-
1 Attachment(s)
randomization mistake
Hi.
I ran a study last Fall where I administered a Stroop. In this particular version I needed to randomize both within and between groups. That is, I had to randomize incongruent and congruent trials within blocks, but also needed to present two separate types of blocks that needed to appear in a specific order one after the other--what I called cong-majority and incong-majority blocks. Please see the attached file. I thought I did this properly (seems straightforward), but when I looked at the data, it became clear, that although the within-group randomization worked, the separate blocks were not presented separately--both types of blocks were interspersed. Could you please let me know where I went wrong?
Thanks!
-
Yes, the WGR values seem fine. For the BGR values, let me ask to be clear--you wanted the the first five blocks to stay together a group and the second set of five blocks to also be kept together? And then you wanted the order of those two groups to be random? Such that they would get either get all the [I]cong[/I] sets first *or* all the [I]incong[/I] sets first?
-
yes this is correct. either all of cong first or all of incong first, as long as they stay together within.
-
I think one constraint on BGR applications is that there can not be a break in the set of items sharing the same BGR value. The few trials that break the BGR blocks may be what's causing the problem here. Can you add them to the blocks using the corresponding BGR values?