Tuesday, January 13, 2009

Law of Large Numbers is Alive and Well

I originally wrote this for the newsletter of a local statistics professional group in late August, 2002. Writing "Time to Play" reminded me of this, and after finally discovering the archived-archive where it was hiding, I have dusted it off to present as a bit of recycled writing. The only trouble was I couldn't decide which of my blogs to put it on; Science and Math, or Math and Games? Since I'm facing a bit of a time crunch this week anyway, I'll post it to both. Problem solved! (but I'll try not to make a habit of it.)

Law of Large Numbers is Alive and Well
at 2002 GENCON Games Fair

“All these people came here to play games?” said one astonished worker to another, as they walked through the midst of several thousand people at the 35th GENCON Games Fair, all busily rolling dice or shuffling cards for a dazzling variety of games. I almost trailed them to see if I could overhear more comments, but I needed to get back to my own session. However, I did continue to think about this pair, and about why people like games.

My own exploration of probability began with the boardgame Risk when I was young, and soon I was writing my own simulation programs on an Apple II+. I was fascinated by the thought of enumerating all the possible outcomes and determining the best strategy for winning the game. This self-study soon expanded to include most of my favorite games, got me hooked on computers, and eventually led me to the study of statistics and a career. I still like to study my favorite games; I love to play them, disassemble the rules, study the probabilities, put them back together again, then go play some more.

Looking over the GENCON crowds gathered in the Midwest Center, I have to believe that there are lots of people out there who are quite interested in probability, and whether they know it or not, interested in statistics too. Some of the people I spoke to are very knowledgeable of the probabilities involved in a game. Furthermore, THEY THINK IT’S FUN.

Now, when was the last time your heard someone describe their college stats class as fun? How many times have you seen people get the “your-a-WHAT” look on someone’s face when you tell them of your profession? I have personally met at least half a dozen people still in statistical-shell-shock, twenty or more years after taking an intro level stats course. How many people have you met that would consider statistics to be fun?

I am very happy being a statistician, but I sometimes think that as a profession, we haven’t done a very good job of selling ourselves. Too many people have the impression that statistics has to be dull, boring, or cryptically complex. Too many people think that statistics are something they can’t understand. It doesn’t have to be that way. I think statisticians need to try harder to increase awareness of what we do, what it’s good for, and how it can be of benefit. I think it’s very important to reach out and find ways to attract new people to our profession, and to let them know that it offers excellent opportunities.

There are plenty of people who are interested. There are plenty of people who grasp the basic concepts. I know there are, because I just saw thousands of them at GENCON. All of them were shuffling cards or rolling dice, all of them putting the Law of Large Numbers to the test, and all of them loving it.
Update - Six Years Later: The need for statistics education has not gone away. There is more and more information being presented to us every day, and our ability to process all that information is limited. A working knowledge of math and statistics is a crucial skill nowadays. I don't mean that every person needs to understand high level probability theory (heck - I don't understand all that stuff - at least not very well), but I do mean that every person needs to be educated as a consumer of information. They need to know what the numbers mean, and when they don't mean anything at all. They should have a basic understanding of how numbers are presented (descriptive stats), what they represent (estimation), and how decisions are made on this basis (inference).
I should add for the benefit of students interested in math and statistics at more than a basic level; there are a lot of jobs available to you even as an undergraduate; there are a graduate statistics programs desperate for good students and willing to pay you to go to school; there is a rich field of careers in math and stats to look forward to, and finally - if you work it right - it might be a lot of fun.