Daily Independent (Ashland, KY)

Columns

May 11, 2014

Charles Romans: Statistically speaking: 05/12/14

ASHLAND — Two plus two will always equal four — that’s common knowledge, and basic math that we all learn as children. But is it really four?

Of course it is . . .  sort of. It all depends upon whether or not the individual parts of the first two are exactly equal to each other, and to the other two parts.

If all four single parts are equal, then the answer is four. It would seem so simple as to almost be childish; and it is until we pass beyond basic, known quantities. The problem is that things aren’t quite that simple.

Research is a complicated process that scientists (as well as corporate marketing departments and others) use to gather information; and hopefully that process will yield manageable results such as which products are beneficial to human health and, I suppose, what flavor of cola most people prefer.

All of the massive amounts of information that these scientists/companies collect is called data — information relating to the subject. That data is then turned into statistics such as frequency of occurrence, effects, and other results.

The problem with statistics is that statistics, though they contain facts, are not facts in and of themselves. Statistics at best are percentages. Statistically, there is full employment in the country when between 4 and 5 percent of the country is unemployed — but that is not the same as everyone having a job.

An unemployment rate of five percent in the United States means that over 12 million people aren’t working; but that statistic isn’t a fact either because it doesn’t allow for the working age citizens still in high school, those over 60 at retirement age, and many other factors.

If we were to say, for instance, that there were 1,000 foot-long hot dogs sold last week then we could also say that there were 12,000 inches of hot dog (with assorted toppings) sold.

That seems reasonable because one foot is twelve inches, and twelve inches multiplied by 1,000 equals 12,000 inches; but a ‘foot long’ hot dog, in reality, is only 10 inches long.

By assuming statistics are facts, we have lost 2,000 inches (roughly 167 feet) of hot dog.

That of course is the problem in a nutshell. For statistics to be of any real use to us we have to understand what variables there are in the calculations or we arrive at an inaccurate conclusion.

And when we consider that it is the conclusion that scientists (and salesmen) are looking for, then we had better be sure that the conclusion is as accurate as possible.

The uncertainty, the need to allow for variables that we might not even be aware of, is why there are more theories than laws in science.

Too many times we have statistics thrown at us that prove this works better than that, or this causes weight gain, or that makes our lawns greener, etc. Each day we drown in a sea of facts that really aren’t.

But, since mostly true at least seems to be partially false as well, we seldom get the actual figures because we might disagree with those who provided them.

CHARLES ROMANS is a freelance writer living in Greenup County.

1
Text Only
Columns