Numeric Data

Next we’re going to talk about different types of numeric data. Let me ask you a question. Is $4 twice as much money as $2? Think about that for a moment. Yes, the answer is yes. I mean, clearly, 4 is twice as much as 2. But let me ask you a different question. Is 40 degrees Celsius twice as much heat as 20 degrees Celsius? Now, you might be tempted to say yes to that. But the actual answer is no. It’s not twice as much heat. The number 40 is twice as much as 20. But if you were to convert that into Fahrenheit or any other unit of temperature, you would find that it is not twice. So what’s going on?

There’s numbers, but these two situations seem to be different. The answer is that not all numeric data is created equal. In this case, for money, there is what we call an absolute zero. That is, zero is the smallest value you can have. When you hit zero, you are out of that thing. With temperature, the zero is arbitrary. We just picked some point along the scale and called it zero. These are very different types of numeric data. And you’re going to come across both of them in the data you analyse. So when we have an arbitrary zero, we call it an interval scale. It is something like temperature.

Or it might be any numeric rating system, like if you rate something on a 1 to 5 scale. The zero there doesn’t have any meaning at all. If there’s an absolute zero, typically when we’re counting things or have some quantity, we call it a ratio scale. Only when we have an absolute zero could you make comparisons like this is twice as much as that when you have those numbers– two very important distinctions for two very important different types of numeric data.