 Chapter 11 in our textbook is correlation and regression. And the what of what we're going to learn in this is about positive and negative correlations that's associations between two quantitative variables. Think about high school GPA and college GPA. But we'll also learn about regression or using one variable to predict levels on another variable. We're going to do a simple version of this sold by variant where you have just two variables at a time. Although regression is a very flexible procedure and you can build equations that have dozens of variables in them to get a more precise prediction of your outcome. Now, in terms of why we're learning this, well, we're learning it because in certain fields, like for instance, economics, regression is a very, very common method. Or any time that you're doing a predictive analysis where you're trying to actually say what you think somebody score on some quantitative outcomes going to be, regression can be the tool of choice. And it's based on correlations. Every time you build a regression model, you're taking a correlation at this point, doing something with it, then taking a modified version of correlation and running through with it. It's a really useful general purpose procedure. And if you're ever going to work with data, you're probably going to do some version of this implicitly anyhow. And that gets us to the what for. What can you do with this? What are you learning it for? Well, again, anytime you're thinking about the association between two variables, think, for instance, how much time you spend on social media and how happy you are. You are thinking of a correlation. You might say they have nothing to do with each other. That'd be a correlation of zero. Or you can say, oh my goodness, for every hour I spend on social media, I feel this much worse or more irritable. That'd be a correlation of negative one, which is a perfect predictive association. Or if you're trying to decide whether how much money you're going to make by working more hours. Now, if you have a hourly job, then hours translate to pay really well. I work this many hours. I multiply times this. I know exactly what I'm going to get. If you have a salaried job, however, you are hoping you're banking on more hours making you more productive in some way, maybe getting a promotion, having a more nebulous and maybe a long-term effect. You can set that up with a regression equation. And you might say, well, the correlation's not really big, but I still can predict, for instance, a long-term outcome, what my salary is going to be next year in five years. Anytime you're looking at the association between these two things or you're thinking about how many different things can contribute, when are you happiest? When do you feel most connected to people? You are doing an intuitive version of correlation when you're looking at just two variables, the association, or you're looking at regression. When you're trying to think, I know that this will make me happier, maybe this much happier. And so these are great tools to have and the general principles you will find apply in about a million situations.