Wednesday, December 23, 2009

Standards and Statistics – In Everyday Use

imageThe other day I was trying to explain the benefits of standardization and the use of statistics to one of my kids.

This led me to write this blog post. I hope it might help you or at least give you a new way to think about Standards and Statistics and how you use them in your everyday lives.

Whether you think about it or not --- We use statistics everyday and we rely upon standards to make our lives easier. I might even go as far to say that standards make our fast paced lives possible.

Why can you drive all the way to work or grandmas house and not remember anything about the drive?

Because there was nothing along the way that caused your brain to interpret a deviation. Nothing caused a reaction. Driving can be mind-numbing. Unless and until something happens to Deviate from the Norm. Then … it’s all up to your reaction time. Hopefully your reactions are fast enough … and your reaction is the right thing to do.

For example, which way do you steer in a skid? If you’ve ever hit black ice … you’ve probably found out very quickly that you steer into the skid. How quickly did you need to react? You responded to a deviation … were you fast enough?

What do I mean? How do we use Statistics and Standard Deviations every day?

  • Every Day Statistics. Whether we are driving down the street or walking on the sidewalk. We use statistics to “predict” the behavior of other people and things. In our cars we expect the person in front of us to keep going – for the most part in a straight line. At traffic lights we expect the light to stay green (or red) for a certain amount of time.
  • We use standards every day too. We expect that the lid will fit on our coffee cup. We expect parking spaces to be wide enough to accommodate our vehicles. Of course, this is not always the case. From a more practical point of view we expect to be able to go to our local hardware store and buy light bulbs or nuts & bolts and we expect them to fit together.
  • Each of these expectations is enabled by standards. Each of these standards is a statistic that has a range that falls within something called the Standard Deviation.

Standard Deviation and You


Hundreds of times a day we unconsciously calculate the standard deviation of the events that unfold around us. We don’t think too hard about the car in front of us – unless and until  - the driver does something unexpected. Something that deviates from what our brains are conditioned to think of as a standard or norm.

Then we react. The Deviation from the Standard Behavior causes a reaction.

. image
How I use Statistics & Standards -
I work in the Enterprise Content Management (ECM) space and one aspect of ECM is the use of workflow or Business Process Management (BPM). What I have said for many years is that a workflow system is only as good as it’s exception handling system. Or put another way – A well designed workflow solution has Graceful Exception Handling.

This is true with our lives too. When an exception is encountered --- something that deviates from the norm --- we are typically required to react. Hopefully you “human workflow system” reacted gracefully.

For example:

  • When driving …
    • In order to avoid an accident we need to react and make real-time decisions.
    • We make assumptions about the other people on the roadway (or sidewalk).
    • We assume they will do certain things. We assume they will NOT do other things

Simple rules like … Driving on the correct side of the road. Not crossing the center line. Using their turn signals to indicate the direction they intend to go.

  • What happens when:
    • A pedestrian steps out into the street – near a crosswalk or not – we need to react and make real-time decisions.
    • You notice a car is drifting into your lane you need to react in real-time, make a decision, and often take action.
    • You realize there are lights flashing behind you. After the initial panic that we all feel … you react.

Nick - First Day of Driving with Dad (small) In case you haven’t figured it out --- our oldest son is learning how to drive. I was trying to explain the probabilities of pulling into the roadway and the likelihood of other drivers to do the unexpected. I’m not sure I achieved my goal, but I do know that he is a very conscientious driver and is on the lookout for deviations from the standards that he has been conditioned to expect.

What do you think?

  • How do you think about Standards and Statistics?
  • Do you realize you are “doing” Statistics every day?
  • Did you ever think that you’d ever use statistics again after college?

Thanks for indulging me in my attempt to explain Standard and Statistics.  If you have another point of view or a better way to think about Standards and Stats please share in the comments or drop me a line at one of my contact points below.

In the meantime have a Merry Christmas and let me know how your holidays stack up to your usual standards.

clip_image001Jeff Shuey is a veteran in the Enterprise Content Management industry. Over the past 20 years he has worked with customers and partners to design, develop and deploy solutions around the world. Jeff is currently the Director of Strategic Alliances at Winshuttle. He has worked for Microsoft, FileNet, K2, Captaris, Open Text, Kofax and Kodak. He speaks and blogs about ECM and the Intersection between Social, Mobile and Cloud Computing.

Follow me on Twitter, check my blog, send email or find me on Facebook, LinkedIn or Google+

1 comment:

Karl Schulmeisters said...

our brains fundamentally work "statistically" Its why Escher drawings work: the brain's image recognition statistics "match" different things depending on what you focus on. And that applies to decision making as well.

I'd add to your essay that I think perhaps you are falling into the trap of assuming a "Gaussian distribution". Those of us who went to Universities that did "curved grading" and took technical subjects, are familiar with the fact that the test results usually were NOT "Gaussian". instead they looked like a "barbell"

And if you apply Gaussian "standard deviations" to a "barbell" distribution, you will actually end up missing MOST of the sample.

It is important to understand that real world probability distributions can be widely and wildly varied