the montha
more concise view from a Mars rocket scientist on "known unknowns",
etc. (followup to last month's quote from Rumsfeld)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1. FAQ: Graphing two responses to view
correlation
-----Original Question-----
From: Singapore
"Recently I bought Design-Expert® software. It is working
well and caters to my needs. Presently, the software allows me to
study interactions between the different factors for a specific response.
I wonder if there is any way to view relationships between the different
responses."
Answer (from Stat-Ease
consultant Pat Whitcomb):
"When you are on the Design layout (the spreadsheet-like view)
in Design-Expert (or Design-Ease® software) go to the main menu
and choose View, Graph Columns. You can plot any two columns from
your experimental matrix, including a pair of responses, against each
other. Look for correlations."
P.S. The software reports the statistic "r"a
measure of correlation that varies from -1 (inverse relationship)
to +1 (directly related). When "r" is zero, the two columns
plotted are said to be "uncorrelated." This occurs when
you perform standard two-level designs, or those laid out "orthogonally"
via Taguchi.
Mark
(Learn more about basic statistics and two-level
designs by attending the 3-day computer-intensive workshop "Experiment
Design Made Easy." See http://www.statease.com/clasedme.html
for a complete description. Link from this page to the course outline
and schedule. Then, if you like, enroll online.)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2. FAQ: Why
do replicated center-points in a central composite design (CCD) for
response surface methods (optimization)?
-----Original Question-----
From: Alabama
"When I set up a central composite design (CCD) for response
surface methods (RSM), the software requested I run the center-point
many times. Why should I do that? Would it be OK just to re-test the
product, or do I have to re-make it?"
Answer:
The CCD you requested from Design-Expert
calls for 6 center-points intended to be run at random intervals with
fabrication done from start to finish, not just re-sampled and/or
re-tested. The book that Pat and I co-authored, "DOE Simplified"
(see http://www.statease.com/doe_simp.html),
provides an example of CCD (in Chapter 8) on confetti that I cut to
varying dimensions before measuring flight time. In this case, I re-cut
the confetti for each replicate of the center-points, the number of
which is specified by statisticians to give a good balance of information
in the middle of the space versus the outer regions. In a related
study, Box and Liu ("Product Design with Response Surface Methods,"
Center for Quality and Productivity Improvement, Report No. 150, May
1998 at http://www.engr.wisc.edu/centers/cqpi/),
cut paper helicopters but then re-dropped them many times. These repeated
measures (not true replication of a run) then were averaged and this
value entered as a response. Furthermore, Box and Liu computed the
standard deviation and entered this as a second response to determine
if some configurations proved to be more variable than othersan
issue related to stable vs. unstable aircraft design.
For more details, see a write-up by fellow Stat-Ease
consultant Pat Whitcomb in the June 1998 Stat-Teaser entitled "Center-Points
in Central Composite Design" at http://www.statease.com/news/news9806.pdf.
P.S. Another Stat-Ease consultant, Shari Kraber,
commented:
"Replication of center-points provides these advantages:
Reduces the prediction error in the presumed optimum of the
design space. (It's assumed that the CCD will be centered on the 'optimum',
so you want the best prediction capability there.)
Provides pure error, used for the lack-of-fit (LOF) test. Simply
re-testing the product underestimates the pure error, making the LOF
test inaccurate and therefore worthless."
(Learn more about RSM designs by attending
the "Response Surface Methods for Process Optimization"
workshop. For a description, see http://www.statease.com/clas_rsm.html.
Link from this page to the course outline and schedule. You can enroll
online by linking to the Stat-Ease e-commerce page for workshops.)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3. Heads-up: Scary e-mails with scientific
flavor -- Read all about typical one on perils of plastic, but do
not pass it on!
-----Original Question-----
From: Non-scientific person
"I just received an e-mail that worries
me very much. It says that plastic wrap causes cancer. Can you find
out if this is true?"
Answer (from Stat-Ease
consultant Pat Whitcomb):
My colleague Tryg helped me track down a
web site that reproduces the widely-circulated letter with the supposed
studies showing carcinogenic properties of plastic wrap and the rebuttal.
See it for yourself at http://urbanlegends.about.com/library/bl-microwave-dioxin.htm.
