 Okay, now we're going to conduct an experiment that has nothing to do with Santa. And we're going to use this little experiment that I came up with that has nothing to do with Santa to help us define or explain what are these variables that we must identify if we're going to have a solid experimental design. So first of all, I'm going to give you my little scenario. Let me tell you what my question is. My question is, does Santa exist? Just kidding. My question is, do high protein diets cause weight loss? Okay? That's a question. Is it a hypothesis? Well, think about it. A hypothesis is a proposed explanation for something. Is that an explanation? Dogs, that's a question. That doesn't count as a hypothesis. So let's make that question into a hypothesis. My hypothesis is going to be, yes, that high protein diets cause weight loss, period. And that I'm explaining a phenomenon of weight loss that happens upon a high protein diet. Now, if we're actually going to do this experiment, we need to identify an independent variable, a dependent variable, and we need to identify a whole doodoo load of standardized variables. Standardized variables. All right, let's identify our doodoo loads. First of all, your independent variable is a variable that you change, and variable. What is a variable? It's a thing. Let me give you an example. It's the thing that you change or that you manipulate. And what is the thing that we're going to manipulate in a high protein diet? In our example, a protein diet. That says diet. So if we're wanting to test this question with a hypothesis that, yes, high protein diets do cause weight loss, then our independent variable is going to be the protein in the diet. So we're going to set up a couple of groups and we're going to have a high protein group and then a low protein group. That's our independent variable, the protein in the diet. It's the thing that we change. The dependent variable is dependent upon the changes and it's what you measure. So it's the thing that you measure. And in the case of our little scenario here, what are we going to measure? Body weight dogs? That's how we're going to know. Did the protein diet have an effect on body weight? So our dependent variable supposedly will change the different treatments of different amounts of protein in the diet. Now, that's pretty straightforward. Do you agree? Talk to me about the standardized variable. What are those things? What are those crazy things that you're talking about? Those are actually the reason why this experiment right here is going to be a disaster because good luck having the standardized variables is going to stay the same. Standardized variables are the things that stay the same. If you have two groups and one of them is getting the protein diet and the other isn't and you're measuring weight loss, you better have their gender the same, their age the same, their exercise level or activity level the same. Let's say their body temperature, how about their house temperature? How about where they live? What kind of job they have? What? Those are all standardized variables. The ideal experiment has no variables that aren't standardized. Your two groups of humans are identical to each other and experience identical conditions and have identical genotypes and genetics and parents, their clones, all of them in both groups. And the only thing that's different between them is your independent variable. What? You can see why doing studies on human beings becomes rather difficult. Doing studies on anything becomes rather difficult because, dude, there's so many standardized variables that can be mixed up and changed around. Once we've identified our variables, that's going to help us visualize groups that we're going to put these guys into. And that's what we're going to talk about next. And these two concepts are really closely linked. So let's see what kind of groups we're going to come up with in our experimental design.