|
|
|
|
|
|
posts relating to data, data based decisions, statistics, SPC, data quality, data analysis, data collection... It is critical to understand common and special cause variation.
Recommended posts: Data is Only a Proxy - Targets Distorting the System - Measurement and Data Collection - Evidence-based Management - Visible Data - Understanding Data
Related: Operational definition - Control chart
This is an interesting video on Deming and American management (by the BBC in 1992): Prophet Unheard. It includes some nice old footage of Deming in Japan. The importance of respect for people is clear and the video also touches on the idea the danger of relying on data (when you do not understand variation and that many important matters and unmeasurable). The video features many snippets of Dr. Deming speaking and includes Don Peterson, Ford CEO; Clare Crawford Mason, If Japan Can, Why Can’t We producer; and Myron Tribus.
Related: Dr. Deming Webcast on the 5 Deadly Diseases – Red Bead Experiment Webcast – Performance without Appraisal – management webcasts
Part two of the documentary explores the Deming Prize, understanding data and the PDSA cycle:
(more…)
Once again the USA was the leading country in manufacturing for 2008. And once again China grew their manufacturing output amazingly. In a change with recent trends Japan grew output significantly. Of course, the 2009 data is going to show the impact of a very severe worldwide recession.
Chart showing the percentage output of top manufacturing countries from 1990-2008 by Curious Cat Management Blog, Creative Commons Attribution.The first chart shows the USA’s share of the manufacturing output, of the countries that manufactured over $185 billion in 2008, at 28.1% in 1990, 27.7% in 1995, 32% in 2000, 28% in 2005, 28% in 2006, 26% in 2007 and 24% in 2008. China’s share has grown from 4% in 1990, 6% in 1995, 10% in 2000, 13% in 2005, 14% in 2006, 16% in 2007 to 18% in 2008. Japan’s share has fallen from 22% in 1990 to 14% in 2008. The USA has about 4.5% of the world population, China about 20%. See Curious Cat Investment blog post” Data on the Largest Manufacturing Countries in 2008.
Even with just this data, it is obvious the belief in a decades long steep decline in USA manufacturing is not in evidence. And, in fact the USA’s output has grown substantially over this period. It has just grown more slowly than that of China (as has every other country), and so while output in the USA has grown the percentage with China has shrunk. The percentage of manufacturing output by the USA (excluding output from China) was 29.3% in 1990 and 29.6% in 2008. The second chart shows manufacturing output over time.
Chart showing the output of the top manufacturing countries from 1990-2008 by Curious Cat Management Blog, Creative Commons Attribution.The 2008 China data is not provided for manufacturing alone (the latest UN Data, for global manufacturing, in billions of current USA dollars). The percentage of manufacturing (to manufacturing, mining and utilities) was 78% for 2005-2007 (I used 78% of the manufacturing, mining and utilities figure provided in the 2008 data). There is a good chance this overstates China manufacturing output in 2008 (due to very high commodity prices in 2008).
Hopefully these charts provide some evidence of what is really going on with global manufacturing and counteracts the hype, to some extent. Global economic data is not perfect. These figures are an attempt to capture the economic reality in the world but they are not a perfect proxy. This data is shown in 2008 USA dollars which is good in the sense that it shows all countries in the same light and we can compare the 1995 USA figure to 2005 without worrying about inflation. However foreign exchange fluctuations over time can show a country, for example, having a decline in manufacturing output in some year when in fact the output increased (just the decline against the USA dollar that year results in the data showing a decrease – which is accurate when measured in terms of USA dollars).
If the dollar declines substantially between when the 2008 data was calculated and the 2009 data is calculated that will give result in the data showing a substantial increase in those countries that had a currency strengthen against the USA dollar. At this time the Chinese Renminbi has not strengthened while most other currencies have – the Chinese government is retaining a peg to a specific exchange rate.
Korea (1.8% in 1990, 3% in 2008), Mexico (1.7% to 2.6%) and India (1.4% to 2.5%) were the only countries to increase their percentage of manufacturing output (other than China, of course, which grew from 3.9% to 18.5%).
