A lot of people have math phobia, and if you do, statistics can be even more frightening. But you want to understand what you read about medical studies, right? Then you're going to need at least a basic understanding of what statistics means even if you don't understand the details.

Fact #1: You don't need to understand a lot about statistics to know what they mean.

Let's start with flipping a coin. It could land heads-up or tails-up. We've been taught from childhood that flipping a coin is random, that is, you can't predict which side up it is going to land. We sometimes use coin flipping in everyday life to start sports games or to break deadlocks. ("Who pays for dinner this time? I don't remember who paid for it last time. Let's flip a coin to decide.") But think about it for a second, is it truly random? Grab a coin, any one. Look closely at the head side and the tail side. Both have an embossed pattern of metal. Can you say for certain by looking at it that the embossing isn't larger on one side than another? I sure can't. That would make one side heavier than the other, and the heavier side would have a tendency to fall downwards slightly more often than the lighter side, wouldn't it? The only way to find out if that is the case is to flip the coin a few times and keep a record of the number of heads and tails. If coin tossing is truly random, you'd expect in the long run to get the same number of heads as tails. If you don't have a real coin or you are too lazy to perform this experiment, click on the "Flip Coin" button below to simulate a coin flip. The box labeled "Result" will tell you whether you get a head or a tail, the box marked "Heads" will keep track of how many heads you get, the box marked "Tails" will keep track of how many tails you get, and the box marked "All" will keep a running log with "H" for a head and "T" for a tail. Mash on that "Flip Coin" button a few times and you'll see how it works.

Result Heads Tails All

As you can see if you mash that "Coin Flip" button enough times, you won't always get the same number of heads as tails. But you should expect that if you mash it enough, the number of heads and the number of tails should get closer. You're going to get tired of mashing that button, I can tell. If you mash the "Run Experiment 1" button below, it will perform 28 coin flips for you. (I'll explain how I chose the number 28 on the next page.) The box marked "Flips" will record the number of coin flips, the box marked "Heads" will record the number of heads, the box marked "Tails" will record the number of tails, and the box marked "Difference" will comment on how many more heads than tails you got. Go ahead and mash that "Run Experiment 1" button now. Notice that the percentage in the "Difference" box is wildly different each time you mash "Run Experiment 1", and it can be quite large. Go ahead and mash that button several times.

Flips Heads Tails Difference

You'll notice that with 28 coin flips you don't always get 14 heads and 14 tails. There's nothing to worry about, that's just life. Or I should say "That's Statistics". It's just the way it is. Sometimes life gives you more heads than tails, and sometimes it doesn't.

If you mash on the "Run Experiment 2" button below, it will perform a hundred thousand coin flips for you. Give it a couple of seconds if you've got a slow computer, because that's a lot of coin flips. Notice that the percentage in the "Difference" box is now quite small every time you mash "Run Experiment 2". That's statistics. The more data you get, the closer you're going to get to "reality". Mash the button several times to make sure I'm not just messing with your mind.

Flips Heads Tails Difference
Fact #2: In statistics size matters because experiments with a small amount of data can give wild results.