statistics - blog - Sister to Sister

 \t \t \t \t   

March 3, 2019, Numbers may not lie but statistics can be manipulated

I love numbers. When I was in my early teens, I loved to do math problems using eight and twelve instead of ten as a number base. I did not believe ten was some kind of universal law of numbers. After all, if we had eight or twelve digits, we would be using eight or twelve as a number base.

When I was a statistical stenographer for several years, I discovered while numbers are true, they can be manipulated and statistics could be a slippery slope. I did not manipulate the numbers, but I did type them and saw how they were used.

Suppose this is a chart of national income for January (dark blue) through June (light blue). After the frantic buying of December and Christmas, January was a bit slow. But it picked up. If you you want to show income levels dropping, compare April (green) to June (light blue) and you have proven your point. But if you want to show income levels have risen, compare January (dark blue) to June (light blue) and you have proven your point. And both will work as long as no one sees the whole chart. This is a simple, straight forward manipulation of statistics. There are so many ways to make statistics say what you want them to say.

There are several books written on how to lie with statistics and several websites with the same information. My daughter majored in finance and minored in economics and she told me her statistics professor introduced the class to the subject. Two such books on the subject are Standard deviations: flawed assumptions, tortured data and other ways to lie with statistics by Gary Smith and How to Lie with statistics by Darrell Huff. And there are a number of websites that will teach you the skill of lying with statistics.

When some one tries to use statistics to influence you, ignore them. Then, if you are really interested, location all of the original figures and do your own statistics.

Numbers don't lie, but statistics are free wheeling.

Let me tell you a joke I heard from Elizabeth Warren. A man was sitting in a bar when he noticed Bill Gates walk in the door. The man announced to all, “On average, this bar is full of millionaires.”

That is how averaging works. If you combine everyone's income, even the lady that works at Walmart for five hours a week, and get an average, it looks really good. But it is because the billionaires are getting so much richer. Even though the middle and lower income families are getting poorer, the averaged income is going up.