Related: posts on manufacturing – Global Manufacturing Data 2007 – Global Manufacturing Employment Data – 1979 to 2007 – Top 10 Manufacturing Countries 2006 – Top 10 Manufacturing Countries 2005 – lean manufacturing resources
The JMP blog has posted some highlights from George Box’s presentation at Discovery 2009
Having worked and studied with many other famous statisticians and analytic thinkers, Box did not hesitate to share his characterizations of them. He told a story about Dr. Bill Hunter and how he required his students to run an experiment. Apparently a variety of subjects was studied [see 101 Ways to Design an Experiment, or Some Ideas About Teaching Design of Experiments]
…
According to Box, the difficulty of getting DOE to take root lies in the fact that these mathematicians “can’t really get the fact that it’s not about proving a theorem, it’s about being curious about things. There aren’t enough people who will apply [DOE] as a way of finding things out. But maybe with JMP, things will change that way.”
George Box is a great mind and great person who I have had the privilege of knowing my whole life. My father took his class at Princeton, then followed George to the University of Wisconsin-Madison (where Dr. Box founded the statistics department and Dad received the first PhD). They worked together building the UW statistics department, writing Statistics for Experimenters and founding the Center for Quality and Productivity Improvement among many other things.
Statistics for Experimenters: Design, Innovation, and Discovery shows that the goal of design of experiments is to learn and refine your experiment based on the knowledge you gain and experiment again. It is a process of discovery. If done properly it is very similar to the PDSA cycle with the application of statistical tools to aid in determining the impact of various factors under study.
Related: Box on Quality – George Box Quotations – posts on design of experiments – Using Design of Experiments
There is no true value of anything: data has meaning based on the operational definition used to calculate the data.
Walter Shewhart’s Statistical Method from the Viewpoint of Quality Control, forward by W. Edwards Deming:
Dr. Deming’s ideas on the theory of knowledge are the least understood and least seen in other management systems. The importance of understanding what data does, and does not tell you, is at least somewhat acknowledged in other management system but is often not found much in the actual practice of management. The execution often glosses over the importance of actually understanding statistics versus using formulas. Just using formulas is dangerous. It may be inconvenient but learning about the traps we can fall into in using data is important.
How often do you see the operational definition used to calculate the data you see with the data you are provided?
via: Shewhart, Deming and Data by Malcolm Chisholm
Related: How We Know What We Know – Pragmatism and Management Knowledge – Measuring and Managing Performance in Organizations – Dangers of Forgetting the Proxy Nature of Data
Interesting lecture on Statistical Learning as the Ultimate Agile Development Tool by Peter Norvig. The webcast is likely to be of interest to a fairly small segment of readers of this blog. But for geeks it may be interesting. He looks at the advantages of machine learning versus hand programming every case (for example spelling correction).
Google translate does a very good job (for computer based translation) based on machine learning. You can translate any of the pages on this blog into over 30 languages using Google translate (using the widget in the right column).
Via: @seanstickle
Related: Mistakes in Experimental Design and Interpretation – Does the Data Deluge Make the Scientific Method Obsolete? – Website Data – An Introduction to Deming’s Management Ideas by Peter Scholtes (webcast)
___________________
Anscombe’s quartet comprises four datasets that have identical simple statistical properties, yet are revealed to be very different when inspected graphically. Each dataset consists of eleven (x,y) points. They were constructed in 1973 by the statistician F.J. Anscombe to demonstrate the importance of graphing data before analyzing it, and of the effect of outliers on the statistical properties of a dataset.
Of course we also have to be careful of drawing incorrect conclusions from visual displays.
