How Do We Know What We Know? - Deming’s SoPK Part IV

Add bookmark

John Hunter

Understanding how you think - and act - based on what you believe you know to be true, is critical to improving your own decisions as a manager, says contributor John Hunter, in the final of a 4-part series on Deming's system of management, SoPK. And understanding the way your mind works - and common mistakes it makes - goes a long way to getting you there.

Read Part I: Systems Thinking and the Three Musketeers

Read Part II: The Trouble with Motivation

Read Part III: Variation, So Meaningful Yet So Misunderstood

"The greatest obstacle to discovery is not ignorance - it is the illusion of knowledge." Daniel J. Boorstin

What do we know that isn’t so? How can we avoid the mistakes we are in danger of making in our thinking? How can we improve the learning process?

These are some of the questions that led Dr. Deming to include the "theory of knowledge" as one of the 4 pillars of his New Philosophy of Management, sometimes referred to as the System of Profound Knowledge or SoPK. While many of his ideas have found their way into other management theories (focus on the customer, variation, systems thinking, innovation, continual improvement, data based decision making, the importance of psychology...) you rarely hear about the importance of understanding how people think –and act - based on what they believe they know to be true. That is core to a theory of knowledge. To put it in the form of a question, "How do we know that what we think we know is really so?"

There are several simple principles related to the theory of knowledge which will help us avoid traps we often fall into because of how we think and because of what we believe to be true.

Confirmation bias

Confirmation bias is one such important concept, and it means that we tend to latch onto evidence that supports our beliefs and ignore evidence that undermines our beliefs. In order to more effectively adjust our beliefs to reality we are well served to question whether we are falling for confirmation bias. Ask: Is this new evidence really convincing or am I just happy because it supports my existing belief? Am I ignoring other evidence because it calls into question my beliefs?

Experiments, Prediction and Learning

Do you subject your beliefs to tests and experiments? To the extent possible you want to rely on evidence for supporting your theories. Well-run experiments provide great evidence. The model used within the Deming System for Managing to gain evidence and insight is the Plan-Do-Study-Act (PDSA) cycle. The PDSA cycle has incorporated within it an understanding of the theory of knowledge.

People learn better when they predict. Making a prediction forces us to think ahead about the outcomes. Making a prediction also causes us to examine more deeply the system, question or theory we have in mind. Also we learn about our understanding of the management beliefs we hold as we examine the results of our predictions.

For example from your predictions (and from the PDSA experiments which test them) you might discover that:

  • You are often overly optimistic (or pessimistic).
  • You are extremely effective at predicting change related to information technology improvements but poor when predicting the results when psychology (people) play a big role in the I.T. change itself.
  • The impact is often as you predicted for the projects during testing (using PDSA) but attempts to actually standardize the improvements across the organization fail –perhaps because the test was not well designed.

Such learning is important both personally and organizationally. Self-insight is critical to improving your own decisions as a manager, and similarly there is a multiplier effect when people throughout the organization increase their self-insight. Cumulatively, all those better informed decisions result in a better performing organization.

Learning about your ability to predict (and your organization’s ability to predict) is a key part of the theory of knowledge. Let’s face it most of us think we have a much better ability to predict the result of proposed actions than actually turns out to be true. A way we discover the reality is when we require ourselves to make a prediction and to examine the results of the actions.


People believe there is much less variation in systems, processes, actions than there is. This underestimation of variation causes people to believe normal variation is not normal, which in turn causes them to search for special causes for the variation. Doing so is a very low yield strategy for improvement. (See more on Deming's view on tampering).

Again, the Deming System for Managing seeks to avoid this trap in how we think about variation by adopting sensible strategies. One of these strategies is to use control charts to show exactly what is reasonable variation for a process. Another sensible strategy is to examine the process and environment in which a person works first, rather than to blame the person (as a special cause), first. Using such tools and approaches helps us in a several ways:

  • We acknowledge that we have biases that cause us to react in certain –and often unproductive ways
  • The tools themselves help counteract our skewed thinking that cause us to adopt poor yield strategies, such as rating and ranking people based on a simplistic analysis of "results."

Psychology and the theory of knowledge

Deming’s System for Managing is based on four areas: the theory of knowledge, understanding variation, appreciation of a system, and understanding psychology. I want to mention the psychology component briefing because of its interrelationship with theory of knowledge. (All four components are interrelated, in fact.)

Here is an example. The way we evaluate an idea is not based on the cold logic we may like to believe it is. If we hear an idea from someone we don’t like and then two days later we hear the exact same idea from a friend we respect, we tend react to those ideas very differently.

As I started to understand the theory of knowledge and apply what I suggested above, I began to question why I believed certain things and how I responded to ideas that were presented. In examining my own reactions, I noticed that I far too heavily weighed the source of the idea in evaluating the idea, myself. I wasn’t consciously doing so, but in looking at the opinions I took, with the lens of the theory of knowledge, I could see I was, indeed, doing so. And then I started to look at others and it amazed me how often I would hear someone criticize an idea from one person and then praise the same idea a week later from a friend.

