it is invalid because the sum of them was not 0.
but I also want to know if the following are right:
[-1 -1 -1 -1 4]
and what it mean.
Thank you very much
-----原始邮件-----
发件人:"Robin CHARK" <[log in to unmask]>
发送时间:2005-09-28 19:53:52
收件人:[log in to unmask]
抄送:(无)
主题:Re: are the two designs same?
One follow-up question. If I want to test the significancy of the first 4 conditions only, can I define it as [1 1 1 1 0]? Thanks. Robin > hi > >> Dear all: >> >> If I had a five conditions fMRI experiment, and made the following two >> t-contrasts using spm2 : >> >> [1 1 1 1 -1] and [1 0 0 0 -1; >> >> 0 1 0 0 -1; >> >> 0 0 1 0 -1; >> >> 0 0 0 1 -1] >> >> Are they all valid? And are they the same? If not, what are they >> meaning separately? >> >> Tank you very much! >> > well, they are not all valid and they are not the same .. > 1. When you enter a contrast, the sum of coef must be 0 and thus [1 1 1 > 1 -1] is invalid (you can try but SPM contrast manager will refuse to do > that) .. if you want to oppose one condition against the others the > correct contrast would be [1 1 1 1 -4] > 2. in the second case you test if any of these contrasts (or > combinations) is significant .. but those ones are correct as they all > sum up to 0 > > best > cyril