 |
 |
Dears, Thanks very much for your good information. Just now I have a question about contrast, I did not understand the difference between [1 0] and [-1 1]. Dear Mclaren, I did not get the point on your sentence: The contrast [1 0] tests whether or not group 1 is significantly greater from 0. Is it possible to explain me more. Is there any references to introduce and teach contrasts? Thanks very much in advanced. Regards Maryam On Fri, Mar 9, 2012 at 6:46 PM, sarika cherodath <[log in to unmask]>wrote: > Hello, > > I would suggest you do as Donald suggested if you have two groups of > subjects varying greatly in age.For eg. Group1 - 7-14 year old and Group 2 > - 30-60 year olds. If you have subjectts from a spectrum of ages varying > from 7-60 years, then you can do as i suggested. > > There is one such study i know where they have looked at the data in the > second way.Here is the link > > http://www.nature.com/neuro/journal/v6/n7/full/nn1065.html (Check figure > 6) > > Regards, > Sarika. > > > On Fri, Mar 9, 2012 at 9:33 PM, maryam momeni <[log in to unmask]>wrote: > >> Dear Sarika, >> >> The images in two groups are not related to the same subjects. I mean my >> subjects are not longitudinal. My images are classified to different groups >> according to their age and because of this I used two-sample t-test. Then >> with this situation, you recommend one sample t-test? >> >> Regards >> Maryam >> >> >> On Fri, Mar 9, 2012 at 4:40 PM, sarika cherodath <[log in to unmask]>wrote: >> >>> Hello, >>> >>> If you want to look at how the activations vary with age, and if >>> you have a continuous distribution or range of ages among your subjects, >>> then you can simply use age as a covariate and do a one sample t test >>> clubbing all subjects together. >>> >>> >>> Hope this helps. >>> >>> On Fri, Mar 9, 2012 at 8:53 PM, maryam momeni <[log in to unmask]>wrote: >>> >>>> Hi Dears, >>>> >>>> I have two groups and I want to compare them corresponding age. >>>> >>>> I used two-sample t-test. I imported the data of each group with out >>>> any covariate. I estimated them using classical method. For comparison, I >>>> used t-test with the contrast: 1 0 and 1 0 that their results are the same >>>> as it is predicted. I have two questions: >>>> >>>> 1- the steps and contrast are true for age analysis? >>>> >>>> 2- the T score are very high with compare to T scores in papers. The T >>>> score in papers are less than 10 but my T score is more than 50 (maximum is >>>> 80). I confused! >>>> >>>> Thanks for your helping me. >>>> >>>> Regards >>>> Maryam >>>> >>> >>> >>> >>> -- >>> Sarika Cherodath >>> Junior Research Fellow >>> National Brain Research Centre >>> Manesar, Gurgaon -122050 >>> India >>> >>> >> > > > -- > Sarika Cherodath > Junior Research Fellow > National Brain Research Centre > Manesar, Gurgaon -122050 > India > >