"Branson A. Boots, how many times do I have to tell you?!? Do "Meow?" "And now you're sick again. You went on another bender and once again, yecch, it's going to come out of you one way or another. I see some of it already has. And guess who is going to have to clean up rejected pieces of rubber again?" "Meow?" "See, look at this chart I made. On the x-axis is the number of times you've eaten some of the fatigue mats. On the y-axis is the number of times you've gotten sick. Notice how it's a completely straight line at a forty-five degree angle? That's because every time you eat part of the mat, you get sick! One eating, one getting sick. Two eatings, two getting sick. And so on. It's a perfect correlation!" "Meow?" "And as I have been trying to tell you, R-Squared is a statistical measure of fit that indicates how much variation of a dependent variable is explained by the independent variable or variables in a regression model. The R-Squared in this case is exactly one. Now, that almost never happens in real life! If you have an R-Squared of one, you usually either don't have enough data points or there is something wrong with your measurement process. The chances that there is nothing else influencing a dependent variable is pretty close to zero in a typical analysis. If we knew that it was that close of a relationship we wouldn't need to study it in the first place. Right?" "Meow?" "Right! And there's only one independent variable here—and that's you eating the fatigue mat! I see the evidence of it downstairs! There are little pieces of it all over the floor, and do I need to mention there are also... processed... pieces on the floor of the kitchen?" "Meow?" "So do you see now that if you don't want to get sick and feel lousy for two or three days, you have to stop eating the fatigue mats? Your companion Shadow never eats them and she never gets sick that way. OK, yes, she is very good at manufacturing hairballs, but that's different." "Meow?" "Yes. I'm really glad we had this conversation." ... |