{domain:"www.qualitydigest.com",server:"169.47.211.87"} Skip to main content

User account menu
Main navigation
  • Topics
    • Customer Care
    • FDA Compliance
    • Healthcare
    • Innovation
    • Lean
    • Management
    • Metrology
    • Operations
    • Risk Management
    • Six Sigma
    • Standards
    • Statistics
    • Supply Chain
    • Sustainability
    • Training
  • Videos/Webinars
    • All videos
    • Product Demos
    • Webinars
  • Advertise
    • Advertise
    • Submit B2B Press Release
    • Write for us
  • Metrology Hub
  • Training
  • Subscribe
  • Log in
Mobile Menu
  • Home
  • Topics
    • 3D Metrology-CMSC
    • Customer Care
    • FDA Compliance
    • Healthcare
    • Innovation
    • Lean
    • Management
    • Metrology
    • Operations
    • Risk Management
    • Six Sigma
    • Standards
    • Statistics
    • Supply Chain
    • Sustainability
    • Training
  • Login / Subscribe
  • More...
    • All Features
    • All News
    • All Videos
    • Contact
    • Training

Lies, Damned Lies, and Teens Who Smoke While Driving

A lack of context is essential to propaganda

Donald J. Wheeler
Mon, 01/02/2012 - 15:56
  • Comment
  • RSS

Social Sharing block

  • Print
  • Add new comment
Body

I

n my February 1996 Quality Digest column I discussed an article out of USA Today. Since that article provides a great example of how we need to filter out the noise whenever we attempt to interpret data, I have updated it for my column today. “Teen Use Turns Upward” read the headline for a graph appearing in USA Today on June 21, 1994. The data in the graph were attributed to the Institute for Social Research at the University of Michigan and were labeled as the “percentage of high-school seniors who smoke daily.”

The portion of this graph covering the 10 years from 1984 to 1993 is shown in figure 1.

https://lh3.googleusercontent.com/_CMMbz69LersWinA1zmZTLgqS7b2Mnu3g3o_N5I1VMeSJm0zQLXTS8BS23dclWUnHy6ulvfJe0wT_LBhe5J6iGUYV426st4UcQdZgvg7CFEyiuHyo9g
Figure 1: Percentage of high-school seniors who smoke daily

Each point in figure 1 is the value found in an annual survey. The 1993 value of 19.0 percent was higher than the 1992 value of 17.3 percent. This was interpreted to mean that more teenagers are using tobacco now than in the past. But are they?

 …

Want to continue?
Log in or create a FREE account.
Enter your username or email address
Enter the password that accompanies your username.
By logging in you agree to receive communication from Quality Digest. Privacy Policy.
Create a FREE account
Forgot My Password

Comments

Submitted by vunderbolt on Wed, 01/04/2012 - 10:34

Wood for my fire

Dear Donald,

Thanks for reminding me of this article. It certainly gives me wood for my fire. (it will be forwarded to some people that should read it) I do a lot of work in marketing and sales. It is possibly the industry where understanding variation should be crucial. I have witnessed Companies "losing" millions by either not even measuring or acting on single events without understanding the context. 

I would be interested in your own and your readers experience in applying XmR charts on sales and marketing data.

Kind Regards

Francois van der Walt

  • Reply

Submitted by dvanputten on Wed, 01/04/2012 - 15:19

In reply to Wood for my fire by vunderbolt

2.66 and 3.27

Hello Dr. Wheeler:

Do the constant values of 2.66 and 3.27 have names? I would like to better understand where these constant values come from.

Thank you, Dirk van Putten

  • Reply

Submitted by dvanputten on Wed, 01/04/2012 - 15:24

In reply to 2.66 and 3.27 by dvanputten

Great Article

Hello Dr. Wheeler:

Great article by the way. I followed all of the math and statistics and understood the genreal concept of noise vs. signal.

Thank you, Dirk

  • Reply

Submitted by dvanputten on Wed, 01/04/2012 - 17:18

In reply to Great Article by dvanputten

Scaling Factors

Hello Dr. Wheeler:

I think I found my answer in your Book, "Twenty Things You Need To Know", Chapter 5, Where Do Scaling Factors Come From?

Thank you, Dirk

  • Reply

Submitted by Tom Hopper on Thu, 01/05/2012 - 04:01

In reply to 2.66 and 3.27 by dvanputten

Constant Values 2.66 and 3.27

Dirk Van Putten, 2.66 is 3 / d_2, where d_2 is an anti-biasing constant that can be found in most textbooks. 3.27 is the D_4 anti-biasing constant.

The Wikipedia page on Indivduals charts (https://en.wikipedia.org/wiki/Individuals_chart) gives a decent overview the charts. Dr. Wheeler's books and other writings go into somewhat greater detail on the constants, as do other books on statistical process control (see, for example, the classic text by Montgomery, though I would caution that Montgomery's general approach to process behavior charts sometimes seems to be at odds with Dr. Wheeler's).

  • Reply

Submitted by JOSE ARREOLA on Thu, 01/05/2012 - 06:21

In reply to Wood for my fire by vunderbolt

XmR charts on sales and marketing data

Hi Francois van der Walt,

I have used XmR charts on sales and marketing data, and I could email examples to you if you want (Spanish). You may also visit Pauls Selden's web page: www.paulselden.com He is author of several works in the field, including "Sales Process Engineering".

  • Reply

Submitted by Donald J. Wheeler on Fri, 01/06/2012 - 07:00

For Francois van der Walt

I have a whole chapter on sales and marketing data in my book Making Sense of Data. This is in addition to the Understanding Variation book cited in the article. I also recommend Dr. Paul Selden's work. Hope this will help. Don Wheeler
  • Reply

Submitted by William A. Levinson on Fri, 01/06/2012 - 10:10

2.66

2.66 is three times the square root of pi divided by 2. The square root of pi over 2 is the estimate of sigma from an average moving range, so three times it yields the width of the 3-sigma control limits for the individuals chart.
  • Reply

Submitted by rgroesbe on Fri, 01/06/2012 - 10:29

Detecting change in successive surveys

The article shows the use of I-MR charts to detect change. With survey data, what is the role of confidence intervals when deciding if there has been a change between successive survey averages?

To use the teens smoking example, what if we had a 99% confidence interval of +/-0.5% for each year of survey data. Then we move from 17.3% in '92 to 19.0% in '93.  Should we be able to say we're 99% confident that the 1992 and 1993 smoking rates are different? Or, in this case, as the I-MR chart shows we cannot conclude there is a change, is our only our conclusion that the difference from 1992 to 1993 is not due to survey error, it's just common cause variation in the underllying process.

  • Reply

Submitted by pillecw on Mon, 01/09/2012 - 19:02

Should be required reading

This article should be required reading for every journalist, journalism major and blogger.  Also every government employee, politician, lobbyist and spind doctor.  Actually, some of those folks probably understand these things and try to use it to advantage (that would be the "damned lies" part).

  • Reply

Add new comment

Image CAPTCHA
Enter the characters shown in the image.
Please login to comment.
      

© 2025 Quality Digest. Copyright on content held by Quality Digest or by individual authors. Contact Quality Digest for reprint information.
“Quality Digest" is a trademark owned by Quality Circle Institute Inc.

footer
  • Home
  • Print QD: 1995-2008
  • Print QD: 2008-2009
  • Videos
  • Privacy Policy
  • Write for us
footer second menu
  • Subscribe to Quality Digest
  • About Us
  • Contact Us