You want to use [-1 0 1 zeros(1,6)]
This will test if the conditions increase over the three conditions in a linear manner. 

If the values don't sum to zero, which they don't have to, then you will be testing whether the weighted sum of the 3 conditions is greater than 0. This is a very different question.

In terms of the oddness of leaving out the middle term. The value of condition 2 will have no impact on the linear relationship. To try this yourself. Do the following in excel.
Set A1=1, B1=5
Set A2=2, B2=rand()
Set A3=3, B3=10

Now, make an X Y scatterplot, add a trendline with equation. Now keep resetting B2 and you will notice that the slope does not change. Thus, condition 2 will not have an impact on the slope. And the slope can be thought of as the condition3-condition1 (which would be [-1 0 1 0 0 0 0 0 0])

Best Regards, Donald McLaren
D.G. McLaren, Ph.D.
Postdoctoral Research Fellow, GRECC, Bedford VA
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
Office: (773) 406-2464
This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
intended only for the use of the individual or entity named above. If the
reader of the e-mail is not the intended recipient or the employee or agent
responsible for delivering it to the intended recipient, you are hereby
notified that you are in possession of confidential and privileged
information. Any unauthorized use, disclosure, copying or the taking of any
action in reliance on the contents of this information is strictly
prohibited and may be unlawful. If you have received this e-mail
unintentionally, please immediately notify the sender via telephone at (773)
406-2464 or email.

On Fri, Jul 8, 2011 at 11:25 AM, Joshua Balsters <[log in to unmask]> wrote:
Apologies if this is already answered on the list, I found this question posted a couple times but without a response.

I have a GLM with 9 conditions. Three of these conditions increase parametrically so I'd be using a contrast vector [1 2 3 zeros(1,6)]; however i believe that t-contrasts are supposed to sum to zero so I'm not sure this contrast vector is correct. If you sum these to zero then you get [-1 0 1 zeros(1,6)]; which also doesn't seem right.

How important is it that t-contrasts sum to zero? and if it is important what's the appropriate contrast for an odd number of linearly increasing conditions?

Thanks in advance