1
00:00:00,000 --> 00:00:01,166


2
00:00:01,166 --> 00:00:06,966
Hello, and welcome to a video on correlation matrices in Python. Let's use the ADD data again.

3
00:00:06,966 --> 00:00:07,499


4
00:00:07,500 --> 00:00:13,566
So, as usual, we're going to be importing pandas as PD, and from Google Colab we'll import drive, so that

5
00:00:13,566 --> 00:00:17,732
we can mount our drive, set our file path, and read in our data through that file path.

6
00:00:17,733 --> 00:00:20,599


7
00:00:20,600 --> 00:00:26,866
Okay, so you can see, for the ADD data set, we have some new variables

8
00:00:26,866 --> 00:00:33,399
that we get to work with. We have GPA, social probability, dropout rates, IQ,

9
00:00:33,400 --> 00:00:33,433


10
00:00:33,433 --> 00:00:39,633
ADDSC, which is some score related to psychology, but I wouldn't

11
00:00:39,633 --> 00:00:45,899
know off the top of my head what it is. Let's say we'd like to know the correlations

12
00:00:45,900 --> 00:00:45,933


13
00:00:45,933 --> 00:00:52,033
between ADDSC, GPA, and IQ. The code to do this is fairly straightforward. So first, let's

14
00:00:52,033 --> 00:00:57,933
subset our data. We'll create a subset data frame, equal to DF,

15
00:00:57,933 --> 00:00:58,233


16
00:00:58,233 --> 00:01:04,399
and it's important to have these two braces here when you're subsetting. This lets Python know we would

17
00:01:04,400 --> 00:01:04,700


18
00:01:04,700 --> 00:01:11,100
like just these variables from this data set. We want ADDSC, GPA,

19
00:01:11,100 --> 00:01:11,133


20
00:01:11,133 --> 00:01:17,133
and IQ. Next, we'll make our correlation matrix. So, I've made a variable called

21
00:01:17,133 --> 00:01:22,366
correlation matrix. That is equal to subset, DF, dot, core.

22
00:01:22,366 --> 00:01:23,566


23
00:01:23,566 --> 00:01:29,699
This is something that's available through regular Python. So, we don't have to import any special libraries to do it. Next, we'll

24
00:01:29,700 --> 00:01:35,800
print our correlation matrix. And that's it. We can see we have a positive

25
00:01:35,800 --> 00:01:35,833


26
00:01:35,833 --> 00:01:41,333
correlation between GPA and IQ and negative correlations between all the the rest of them.

27
00:01:41,333 --> 00:01:43,233


28
00:01:43,233 --> 00:01:47,933
All right, that's all we have to say about correlation matrices. Happy coding!