Be Skeptical Of Statistics

Statistics can be true and still mislead you.  It is not the fault of the statistic, it is the fault of the way we process information and of the way that information is presented to us.

Here is an example.  Subaru is now copying a late ’70s Volvo ad.  Subaru says, “90% of the Subarus sold in the past 10 years are still on the road.”  What does that tell you?

If you think what the advertising agency would like, it tells you that Subarus are reliable and durable.  The advertiser is using truth to manipulate your thinking.  Still manipulative though.

What is missing?

As I have pointed out many times before, you cannot assess a summarized statistic unless you know the population of all events.  In this case you need to know that, in the United States, of all the Subarus sold in the past 10 years, 65% have been sold in the past 5 years.  We immediately see that the population data is skewed towards the recent and durability is intuitively a long time idea.  We know very little about how many of those sold between 6 and 10 years ago are still on the road.  That might tell us more about their value in use over time.

Here is a fun statistic.  Of all Tesla “S” vehicles sold and delivered in the past 20 years, more than 98% are still on the road.  Useful?  Not really.  All of the Tesla “S”  vehicles sold, have been sold after 2011 and 92% of those after 2012.

If you are unaware of the population of all events, you do not know the meaning of the statistic.

Averages work that way, too.   Take the total of the characteristic, divide by the number of instances and there you are.  The average Canadian has 1.97 legs.  The fact that not a single member of the population has that configuration should tip you off that there is something amiss.

Pay attention to things like the yield on your portfolio, but in context of the entire dataset.  Making 15% this year says nothing about next year.  It does not even help you guess last year.  It is just a number until you fit it into a bigger scheme.

Statistics can be both true and without meaning.  Be cautious.


Don Shaughnessy is a retired partner in an international public accounting firm and is presently with The Protectors Group, a large personal insurance, employee benefits and investment agency in Peterborough Ontario.

Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: