Dear Dr. McLaren,
Thanks for responding to me. The design is event-related with varying numbers of sessions due to blocks being removed due to movement and such. Is the weighting with whole integers still a problem in the second level analysis if the contrast sums to zero? For example, 2 session subject (1 -1 0 0 1 -1 0 0) and 5 session subject (1 -1 0 0 1 -1 0 0 1 -1 0 0 1 -1 0 0 1 -1 0 0).
Andrew
|