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

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Chi square test?

- Subscribe to 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
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

01-13-2017 04:52 PM

Hello:

Assume we try to investigate whether there will be potential exposure to toxin A if people have high intake seafood diet. I plan to use **Chi square** test to compare the urine matabolite with the normal diet and high seafood intake diet in three months. There will be 30 participants in the normal diet group and 30 participants in the seafood intake group. And the urine metabolite is the continuous data.

Am I select the right statistical model? How to write a SAS code? Is it used Proc freq / Chisq? But how? Thanks.

Accepted Solutions

Solution

01-19-2017
11:16 AM

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

01-13-2017 05:03 PM

If you don't know how to use a PROC the first step should be the SAS documentation

Specifically the examples.

The next is Lexjansen.com

The third is here

If you're just measuring the presences of exposure/incident of exposure then Chi Square is appropriate. If metabolite is continuous I think you're dealing with a T-Test - possible a paired t-test because you have before and after for two different exposure groups.

T-Test has PROC TTEST and/or PROC ANOVA and/or PROC GLM that may be applicable.

All Replies

Solution

01-19-2017
11:16 AM

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

01-13-2017 05:03 PM

If you don't know how to use a PROC the first step should be the SAS documentation

Specifically the examples.

The next is Lexjansen.com

The third is here

If you're just measuring the presences of exposure/incident of exposure then Chi Square is appropriate. If metabolite is continuous I think you're dealing with a T-Test - possible a paired t-test because you have before and after for two different exposure groups.

T-Test has PROC TTEST and/or PROC ANOVA and/or PROC GLM that may be applicable.