Go back to the main site to clarify the truth (if any) behind other
Internet-spread rumors which seem to run rampant all over the globe.
Here's my first response to the question, before
Tryg found the specific answer:
"It would take a lot of digging to follow-up on all the issues
brought up by this inflammatory e-mail. I'd be very surprised if there's
anything to it. Generally these things start with enough elements
of truth to sound plausible but from there they play on fears in an
unscrupulous way.
It boils down to this: Do not believe anything
you read that's not reviewed by peers, such as technical articles
in reputable scientific journals. Even reputable scientists may publish
results that later prove to be erroneous, so (in my opinion!) it pays
to be skeptical at all times about everything."
I received this reply from the inquirer:
"Thank you! I didn't forward it along and I'm glad now."
In my opinion, anything you people reading
this newsletter can do to quell rumors like this would be a service
to humankind. Mark
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4.
Events alert: Stat-Ease exhibiting and hosting roundtable discussion
at Joint Statistical Meetings in San Francisco
Second notice:
See http://www.amstat.org/meetings/jsm/2003/
for details on this year's Joint Statistical Meetings (JSM) on August
3-7, 2003 in San Francisco, California. JSM is the largest gathering
of statisticians held in North America. It is sponsored by the American
Statistical Association (ASA), the International Biometric Society
(ENAR and WNAR), the Institute of Mathematical Statistics, and the
Statistical Society of Canada. Activities include "oral presentations,
panel sessions, poster presentations, continuing education courses
and an exhibit hall with state-of-the-art statistical products"
(such as Design-Expert software at booth
205!). A Stat-Ease consultant will lead roundtable Session
140 over lunch on Monday 12:30 PM, sponsored by the Section on Quality
& Productivity. The topic is "Practical versus Statistical
Aspects of Altering Central Composite Designs." See http://makeashorterlink.com/?Z336269C4
for more details. It will be interesting to share knowledge with expert
practitioners of response surface methods (RSM).
Click http://www.statease.com/events.html
for a list of where Stat-Ease consultants will be giving talks and
doing DOE demos. We hope to see you sometime in the near future!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5. Workshop alert: Coming
to San Jose later next month, Philadelphia class in September, plus
much more
Second notice:
I am slated to teach the August 5-7 Experiment Design Made Easy (EDME)
workshop in San Jose, California. I hope to see a good turn-out of
students from the Far West and elsewhere. Mark
For you East Coasters we offer Experiment
Design Made Easy on September 9-11 in Philadelphia, PA. For other
workshops and other locations (mainly at our home training site in
Minneapolis), see http://www.statease.com/clas_pub.html
for a schedule. To enroll, click the "register online" link
on our web site or call Stat-Ease at 1.612.378.9449. 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. Call us to get a quote.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I hope you learned something from this issue.
Address your general questions and comments to me at:
mark@statease.com
Sincerely,
Mark
Mark J. Anderson, PE, CQE
Principal, Stat-Ease, Inc. (http://www.statease.com)
Minneapolis, Minnesota USA
PS. Quote for the
montha more concise view from a Mars rocket scientist
on "known unknowns", etc. (followup to last month's quote
from Rumsfeld):
"We want to research
what we call the 'known unknowns.' This will reduce total risk in
the face of unknown unknowns, the true surprises out there."
John Charles,
senior staff scientist at NASA's Space and Life Sciences Directorate,
speaking about a possible voyage by humans to Mars
Trademarks: Design-Ease, Design-Expert and Stat-Ease
are registered trademarks of Stat-Ease, Inc.
Acknowledgements to contributors:
Students of Stat-Ease training and users of Stat-Ease software
Fellow Stat-Ease consultants Pat Whitcomb and Shari Kraber (see
http://www.statease.com/consult.html for
resumes)
Statistical advisor to Stat-Ease: Dr. Gary Oehlert (http://www.statease.com/garyoehl.html)
Stat-Ease programmers, especially Tryg Helseth (http://www.statease.com/pgmstaff.html)
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 (see above)
Click here to add your name to the FAQ
DOE Alert newsletter list server.
Statistics Made Easy
DOE FAQ Alert ©2003 Stat-Ease, Inc.
All rights reserved.