For all four datasets:
| Property | Value |
|---|---|
| Mean of each x variable | 9.0 |
| Variance of each x variable | 10.0 |
| Mean of each y variable | 7.5 |
| Variance of each y variable | 3.75 |
| Correlation between each x and y variable | 0.816 |
| Linear regression line | y = 3 + 0.5x |
Edward Tufte uses the quartet to emphasize the importance of looking at one’s data before analyzing it in the first page of the first chapter of his book, The Visual Display of Quantitative Information.
Related: Great Charts – Simpson’s Paradox – Seeing Patterns Where None Exists – Visible Data – Control Charts – Edward Tufte’s: Beautiful Evidence
The system is responsible for 90, 92, 94, 97% of problems – W. Edwards Deming. Fix the system, don’t blame the people. When you seek system fixes you approach situations differently than if you search for people to blame.
By the way, I am often asked about the data supporting Deming’s contention that the system was responsible for 97% of the problems. This statement was not based on a set of data but on Dr. Deming’s decades of experience. And he increased the percentage over time – as he learned more.
Roads that are designed to kill
One of the ways they began to protect people was to put barriers down the center of two-lane roads. They showed that this could be done cheaply. When Mylar – a strong polyester film – is supported by closely spaced plastic poles, it can keep cars from crossing the median. When the Swedes used this type of center barrier to separate the traffic going in opposite directions, they effectively prevented head-on collisions and the death rate on these roads fell by 70 percent to 80 percent.
Global health research shows more improvements can save lives. For example, Ghana put in rumble strips – small bumps spaced closely together – across all the roads leading into the capital city of Accra, reducing fatalities by 35 percent. Research has shown that speed bumps on roads are one of the “best buys” in all of global health.
…
Most people think we are doing all that can be done to keep our roads safe. They are wrong. Road traffic injuries kill more than a million people a year worldwide, including 40,000 a year in the United States.
Is a situation killing 40,000 people in the USA a year a health care issue? It sure seems to me it would be. It probably isn’t a disease management issue though (some might try to say bad roads are a disease but I wouldn’t say that). I think this is one, of many examples, that shows that we have a disease and injury management system not a health care system (in addition to illustrating systems thinking, effective root cause analysis, PDSA, innovation, respect for people…).
Related: Find the Root Cause Instead of the Person to Blame – Traffic Congestion and a Non-Solution – Checklists Save Lives – Saving Lives: US Health Care Improvement – The Economic Benefits of Walkable Communities – SWAT Raid Signs of Systemic Failures – System Improvement to Respond to the Dynamics of Crowd Disasters – The Leading Causes of Death
Google does a great job of using statistical and engineering principles to improve. It is amazing how slow we are to adopt new ideas but because we are it provides big advantages to companies like Google that use concepts like design of experiments, experimenting quickly and often… while others don’t. Look Inside a 1,024 Recipe Multivariate Experiment
via: @hexawise – My brother has created a software application to provide much better test coverage with far fewer tests using the same factorial designed experiments ideas my father worked with decades ago (and yet still far to few people use).
Related: Combinatorial Testing for Software – Statistics for Experimenters – Google’s Website Optimizer allows for multivariate testing of your website. – Using Design of Experiments
Making things visible is a key to effective management. And data in computers can be easy to ignore. Don’t forget to make data visible. Paul Levy, CEO of Beth Israel Deaconess Medical Center in Boston recently hosted Hideshi Yokoi, president of the Toyota Production System Support Center and wrote this blog post:
The highlight? At one point, we pointed out a new information system that we were thinking of putting into place to monitor and control the flow of certain inventory. Mr. Yokoi’s wise response, suggesting otherwise, was:
“When you put problem in computer, box hide answer. Problem must be visible!”
The mission of the Toyota Production System Support Center to share Toyota Production System know-how with North American organizations that have a true desire to learn and adopt TPS.
Related: The Importance of Making Problems Visible – Great Visual Instruction Example – Health Care the Toyota Way
Where are the Shareholders’ Mansions? CEOs’ Home Purchases, Stock Sales, and Subsequent Company Performance by Crocker H. Liu, Arizona State University and David Yermack, New York University – Stern School of Business
That we put in power CEO’s that see themselves as nobility with the right to build castles (and many of these CEO castles dwarf all but the most conspicuous castle built by nobility) by taking the wealth produced by others from corporate coffers is a sign of our failure to select acceptable leaders for companies.
