The Bell Curve Defined
How tall are you compared to the average person? How much does a dog weigh compared to others of its breed? What is your IQ compared to your friends? All these questions can be answered with the help of the handy-dandy bell curve. This lesson will introduce you to the bell curve and show you how, with the right information, you can figure out just how smart you are compared to the rest of the world.
Some things you look at and wonder how they came up with the name for it. Have no fear, the bell curve is not one of those things. Just look at it - it's a bell. While the nomenclature is simple, the idea of what makes a bell curve and how to use it are a little trickier.
The bell curve, sometimes referred to as a normal curve, is a way of graphically displaying normally distributed data. What is normally distributed data? It's a data set with an identical mean, median, and mode. Put more simply, normally distributed data has a perfect average. Look at a bell curve and you'll see it is perfectly symmetrical. There is exactly 50% of the data on either side of the midpoint. You'll also notice that the largest amount of is at the center of the curve and it tapers off as it gets higher and lower, giving it the distinctive bell shape.
A typical bell curve
 |
The beauty of the bell curve, and why it is so important to understand, is that the information it displays has certain characteristics. As mentioned above, the average of the data is perfectly in the middle, but you can also see there are percentages given for each 'chunk' of data. Each one of these data chunks represents one standard deviation, but don't worry about that term for now. Just think of them as chunks of data. In a bell curve of normally distributed data, those percentages given will always be true. The middle chunk will always add up to 68.2%, the next chunk is 27.2%, the next is 4.2%, and the last little bit is 0.2%. Note that because of rounding, they don't add up to 100%.
It is important to note here that this is true only for perfectly normal data. You may encounter graphs that look like a bell curve but the data isn't perfectly normal. In other words, the graph won't be perfectly symmetrical and it doesn't have identical mean, median, and mode. This is what we call an approximately normal data set. These are usually quite close to a bell curve, and sometimes you can get away with using these bell curve 'chunk' percentages, but know that they won't be super accurate unless the data is perfectly normal.
Now, on to proving that you have a higher IQ than half the world!
