I’ve had it with average and why you should too

Who’s the biggest out of these two guys?

image (1)

Is it the guy on the left?
He’s barrel chested, bulky looking guy.
Could be. But he’s kind of short.

What about the guy on the right?
He’s biggest cos he’s taller? More an all over a lengthy kind of dude.

How about, instead of guessing, we check how their dimensions compare with the average?

The below pic shows how the individual dimensions of each man compares to the average for a man,  from height to leg length.



This pic shows (highlighted in yellow) how each part of the guy on the right compares with the average, whether it is below average or above it.  We can see there are four dimensions below average, weight, shoulders. chest and hips.
And he is above average on 4 dimensions too, height, reach, torso and legs. He is just on average on waist.


The guy on the left similarly is below average on 3 dimensions, and above average on 4. He is average on 2 dimensions, reach and chest.


So it looks like these two people are both NOT average, in most of their dimensions.
they are more UNLIKE the average than they are similar to the average.
Look at the pic below, between the two of them there are only 3 dimensions that are average.


So, these men are total freaks of nature, because they are un-average and not typical of the whole of the human race. Right?


It turns out that being untypical and not being average is actually the vast majority of people. Most people do not fit into an average on most of their dimensions.
Average is highly unlikely indeed.
So unlikely, that in one story an average sized man did not exist at all…

In the early 1950s, the U.S. air force measured more than 4,000 pilots on 140 dimensions of size, in order to tailor cockpit design to the "average" pilot. But it turned out the average airman didn't exist.In the 1950s USAF planes were much faster than they had ever been before. They were jets!
They moved faster and required faster pilot reaction times.

There were increasing amounts of accidents where pilot error was diagnosed.
They looked at the cockpits and decided that they needed to be ergonomically designed around the pilot. In the existing cockpits they couldn’t see the instruments without straining or twisting their head and reaching for a switch or pressing a pedal wasn’t quick or easy.
So the USAF decided to standardise the cockpit around the average pilot. They measured 4,000 pilots on 140 dimensions including such tiny things as thumb length and distance from eye to ear.
They gathered all this data so as to be able to design the perfect cockpit, based on the average of 4,000 men, in 140 dimensions.
This final set of data of average dimensions should have represented the best fit for the majority of pilots. The standardised cockpit would be designed around the average pilot’s size and would therefore be the best design for the USAF. What could go wrong?

A scientist who worked for the Air Force was puzzled and thought just how many pilots are average?
He chose the 10 dimensions most important for cockpit design, height, arm reach etc He compared these 10 individual pilot’s dimensions against the average to find out how many pilots were average.
He gave a generous 30% leeway to sit around the average of each dimension so even though the exact average height was 5’9” he defined the height of the “average pilot” as ranging from 5’7”  to 5’11”.
Most of his colleagues thought that the vast majority of pilots would fall under the average dimensions.

Turns out, that out of the 4,000 pilots measured, not one single pilot was average.
Yes, the pilots that produced the data for the “average pilot” were not themselves average.
Even when he only looked at 3 dimensions, just 3.5% of all the pilots fell within 30% of the average on all 3 dimensions. So even on a mere handful of dimensions, there were a tiny amount of average pilots.

So what? We’re not gents outfitters. What’s it matter?

It shows that we use the idea of “the average” to be a short-hand substitute for “most people”. So much so that when we say or hear “on average” we think “most” or “typically”, “normally” or even with a tiny step from there to “preferably” and we would be incorrect to do so.
The two men and the USAF pilot measurements above show this is not the case.
These two stories come from an amazing book called “The End of Average” by Todd Rose.

The book uncovers the assumption of an average that lies hidden underneath all sorts of domains. Ones where the word “average” is never actually said out loud, so you wouldn’t know that it is being used because it is often buried in other bigger assumptions that comes from “the average”.
For example in tests or other form of appraisals, at school or at work, people are graded and then ranked according to where they are against the average. If none of the 4,000 pilots were average, why would anybody be? And if average doesn’t exist then why would “above average”?
This book The End Of Average contains way more than I can do justice to here. If you’re at all interested in understanding how people and organisations work, then buy it.
And as a shortcut, ditch any measure tainted by average.

This entry was posted in command and control, data, knowledge, statistics, systems thinking and tagged , , , . Bookmark the permalink.

7 Responses to I’ve had it with average and why you should too

  1. micheleplayfair says:

    Reminds me of Lake Wobegon, “where all the women are strong, all the men are good-looking, and all the children are above average.” I’ve always maintained that “One Size Fits All” really means One Size Fits Nobody.


  2. Frank Wood says:

    I’m not a mathematician but this argument seems oversimplified. Using averages can be a very useful tool but only as a tool amongst other tools.

    There a term in Statistics called Standard Deviation and this (in the case of the pilots) would be able to show us just how many pilots were close to the average and can give a good idea on how to ergonomically design the cockpit.

    It’s only when people don’t use the term “average” properly that problems occur.

    Here’s someone explaining Standard Deviation better than myself



    • ThinkPurpose says:

      love the phrase “I’m no [insert profession here] BUT” so kudos for that.

      I’ve been mis-accused in the past of having a beef with IT, cos I always question why IT as a solution when it is suggested as a default answer to a hazily defined and poorly understood problem.
      I’m not anti-IT, I’m anti default thinking using misapplied concepts. Same as averages. I know you need an average to make a control chart, so I’ll use it.
      Im using IT right now to post this comment.
      This post was an attempt to tempt the reader to find out more about the idea of average, and hopefully the actual book.
      I missed the punchline from the story about the pilots and the cockpit. In the end they didn’t attempt to design a standardised cockpit, they built one that would fit virtually all pilots by demanding that manufacturers build ADJUSTABLE cockpits that would fit pilots that fell in the 5% to 95% range in each of a set of dimensions.
      Adjustable seats, straps on helmets, the sorts of things we see in cars today. So instead of forcing the individual into the system, they designed the system to fit around the individual.
      So, actually, typing that last sentence has just made me realise I missed out the point of this blog post, cos that’s the point of the actual book.
      It’s VG, I couldn’t do it justice here, so just go and buy it instead.


      • Frank Wood says:

        I’ve been accused of being a Luddite so it was a strange feeling for defending the use of averages in these contexts.

        Of course needless to say they used the Standard Deviation tool when making the adjustable seats etc albeit allowing for a large deviation but deviation nonetheless and where did they get the deviation from? Averages of course 🙂

        Did you read Mark’s critique on Amazon.com? A brilliant piece on the dangers of jumping to conclusions and how crappy life was before averages were used. Food for thought.


        • ThinkPurpose says:

          I have read it just now, after scrolling past about 30 other glowing 4 and 5 star reviews, I finally got to the 1 star review you mention.
          Honestly I don’t recognise the book at all from that review. Nothing he says I recall from the book. The bullet points at the end of his review, starting with sharing stating the book says that skills are a myth and people are disposable is the TOTAL opposite of an entire chapter in there about organisations should treat individuals like individuals, for the sake of staff and customers and the organisation. It’s total balls, beginning to end.


  3. Charles Beauregard says:

    Here’s a picture I’ve used a few times (either this version or a stick person version drawn on a flip-chart) to demonstrate one of the flaws of averages: http://tenacioustortoise.com/wp-content/uploads/2009/09/flaw_of_averages.png


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s