Print

---

I know there have been multiple threads regarding contrasts involving empty EVs, and the general take home message is that you should remove any subjects with empty EV when looking at any contrasts involving that EV in the higher level analysis. But I was wondering if there is a way around it in the following case.

Let’s say it is a memory experiment with 3 different conditions (c1,c2, c3) and all can be divided into remembered (R) and forgotten (F), and I’m interested in R>F contrast for each condition as well as average activation for R, F, and R>F across conditions, so that my design is

c1_R   c1_F   c2_R   c2_F   c3_R   c3_F   

c1 R>F 1    -1        0         0         0         0     

c2 R>F 0         0          1         -1        0         0 

c3 R>F 0         0          0         0         1         -1

R 1    0          1         0         1         0

F              0         1          0         1         0         1

R>F         1         -1         1         -1        1         -1

Let’s say sub1 has no forgotten trial for condition 1, so c1_F is an empty EV. I’ll have to exclude this subject for the first contrast, c1 R > F, but I was wondering if I could keep the subject for F and R>F by modifying the contrast as

F              0          0          0          1.5         0        1.5

R>F 1          0          1          -1.5        1        -1.5  

And then use the resulting COPEs in the higher level analysis? In particular I wasn’t sure if I’m allowed to use decimal numbers like 1.5 in the contrast matrix to make the scaling equivalent across subjects…

Thank you in advance for your input,

Ami Tsuchida