Email discussion lists for the UK Education and Research communities

## SPM@JISCMAIL.AC.UK

#### View:

 Message: [ First | Previous | Next | Last ] By Topic: [ First | Previous | Next | Last ] By Author: [ First | Previous | Next | Last ] Font: Proportional Font
 LISTSERV Archives SPM Home SPM May 2017

#### Options

Subject:

Re: Two-Sample t-Test

From:

Date:

Mon, 22 May 2017 17:45:30 +0200

Content-Type:

text/plain

Parts/Attachments:

 text/plain (38 lines)
 ```Dear Sascha, I understand that your contrast is a t-contrast. T-constrast have a direction, for this reason it is used when we compare between groups, while F-Contrast is for showing an effect of interest. If you have group 1 .vs. group 2 and put a t-constrast 1 -1 you are supposing that group 1 is major that group 2. If, in the contrary you créate a t-contrast -1 1 you are supposing that the effect of the second group is major. In other words, if you have significances in the contrast 1 -1, the meanign is that you have more activity in the group 1 than group 2. In the other case, the group 1 shows less activity that the group 2. These meaning is more or less, it depends that the study that you are doing. I hope that I help you! If I am wrong, please SPM users, tell me! Best, Héctor Espinós INCLIVA Public Health Institute El 2017-05-22 11:42, Sascha Frölich escribió: > Dear all, > > I have a question regarding the two-sample t-test in SPM: > > When comparing group 1 vs group 2 (ie. contras 1 -1) I get a different > result from when I use contrast -1 1 (group2 vs group 1, as far as I > understood). How is this possible? Shouldn't the t-test by symmetric? > What do the different results mean? > > Thanks! > > Best, > Sascha ```