Related: Another Year of CEO’s Taking Hugely Excessive Pay – Excessive Executive Pay – Exposing CEO Pay Excesses – Narcissistic Cadre of Senior Executives – 9 Deadly Diseases
When Economic Incentives Backfire by Samuel Bowles, Sante Fe Institute
Punished by Rewards, by Alfie Kohn, is a great book on this topic. The area of “motivating” employees is one it is often hard for managers to learn. Even managers that have been studying Deming, Ackoff, Ohno… for years still have trouble with the idea that trying to find the right incentive scheme to motivate the right behavior is the wrong approach. Read the The Human Side Of Enterprise by Douglas Mcgregor (in 1960) to re-enforce the understanding of human motivation provided by Toyota’s respect for people principles.
Managers need to eliminate de-motivation in the work systems not try and find bonus schemes to motivate behavior. Eliminating de-motivation is often much more work. You can’t just get some money from the bonus pool and start giving it away. You have to manage. But if you are a manager you shouldn’t be afraid to actually manage the system and make it better.
Related: “Pay for Performance” is a Bad Idea – Reward and Incentive Programs are Ineffective — Even Harmful by Peter Scholtes – The Defect Black Market – What’s the Value of a Big Bonus? – Problems with Bonuses – Losses Covered Up to Protect Bonuses – Stop Demotivating Employees
Secret of Googlenomics: Data-Fueled Recipe Brews Profitability
Google continues to make bold moves putting faith in their ability to find innovative solutions that others reject as impossible. It is a challenging but interesting path to success, for them, at least.
Related: Google Should Stay True to Their Management Practices – Google’s Answer to Filling Jobs Is an Algorithm – The Google Way: Give Engineers Room – Google Website Optimizer – Google: Experiment Quickly and Often – posts on innovation in management
Revealed Preference: the preference consumers display by their action, in contrast to what they may say they prefer. While surveys may be useful people often say they will do one thing and actually when given the choice to do so, don’t.
Normally what matters is not what people say they want but what they actually will choose. For that reason revealed preference is a better measure than stated preference. Stated preference is often used as a proxy for actual preference (which may be fine) but it is important to understand that it is just a proxy for actual preference.
See more explanations from the Curious Cat Management Dictionary.
Related: Packaging Improvement – All Models Are Wrong But Some Are Useful – Dangers of Forgetting the Proxy Nature of Data – Confirmation Bias – Be Careful What You Measure
Dr. Deming long ago stated in his 14 obligations of management: “Eliminate numerical goals, numerical quotas and management by objectives.” I think he was right then, and is right now. A goal can help set the scope of the effort. If you are aiming for 2% improvement different strategies may make sense than if you are aiming at 50% improvement. But that type of use is rare. The problem with goals is what actually happens in organizations. They create serious systemic problems and should be avoided (other than in setting the scope). They are deeply ingrained in the way many people think, but we would be better if we could eliminate the use of goals, as they are used now (mainly as arbitrary numerical goals).
Ready, Aim … Fail, Why setting goals can backfire
Today, as the economic situation upends millions of lives, it is also forcing the reexamination of millions of goals – not only the revenue targets of battered firms, but the career aims of workers and students, and even the ambitions of the newly installed administration. And while it never feels good to give up on a goal, it may be a good time to ask which of the goals we had set for ourselves were things we really needed to achieve, and which were things we only thought we should – and what the difference has been costing us.
Related: Measuring and Managing Performance in Organizations – Arbitrary Rules Don’t Work – The Defect Black Market – Goals can Distract from Improvement – Be Careful What You Measure
Dr. Deming used the red bead experiment to present a view into management practices and his management philosophy. The experiment provides insight into all four aspects of Dr. Deming’s management system: understanding variation, understanding psychology, systems thinking and the theory of knowledge.