This also shows one of the real weaknesses in performance appraisals. When a manager likes a person, the manager is much more likely to appreciate the person’s work, ideas, and contributions and thus give a higher rating to that person.

Misinterpretation and Misunderstanding

Correlation is not causation but we often fall into the belief that correlation does mean causation. Correlation means that two or more things might be observed to occur together. For example, a study might show that "boys who play soccer are more likely to have unprotected sex" — a correlation between playing soccer and unprotected sex — but then people incorrectly conclude that playing soccer causes boys to engage in unprotected sex. That would be causation: cause and effect.

Confirmation bias helps explain why we often fall into this trap of confusing the correlation with causation. Once again, making a prediction not only helps us understand how we learn and how to learn better; it also helps prevent falling into the correlation/causation trap in the first place. For example, analyzing existing spreadsheet data can be useful but it also has dangers, and one risk is that the data already has correlations and assumptions embedded in it. This can cause us to make the incorrect conclusion that the correlations are causations. With an understanding of theory of knowledge we will be more mindful of this trap of confusing causation with correlation –and we can make better predictions and decisions.

There is no true value of any characteristic, state, or condition that is defined in terms of measurement or observation. - Dr. W. Edwards Deming

The "value" is in the context for a given operational definition. Understanding that a value must be interpreted via context, leads us to question any data that doesn’t provide the operational definition for how the data was created. And this leads to better understanding. Otherwise, without having the operational definition we are likely to draw incorrect conclusions from data. Let’s look at a simple example:

The height of senior executives is much greater than the average population. Are tall people inherently much better leaders in organizations? It may have made some sense historically to have the biggest people in the group lead the hunt but it sure doesn’t seem like there is a good reason to believe height is a great value for leaders in contemporary society. You could take from that data about tall senior executives that in fact height is a "value" of great value. Or you could take from it that we continue to make judgments based on unimportant factors because unconscious evaluation criteria (such as height is better) are baked into our psychology and those criteria shape what we believe. Understanding the power of unconscious evaluation critera is a core aspect of Deming’s teachings about theory of knowledge, and why he urged us to state a theory and a prediction, to test it, and to study the results before we put our theory into practice (Plan-Do-Study-Act).

Some beliefs that seem to be baked in to our unconscious, and for which we tend to have a favorable confirmation bias include management practices such as pay-for-performance, management by fear, emphasis on short-term profits, mistaking natural variation for special cause variation, and the use incentives and rewards. (Deming articulated those and others.)

Next Steps

If we can break from such beliefs that are not useful in modern organizations, we can improve our decisions. Having a Deming-based theory of knowledge will help us break from those beliefs and it will help us be more thoughtful as we learn to question other management beliefs we hold –many of which simply are not useful –or cause harm.

Understanding the theory of knowledge within the context of the Deming’s System for Managing helps us more effectively and consistently learn and improve the processes and systems we work with.

Gaining an understanding of how the theory of knowledge is integrated into Deming’s System for Managing is easy for some people, but not easy for others (ah, variation). My intent was for this article to get you started along the trail of discovery but it only scratches the surface of the theory of knowledge because theory of knowledge is a very rich subject.

If you're interested, you can read more about how the theory of knowledge relates to management on my CuriousCat Management Improvement blog.

Copyright 2011 by John Hunter



Deming's SoPK Series on PEX Network

System of Profound Knowledge (SoPK) is the main subject of Deming’s second book on management, The New Economics. SoPK is Deming's system of management and has four interdependent areas:

  • Appreciation for a System –how to lead a system, and systems thinking
  • Knowledge about Variation, including statistical variation
  • Theory of Knowledge -the study of how we know what we know
  • Psychology - understanding the human aspect of management, and especially intrinsic motivation vs. extrinsic motivation.

Dr. Deming pointed out that one need not be an expert in any of the four elements, but viewing the world through the lens of the four elements with some proficiency would provide the viewer with profound knowledge of how to lead, diagnose data and issues, plan for the future, have everyone work together to optimize the system, innovate, and create an exciting win/win environment for customers, suppliers, employees, and managers alike.

Read Part I: Systems Thinking and the Three Musketeers

Read Part II: The Trouble with Motivation

Read Part III: Variation, So Meaningful Yet So Misunderstood


Editor’s Note: The columns published in THE DEMING FILES have been written under the Editorial Guidelines set by The W. Edwards Deming Institute. The Institute views these columns as opportunities to enhance, extend, and illustrate Dr. Deming’s theories. The authors have knowledge of Dr. Deming’s body of work, and the content of each column is the expression of each author’s interpretation of the subject matter.