So now we know how to map values from one matrix to another, we might want to see mapping one column could effect mapping another column. The problem is that sometimes we don’t know how the columns interact with each other though. In other words, we don’t know their joint distribution. We might have a hunch about the joint distribution if we look at the pairwise correlation between two variables. So, say Randy wants to get the 411 on which dimension he is in based on where Tina is employed — if she is employed in that realm. So say he looks at the Houston and Indianapolis realms and sees that she’s employed at a computer store or a printer store. And since she works at those stores in justified dimensions, he can further deduce that it is a good dimension and Tina is in fact employed by good Anton. Or is he … meaning Anton being good or … well, you know how to find out by now! And yeah, Tina might have a more exciting life in the unjustified dimension … but does she have access to a nearby Chipotle there? Or a Barnes & Noble where she could order a copy of the Order of The Dimensions series? Or an AMC theater showing the Order of the Dimensions movie or … yeah, I’ll stop. But anyway, the point is working at a computer or printer store at a shopping center has its benefits. Like she’s most likely near a CVS or Walgreens where they may also carry DVDs of … well, you get the picture. But still, you can always check out the Order of The Dimensions series at Barnes & Nobel until next time!
And today we’ll talk about the ICC, or intraclass correlation coefficient, which basically tells us whether most of the variation takes place between subjects or within subjects. Like say we have different realms (no, I did not say dimensions — I said realms!) and we want to see if the differences between justified dimensions, I mean, realms or if the differences between them are greater. So if the five main characters within them are similar and, say, have similar lifestyles, the ICC will probably be higher. But — and now comes my version of the exciting part — if bad Anton got switched with good Anton in some of those dimen … realms, there could be more variation within them than between them so the ICC could drop. And how do we know if bad Anton switched himself with his good alter? Well, that could be tricky. Even trickier than finding the cat in this photo.
What? You did? That fast? Well, good for you! Until next time … when I’ll pull one over you yet. Can’t believe it took you that fast as it took me … never mind.
So I went to this Statistician of the Year Award Dinner the other night — same one I attended last year and thought about what message I got from it that I could convey here. And, although there were no cool Marilyn Monroe pix — and I thought the honey glazed salmon tasted better last year too, the topic was once again interesting in that it talked about how some data we analyze could be extrapolated too much. Like the famous ice cream-murder example. Never heard of the famous ice cream-murder example? No? Well, here it is! Or how in a recent study that the speaker talked about, women were found to wear red more often at certain times. Which made me very concerned, as I do not have much red in my wardrobe so what does that make me?
So of course, the next step was trying to relate them to my trilogy. So now, let’s say Randy again wants to figure out what dimension Anton is in and goes by the average climate of where his target could be. But even that is not as simple as it seems. Like if Randy determines that Anton is in a warmer climate, he could be in Hawaii (an unjustified dimension) or in Houston (a justified dimension). Or considering cooler climates, Anton could be in Finland (an unjustified dimension) or in Detroit (a justified dimension). So I guess my take-home message today is even though correlations are cool, their associations should be interpreted cautiously and maybe adjusted for different factors. Like in our case, Anton’s marital status, occupation, and his proximity to the masterboard. Masterboard which kind of looks like Hal …
Okay, not really. Now, you’ll have to read my trilogy to see what it really looks like, won’t you? Or else, you can join me next time when I think of another way to use stats to pimp out my books.
So I’ve decided to talk today about different correlations. You know, basically how well things agree with each other. So for example, there’s the correlation between two continuous variables, or the correlation between a continuous and binary variable, or variable taking only two values, or a correlation between two binary variables. Now the second type of correlation is called a point-biserial correlation and could help us find the association between what realm we’re in and Jane’s physics grade, which would be higher in the justified dimensions. Or Tina’s paycheck and the dimension we’re in, which might be a little lower in the justified dimensions. So why the heck would she wan t to be in a justified dimension? Well, again, you can just check out my trilogy here and find out. Finally, the third type of correlation, called the phi correlation, is between two binary variables, and for example, can be used to see if Tina is married in a certain dimension or not. So in which dimensions is she married? Again, my trilogy! Hello! Not that there’s anything wrong with being single by the way. Just ask Beyoncé.
I mean, yeah, she’s actually married and has a kid and … um, okay I’ll think of another example next time. Until then, “If you liked it then you should’ve put a ring on it … ” Yes, I’m stopping now.