From the course: Everyday Statistics, with Eddie Davila

The normal distribution is everywhere

- [Instructor] Perhaps when you were in school the instructor would put up the entire class' test results on a chart. Often it would look like a bell. In fact, they might have referred to it as the bell curve. Actually, you probably had a number of instructors that showed this to you. Maybe you even started to wonder if it was an academic conspiracy. Well, I'm here to tell you it's not a conspiracy. In fact, we don't just see bell-shaped distributions on test scores. They're everywhere. All around us. And while we might refer to this curve as the bell curve, statisticians refer to it as a normal distribution. And as I said, so many things in nature and society are normally distributed. If you measure the blood pressure of a large number of humans, the results will look like a bell curve. The temperature in London over the last 10 years. The sizes of watermelons grown in a certain state. The heights of NBA basketball players. Shoe sizes for women. The weights of newborn babies in Japan. When the number of observations is large enough, the data will resemble a normal curve. Now, not all normal curves are created equal. Some are tall and skinny. Some are wide and fat. And some are somewhere in between. And what's even more amazing, armed with the normal distribution and just a little bit of statistical know-how, you can estimate how common a single data point might be. Why? Well, when you have a normal distribution, about 68% of all data points will be within one standard deviation of the population average. 95% of data points will be within two standard deviations of the population average. And 99.7% of all data points will be within three standard deviations of the population average. So since you know that many things in nature, like blood pressure, are normally distributed, if you had the population's average blood pressure and you could calculate the standard deviation for the population, you would be able to see if someone's blood pressure was pretty normal, within one standard deviation of the average. Or maybe a huge outlier. That is, three or more standard deviations from the mean. As I said, the test score bell curve is not a conspiracy. And it's something you'll likely encounter in nearly any career. Take the time to get the basic statistics foundation that will help you discover the importance of the normal distribution.

Contents