Red Bead Experiment by Steve Prevette
Related: Fooled by Randomness – Performance Measures and Statistics Course – Performance without Appraisal – Exploring Deming’s Management Ideas – Eliminate Slogans
Lynda Finn, President of Statistical Insight, has written an article on how to create a control chart for seasonal or trending data (where there is an underlying structural variation in the data). Essentially you need to account for the structural variation to create the control limits for the control chart. She also provides a Minitab project file. Both are available for download from the Curious Cat Management Improvement Library.
Related: Control Charts in Health Care – Common Cause Variation – Managing with Control Charts – Measurement and Data Collection – Fourth Generation Management
From the excellent xkcd comic.
Related: Correlation is Not Causation – Does the Data Deluge Make the Scientific Method Obsolete? – Understanding Data – Theory of Knowledge – What Makes Scientists Different 
– Dangers of Forgetting the Proxy Nature of Data – Seeing Patterns Where None Exists
Statistics for Experimenters, second edition, by George E. P. Box, J. Stuart Hunter and William G. Hunter (my father) is now available in Spanish.
Read a bit more can find a bit more on the Spanish edition, in Spanish. EstadÃstica para Investigadores Diseño, innovación y descubrimiento Segunda edición.
Statistics for Experimenters – Second Edition:
Book available via Editorial Reverte
Related: Statistics for Experimenters Review – Correlation is Not Causation – Statistics for Experimenters Data – posts on design of experiments
What’s the Value of a Big Bonus? by Dan Ariely
When I recently presented these results to a group of banking executives, they assured me that their own work and that of their employees would not follow this pattern. (I pointed out that with the right research budget, and their participation, we could examine this assertion. They weren’t that interested.)
This is an interesting look at an effect of bonuses. We all know monetary bonuses can influence behavior. The problem is the type of behaviors that result. Huge bonuses, for example, create huge incentives to risk the future of the company for the chance at a huge bonus for the executive. Extrinsic motivation leads to many problems.
Problems with bonuses: Losses Covered Up to Protect Bonuses – “Pay for Performance” is a Bad Idea – Problems with Bonuses – Book: Punished By Rewards: The Trouble With Gold Stars, Incentive Plans, A’s, Praise, and Other Bribes by Alfie Kohn – posts on executive pay
I continue to tilt at the robber barron CEO pay packages (2007 post on CEO pay abuses).
| 2007 pay rank |
Company | CEO | Pay | 5 Year Pay | CEO % of 2007 Earnings | |
|---|---|---|---|---|---|---|
| 1 | Apple | Steve Jobs | $646,600,000 | $650,170,000 |
|
18.5% |
| 2 | Occidental Petroleum | Ray Irani | $321,640,000 | $509,530,000 |
|
5.9% |
| 3 | IAC | Barry Diller | $295,140,000 | $512,270,000 |
|
Company Lost Money |
| 4 | Fidelity National Financial | William Folley | $179,560,000 | NA |
|
138.4% |
| 5 | Yahoo! | Terry Semel | $174,200,000 | $432,490,000 |
|
26.4% |
| 7 | Countrywide Financial | Angelo Mozilo | $141,980,000 | $295,730,000 |
|
Company Lost Money |
| 13 | XTO Energy | Bob Simpson | $72,270,000 | $215,280,000 |
|
4.2% |
Data via: Forbes CEO Compensation (Total compensation for each chief executive includes the following: salary and bonuses; other compensation, such as vested restricted stock grants, LTIP payouts and perks; and stock gains, the value realized by exercising stock options.) and Google Finance (using 2007 earnings – Countrywide from SEC). I realize this chart could be improved by spending more time (the effect of stock options exercised in one year distorts things a bit but the excess are so massively huge that the clarity of the data does not need to be very precise).
(more…)
Curious Cat Management Improvement Blog © curiouscat.com 2005-2006 powered by WordPress - Curious Cat Alumni Directory
