 Hi, this is Dr. Don. I have a problem out of Chapter 9 on regression, simple linear regression. And in this problem, we're given some data which I'm not showing here, and it's the links and girths of harbor seals. How big around the middle are they? We need to find the regression equation, reconstruct a scatter plot with a regression line, then use the regression equation to predict the value of y for each of the given values of x, if meaningful. If the x value is not meaningful, predict the value of y, explain not. Well, remember we can only predict safely for x values that are within the range of our data. If we're outside the range of our data, then we're on shaky ground, so we're not being meaningful. We're going to do this using StatCrunch. So, I'm going to open up StatCrunch now. Okay, I have StatCrunch open. Remember you can click on the little blue rectangle if you're in my StatLab and have the data open directly or you can copy and paste it in. As I like to do is I can control where it goes. Anyway, we have our x values listed there, the length, the y value, the girth, and we're just going to go, first of all, because we want to check and make sure that we're not predicting out of our data range, we want to find the maximum values. Now, this is not a large amount of data, but if you had a large amount of data, you couldn't just inspect it very easily. So, what I like to do is go to Summary stats, rows, select my column, move this again, and I want to scroll down here until I get the min and the max, and click Compute. And I'm going to bring this answer over here, and you can see the minimum x we have is 123, the max x we have is 168. So, if our x value that we're predicting for is less than 123, it's not reliable. If it's more than 168, it's not reliable. And that's good to know. We're going to get the regression by going to stat, regression, simple linear. And it brings up this dialog box, and we want to select the x variable, which they tell us nicely, x, and the y variable. We're going to perform the hypothesis test, and we don't have to change anything here. But down in this section, we want to put the values of x that we're going to predict for. And you click in there, and you start typing 140, 172, 164, 158. And the level that we're going to predict for, the 95%, we don't have to worry about, we can just leave that default, and we're going to click compute. And we get our answers up here. And we get the regression equation there, y equal 9.758 or 9.759, which is the intercept. And the slope is 0.695 times x, the length. We've got our correlation coefficient, r of 0.919, which is pretty highly correlated. And then we can look down here on the bottom, and we can get our predicted values of y for each value of x. Now 140 is okay, 140, the predicted value y is 107.09, or 107.10. The 172 is outside our range, that's more than our max, therefore that's unreliable. 164 is inside the range, 123.78, and 158 is in the range, so that's okay too, 119.61. To get our graph, we just click the little arrow there, and here's our graph, and then we just need to compare this to the graph that we have on our options and pick the right one by looking where the intercept is, looking to see if it's a positive from lower left to upper right slope, and then looking where the different data points. So you just select the right answer that way. So that's a quick way to do this using StatCrunch. In the next video, I'll show you how to do the same problem using Basic Excel. Hope this helps.