turn on suggestions

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

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- Comparing means of unequal distributions

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
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 2, 2016 4:46 PM
(1322 views)

Hi,

I have measurements from two groups with a large variance (for which I have already corrected by a Log10 transformation) and unequal distributions (Bartlett p=0.009).

What is the proper way to compare the means of these two groups?

The data can also be paired. Is it correct to use the paired t-test on the Log10 data?

Thanks

4 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 2, 2016 6:57 PM
(1313 views)

It is proper to do a paired t test on the transformed data.

Jim

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 2, 2016 9:17 PM
(1307 views)

Thanks a lot.

And the groups in general?

Also a t test on Log-transformed data?

And the groups in general?

Also a t test on Log-transformed data?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 3, 2016 4:18 AM
(1292 views)

You will want to used the log transformed data since that transformation results in a normal distribution. And you would want to use a paired t-test if the data warrants it, to increase the power of the test.

Jim

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 3, 2016 10:34 AM
(1272 views)

Thanks a lot, Jim