Lies, damn lies and statistics
Jack Balshaw 7/10/01

People are forever using numbers to convince us that we should support or oppose something.  At other times the purpose is to convince us something is or isn’t fair.  Numbers are thrown about in ways that can either clarify or confuse our understanding of an issue.  The problem arising from use of numbers is that, if you don’t know just what they represent or you assume they mean one thing and not another, you can be seriously misled. 

A favorite of mine is from those who encourage you to save simplistically for retirement by putting away a fixed amount monthly at an estimated interest rate for a long number of years.  The amount of money calculated as available for your retirement is impressive.  Unfortunately, they don’t usually deduct what you would have to pay in taxes, nor do they correct for the reduced purchasing power due to the effect of inflation on the value of money during all those years. 

We’re often told about countries where the annual per capita income is only a few hundreds or thousands of dollars.   The impression is that people in that country live poorly.  But, what we’re not told is that housing and food may be extremely inexpensive by our standards.  The net result might be that these people can live just as well as us ( maybe without the jet skies and SUV’s) and live lives of contentment.   OK, so some don’t.  

 Another example is when businesses want to convince us that they’re hardly making a profit. Typical is the information from grocery chains.  Their profit is said to be only about one percent of sales. This doesn’t seem like much when any of us can get better than four percent on a Certificate of Deposit but actually it’s almost free money.

 The money paid for groceries during the week pays the workers’ salaries at the end of the week and, because most of the products sold don’t have to be paid for until several weeks after receipt, other cost have already been collected via the customers’ purchases before the bill comes due.   Profit on actual cash invested could easily be in the 15% to 20% range.

 You might think that having a automotive fleet average of 25 miles per gallon is pretty good.  That’s until it comes out that trucks aren’t included in the fleet and all pickups, vans and SUVs are counted as trucks and not passenger cars.  All of a sudden, 25MPG isn’t such a big deal. 

 The common reaction to increased interest rates is that it’s a bad thing.  But how about all the seniors living off the interest on their life savings?  To them it’s a good thing, the interest is part of the money they counted on to meet their needs.  

Percentages are often used instead of actual numbers, and vice/versa, depending on the impact intended by those supplying the numbers.  A neighborhood might accept a 10% increase in traffic but would balk if told that meant 1000 cars a day.   On the other hand, two murders in town might be accepted more readily than a 100% increase.  A 10% increase in the number of skydivers killed in a year might mean just one additional death, while a 1% increase in the number of cancer deaths might mean a thousand additional deaths.  

 In general, when the numbers look bad, those responsible use percentages.  Of course their critics usually take the opposite view and use the actual numbers.

The changes in the number of millionaires, auto deaths, people on welfare, toxins in the air, failing students, etc., are all reported in such a manner as to shed the best light on the organization reporting the changes. A ten second delay at an intersection might not seem like much, but if it was only five seconds last year, that’s a 100% increase.

It may not be important to truly understand the numbers or percentages themselves.  But understanding the significance of how numbers are presented could cause you to have more or less confidence in other information an organization might ask you to believe.