PMI BLOG

Data interpretation, measures for learning

Author: Des Kelly

or..How to Avoid Being the Chicken Licken of Management

A classic story of the risks of poor data interpretation is told in the folk tale with a moral, Chicken-Licken. When an acorn falls from a tree on to Chicken-Licken’s ‘bald pate’, he looks up into the clear blue sky and concludes that ‘the sky is falling’. Crafty Fox-lox sees an opportunity to feed his family by capitalising on the hapless hero’s poor judgement and Chicken Licken gets eaten.

Chicken-Licken took a piece of data, developed a theory and made a prediction of the future. He misinterpreted the data, which led to a judgement, which became a theory, which became fact. A high-risk strategy!

While opinion supported by correctly analysed and interpreted data is knowledge, opinion without correctly analysed and interpreted data is just that, an opinion.

Measures provide data that leaders can use to generate theories. Theories can be tested and used in due course to enable the decision-making process and ultimately determine the success or even potential failure of that business.

Measures have the potential to be very powerful resources but must always be seen in the context of business improvement and not used for punishment or reward. There is a very real risk of destroying the bond between the leader and the employee which will ultimately shut off the stream of knowledge that keeps the business sustainable and able to grow.

Understanding ‘whats going on’

One of the key skills leaders’ need is the ability to understand what is going on, interpret what data is telling them, develop and test robust theories, predict outcomes and decide. Simply relying on ‘gut’ instinct, experience or opinion alone to form the basis of these decisions can lead to catastrophic results (think of Motorola who, despite having a long history of innovation and sustained growth in the late 90s, ignored data which indicated the rise of digital mobile phones and instead invested heavily in analogue technology).

When a leader can understand what the data is communicating and can discern what is true or right coupled with the good sense and insight to understand what the right actions should be, they have what is known as ‘occupational wisdom’ – the light that shines so others may follow.

The alternative to the development of occupational wisdom is to use chance as the determinant of decisions. This is better known as gambling.

Making knowledge-based decisions

How can we ensure that our that our theories and decisions are knowledge-based?
How do we avoid becoming the Chicken Licken of management?
How can leaders develop occupational wisdom and cast-off the gambling technique?

We can gain insight into this from a small number of very perceptive thinkers.

One who had nothing to do with management and organisations but had great insight was Sir Arthur Conan Doyle who gave Sherlock Holmes the words of wisdom;

I never guess. It’s a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.

In management, decisions are based on robust theories that have been devised from facts that emerge from the rigorous analysis of the data.

W.E Deming the management guru frequently said “Management is prediction” in the forward for the book Quality Improvement Through Planned Experimentation, by Ronald Moen, Thomas Nolan, and Lloyd Provost Deming wrote:-

Prediction is the problem, whether we are talking about applied science, research and development, engineering, or management in industry, education, or government…The question is, ‘What does the data tell us? How do they help us to predict?

It’s surprisingly easy.

3 steps to wisdom!

PDSAThe leader only need to have an appreciation of three things;

  1. To understand that all data gathered through measurement will vary. The trick is to understand what this variation is trying to tell us.
  2. To appreciate that good theories are the consequence of continual learning. As articulated through the PDSA model.
  3. To know that for the second to be good you need to be able to do the first!

The great news is that to benefit from understanding variation you don’t have to be a scholar or a statistician. You just need the willingness to see what the data is truly telling you.

The technique developed to help leaders predict the likely outcome is more frequently called today Process Performance charts. Historically known as Control charts these use very simple number crunching to permit simple analysis to present profound knowledge. It’s a straightforward concept because there are only two types of variation, Common Cause and Assignable Cause. What we want to do is determine when we have either.

When the leader understands this simple concept, wisdom follows.

Learn more about control charts:
Watch our Videos
Attend a webinar
Learn about control charts on our Lean Six Sigma Green Belt course.


About the author

Des Kelly

Des Kelly is a Director Consultant at PMI with over 25 years experience helping clients around the world to develop successful improvement strategies. He has led transatlantic transformation programmes for some of the world’s largest organisations.

If you have any questions for Des, please do get in touch.

Follow Des on LinkedIn

 

Menu