 In this video, we're going to look over some tips on how to study for the test on statistical inference. This is the homepage of the website, the course website. Go look at the left side under course materials, the handouts page, and then inside the handouts page, exam prep. Here we are. The first thing you see under prep for exams is the formulas and tables packet. That's what you're going to need to have during the test. If it's an in-class exam, then you'll be provided with it. It depends on your teacher. If it's an online exam, you should make sure to print it out and take it along with you. You've got formulas, you've got tables, everything you need in order to take the test. Back to handouts. Okay. Now, under study sheets, you have some materials that were taken from the lecture notes. You might call them summaries, some good summaries to look at, to review, make sure you know everything for the exams. Of course, if you've gone through the lectures carefully, you already know this and you probably don't need them. Let's look at the ones that might be relevant to the inference exam. Steps in hypothesis testing, of course. Inference overview. Let's take a look at that one. That's the kind of thing that helps you, not for all of inference. This doesn't cover confidence interval estimation, but for hypothesis testing, we've learned a lot of different types of hypothesis tests. They're all very, very similar, which is a good thing because once you learn one, you've learned all of them with minor differences. But you might need some kind of a guide to help you figure out when you read a problem, what kind of problem is it? And here's your guide. What's the parameter? We've learned mu and p, the population proportion. What test statistic will you be using, z or t? Is it a one-tail or a two-tail test? And is it a one-sample or a two-sample test? Back here. Hypothesis testing. When you work out a problem for homework, you want to make sure that nothing is missing. Here are all the different parts of a hypothesis test that must be there. And this is what I'm talking about. Here's an example. You need four parts. If you're missing any one of these parts, it's not a hypothesis test. You need the hypotheses, H0 and H1. Without these, it's not a hypothesis test. You need to define the parameters of the test. You want to know when you're going to reject the null hypothesis and when not. So you need to choose the test statistic. In this case, it's z. You need to find the critical values from the table. You need to set up the decision rule in the unshaded area in the middle here. That's where we say no, we can't reject the null hypothesis. It might be true. And in the shaded areas in the two sides, the two tails in this case, that's the region of rejection. And then the third part is what we get from the data, from the empirical data, the calculated value of the test statistic. And then finally, you need to come to a conclusion. You compare your calculated value of the test statistic with the critical values that you got in the second step. And you either reject the null hypothesis or you don't reject the null hypothesis. But in any case, you need to have that final step. It's not a hypothesis test without each of these four steps being present. Again, this is something from the lectures to understand the difference between substantiating a claim and refuting a claim that's different paradigms in hypothesis testing. Here's something relatively new. If you've been following the course for a while, you may not have seen it. It was just put up last week. It's a lecture that's a review session of certain problems in inference. And here you have it in narrated PowerPoint form and as a YouTube video, just like all the other lectures. And that would be something very, very good for you to follow up on. And this is a good point to make at this point and all throughout the semester. The best way to study for exams in this course is to do lots of problems, problems, problems, problems, homework, homework, homework. Find problems wherever you can. Go to problem solver books. That is the only way that you're going to... You can't just study theoretically and say to yourself, oh, I know that. And then when you come to the exam, hope that you're going to get 100. It's very easy to get 100. It's very easy to get an A in this course, but you do need lots and lots of practice problems. So this review lecture gives you practice problems in inference. And in addition, we have more practice problems right below it. You can ignore the ones for the midterm and sampling distributions because that's done. We're talking about the next test, the inference test. And here you have two practice sets for inference problems. Let's take a quick look at it. If you download this, you will... There we are. You'll see all different kinds of problems that you can work on. The more, the better. Here we are back again. This is an interesting one. Inference popery. One of the problems that a lot of students have when they do the lecture, they do the homework. It's, let's say, a one-tailed, one-sample t-test. And they get it all right and they're great. The problem happens when you're at a test. And the test could have problems from any of the inference topics that we've studied and there's at least four weeks of them. And you want to practice doing problems where you don't know in advance what chapter or what topic it came from. So let's take a look at that. Make sure you read the first page very carefully because this is actually taken from a review for the final exam where there might be more problems than you need. And so you read this carefully and you only do the problems related to statistical inference. What we were talking about, mu versus p, z versus t, and so on. But the nice thing about this practice is that the problems are all over the place. And where are the solutions? We'll go back to the handouts. The solutions are right here. And you get an idea of why it would be very difficult for me to redo this and take out the irrelevant problems. One day I'll probably do it, but I got to say that I'm not looking forward to it. So these are the solutions, the problems we had before, you saw them. What else we got? Anything that says practice for the final exam, of course, is going to include inference. I might have a review or pointers for studying correlation and regression separately, but this is supposed to be purely for the inference sections. Before I sign out, let's just go to the lectures, the overview of the lectures. So you have an idea of what we're talking about. When I say the inference exam, the inference exam would cover all these topics. There you go. Starting from introduction to statistical inference. And there you go. Through all the two sample tests. I'm not including chi-square because we don't always do it every semester. So if your teacher did do the chi-square distribution and you're responsible for it, then obviously you would want to include that. All right. I hope this was helpful. Good luck to everybody on your exams and on your progress through your professional careers.