1.3 How do you tell a story with numbers? - Video Tutorials & Practice Problems

<v ->So when you're telling a story with numbers,</v> there are certain pitfalls that you might need to think about there's certain ways of describing the numbers, there's certain ways of telling your story that you really need to keep in mind. So let's look at this example here. I call this the flaw of averages. Because if you look at this graph, it's true that the average depth of the water is 14 inches, but depending on what you're trying to convey, that might not at all be the important thing to discuss. Means suppose what you're trying to convey is whether it's safe for you to walk across this body of water on foot. Well, I don't think it might be the best thing to do. Because you can see it, it's deepest, it's about eight feet. Even though the average depth of the water is 14 inches. And so knowing what you're trying to talk about, knowing what the listener to your story needs to know, knowing what decision you want to make based on the data, can help you figure out what you want to tell in that story. And it could be that the average is the important thing to tell. But in this case, if you're trying to make a safety decision for walking across this water, I think the deepest part of the water might've been the number to really communicate in your story. And so think about what the decision is that people are going to make based on your data as you construct the story that you're communicating with. Here's another example for average household income. So average household income seems like a pretty straightforward number. But with every average it's based on a sample. And so what happens if your sample is skewed? Suppose for example, just randomly, your sample for average household income includes Bill Gates. Well, now all of a sudden the graph looks a little bit different than it did before. And this is just because you have one outlier in your data. And so maybe looking at the data to make sure that you don't have some one data point or a couple of data points that are wildly outside the norm for all the other data that you have, that can really hurt your ability to tell a story. It can actually completely confuse the story. And so we don't want to really do that. We don't want to tell people that $3 million is average household income because Bill Gates was in it. And so use the data to tell the story. So don't tell averages when really the deepest part of the water is what you need to do. You need to select the metrics that best explain your data or that reduce uncertainty. And ask yourself, do averages tell the story better or ranges? Try to find visuals that really illustrate what you're trying to get across. And understand that narrowing the problem can often help clarify the data. So if you're using lots and lots of data, maybe try and narrow down what data you using to just the data that's important for a particular problem. If you have a problem that has three or four decisions that you need to make, maybe you need to tell a story for each decision, and actually narrow the data for each decision. So use your data to tell the story that helps you make decisions better.