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
- /
- Analytics
- /
- Stat Procs
- /
- T TEST

Topic Options

- 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

09-23-2010 02:43 PM

Hello All,

I have two datasets: one & two. I want to test if there is a difference in the mean score (variable 'SCORE') of two persons in these two datasets. The variable PERSON has two persons (A & B) in it in both datasets.

Could anyone please suggest me a code to deal with two datasets in PROC t test ?

Kindest Regards,

Kriti

I have two datasets: one & two. I want to test if there is a difference in the mean score (variable 'SCORE') of two persons in these two datasets. The variable PERSON has two persons (A & B) in it in both datasets.

Could anyone please suggest me a code to deal with two datasets in PROC t test ?

Kindest Regards,

Kriti

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

09-23-2010 03:57 PM

you need to provide more detail. We need some sample data and a more precise statement of how the variables relate to each other and to PERSON.

In general, you need to combine the two datasets into one. You can do that with the DATA step. If the two datasets represent pairings, then the MERGE statement would be needed; otherwise SET. In SQL, the comparable clauses are JOIN and UNION.

If your two scores are paired, then you need to take the difference and compare that to 0 using PROC MEANS. If they are independent, you would use the TTEST procedure with some classification variable.

The BY statement takes care of the person variable.

Doc Muhlbaier

Duke

In general, you need to combine the two datasets into one. You can do that with the DATA step. If the two datasets represent pairings, then the MERGE statement would be needed; otherwise SET. In SQL, the comparable clauses are JOIN and UNION.

If your two scores are paired, then you need to take the difference and compare that to 0 using PROC MEANS. If they are independent, you would use the TTEST procedure with some classification variable.

The BY statement takes care of the person variable.

Doc Muhlbaier

Duke