turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- SAS Procedures
- /
- Correlation data interpretation

Topic Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

10-25-2017 09:29 PM - last edited on 10-25-2017 11:13 PM by Reeza

Hi, I am new to SAS so my question is just very basic.

I am trying to perform a correlation test using the code below:

proc corr data=meat;

run;

I got the following results:

I see two numbers in each cell. What does each represent?

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Sara73737

10-25-2017 09:30 PM

Hi, I am new to SAS so my question is just very basic.

I am trying to perform a correlation test.

I have tried the following code to estimate the relationship among the variables.

proc corr data=meat;

run;

I got the following results:

I see two numbers in each cell. What does each represent?

Thank you

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Sara73737

10-25-2017 11:13 PM

2 identical posts merged.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Sara73737

10-25-2017 11:10 PM

Look at the table heading.

First value is the Pearson Correlation Coefficient.

Second value is the P-Value, Prob > |r| ...., is the correlation significant.

UCLA has a section with annotated output that can be helpful.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Sara73737

10-26-2017 11:08 AM

And if you don't want the p-values, you can use the NOPROB options, like this:

```
proc corr data=meat noprob nomiss;
run;
```

Here is a link to the PROC CORR documentation, in case you want to read more details.