Pet Tricks

For some people, numbers are like pets. They groom and bathe them. They feed them. They use them to fetch information. They even teach them how to do tricks!

For example, Moshe Adler writes,

Gov. Andrew Cuomo wants to freeze the salaries of state employees, but The New York Times soberly cites numbers showing, it says, that the governor “may need to go further.”

What are the numbers, then? “The average salary for New York’s full-time state employees in 2009 (even before the last round of raises) was $63,382, well above the state’s average personal income that year of $46,957,” the paper noted. There is nothing wrong with the numbers, but putting them together, side by side, suggests either that the editorial board of The New York Times does not know that you must compare apples to apples or that it does not know what the words per capita mean. (In either case, this is perhaps evidence that the private school system is also in need of reform.)

The first figure is the average salary of a state employee, and in a family in which both parents work full time and for the same pay, the combined income from salaries is $126,764. The second figure is the personal income per capita, which means that if personal income were distributed equally, every family of four in the state of New York would have had an income of $187,828. It is clear, therefore, that the incomes of the families of state employees do not come even close to the average family income and, as a matter of fact, neither does the pay of the vast majority of New York families. In 2009, median income for a family of four in the state was only $82,587. Both public and private workers are due a raise...

I realize that it could be argued that the numbers themselves are not the problem; it's how people manipulate them. I would counter that manipulation is an inherent part of the story of numbers themselves. Numbers are rather meaningless beyond their presentation to humans. So, how we present them often involves trickery.

Here's another example, if I was trying to lure tourists or businesses to this area and told them that precipitation in South Bend was 10 inches above average for the entire water year, the average person might think that figure wasn't all that exceptional. Ten additional inches spread over 12 months only amounts to an increase of about 0.83 per month. That total would be nothing to get excited about.

However, we have realized that 10" above normal for a water year in 2 weeks less than the halfway point of the entire year! Here we are in the middle of March and our rainfall total for the water year (10/1/10 - 9/30/11) stands at 93.5 inches and counting. In other words, if starting tomorrow and running for the next 6+ months we experienced nary a drop of rain, we would still end up 10" above normal for the year.

Needless to say, it will not be dry for the next 6 months. The forecast for the next 7 days calls for rain. At the rate we're going, we will end up 25" or more above average.

My point here is that numbers can be utilized in many different ways. They can be massaged to back up almost any point of view. So, it is important when viewing numbers to make sure we understand how they were derived AND that numbers shown side-by-side are comparable in the first place.