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 July 2011

#### Options

Subject:

Re: differences between linear regressions

From:

Date:

Tue, 12 Jul 2011 10:12:04 +0100

Content-Type:

text/plain

Parts/Attachments:

 text/plain (56 lines)
 ```Hi Chiara Following your design matrix I would say that 0 0 1 -1 0 is the right thing to do ie will tell you in which voxels slope sample 1 > slope sample 2. Usually I also run something like 0 0 1 0 0 and 0 0 0 1 0 (or using -1 in your case) allowing to have a more 'qualitative' interpretation, that is sample 1 and sample 2 can both have a significant correlation but one is stronger than the other or maybe only one sample shows the correlation and that mean that in your patients there is no effect of age which changes the way you can interpret the results Hope this helps Cyril > Dear SPM users, > > in my latest analysis I have two samples and I want to study the dependance of certain ROIs volume upon age. > > First I checked if the two samples come from two different populations: I prepared a full factorial considering one factor (=group), one covariate (age) interacting with the factor and another covariate (scolarity) not interacting. The matrix has therefore 5 columns: in the first two there are the images (sample 1 and sample2), in the third and the fourth there is the age of the two samples (age sample 1 and age sample 2) and in the fifth there is the scolarity of all subjects. > > In certain ROIs I found clusters in which there is a main effect of group and, in these clusters, volume turns out to be negatively correlated with age in both groups: the latter information comes from a separate multiple regression analysis. > Now I want to check if the slopes of the regression lines are statistically different. How should I do? I have tried creating, for instance, a contrast with 0 0 1 -1 0, but I am not sure... > > Can anyone help me? > > Thanks in advance, > Chiara > > -- Dr Cyril Pernet, Academic Fellow SBIRC fMRI Manager SFC Brain Imaging Research Center Division of Clinical Neurosciences University of Edinburgh Western General Hospital Crewe Road Edinburgh EH4 2XU Scotland, UK [log in to unmask] tel: +44(0)1315373661 http://www.sbirc.ed.ac.uk/cyril http://www.sinapse.ac.uk/ The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. ```