3DANOVA3 Error

AFNI Message Board
1

Hello,

I have a 2-way mixed design (2 conditions at 2 time points)

I keep receiving the following error:

++ 3dANOVA3: AFNI version=AFNI_19.2.10 (Aug 7 2019) [64-bit]
++ Authored by: B. Douglas Ward
** FATAL ERROR: Program 3dANOVA3: must have equal sample sizes for 3dANOVA3
** Program compile date = Aug 7 2019

when I run:

#!/bin/bash

Bank=4

echo $Bank

3dANOVA3 -type 4 -alevels 2 -blevels 2 -clevels 14
-dset 1 1 1 stats.VGT08pre.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 1 2 1 stats.VGT08post.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 1 1 2 stats.VGT12pre.MNI+tlrc’[‘“$Bank”’]’
-dset 1 2 2 stats.VGT12post.MNI+tlrc’[‘“$Bank”’]’
-dset 1 1 3 stats.VGT13pre.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 1 2 3 stats.VGT13post.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 1 4 stats.VGT17pre.MNI+tlrc’[‘“$Bank”’]’
-dset 2 2 4 stats.VGT17post.MNI+tlrc’[‘“$Bank”’]’
-dset 1 1 5 stats.VGT18pre.MNI+tlrc’[‘“$Bank”’]’
-dset 1 2 5 stats.VGT18post.MNI+tlrc’[‘“$Bank”’]’
-dset 2 1 6 stats.VGT19pre.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 2 6 stats.VGT19post.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 1 7 stats.VGT21pre.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 2 7 stats.VGT21post.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 1 1 8 stats.VGT24pre.MNI+tlrc’[‘“$Bank”’]’
-dset 1 2 8 stats.VGT24post.MNI+tlrc’[‘“$Bank”’]’
-dset 2 1 9 stats.VGT28pre.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 2 9 stats.VGT28post.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 1 10 stats.VGT31pre.MNI+tlrc’[‘“$Bank”’]’
-dset 2 2 10 stats.VGT31post.MNI+tlrc’[‘“$Bank”’]’
-dset 1 1 11 stats.VGT33pre.MNI+tlrc’[‘“$Bank”’]’
-dset 1 2 11 stats.VGT33post.MNI+tlrc’[‘“$Bank”’]’
-dset 1 1 12 stats.VGT35pre.MNI+tlrc’[‘“$Bank”’]’
-dset 1 2 12 stats.VGT35post.MNI+tlrc’[‘“$Bank”’]’
-dset 2 1 13 stats.VGT36pre.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 2 13 stats.VGT36post.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 1 14 stats.VGT37pre.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-dset 2 2 14 stats.VGT37post.MNI.LRFlip+tlrc’[‘“$Bank”’]’
-fa condition
-fb time
-fab time_cond_interact
-adiff 1 2 CCFESvsVG
-bdiff 2 1 PostvsPre
-aBdiff 1 2 : 1 CCFESvsVG_Pre
-aBdiff 1 2 : 2 CCFESvsVG_Post
-Abdiff 1 : 2 1 PostvsPre_CCFES
-Abdiff 2 : 2 1 PostvsPre_VG
-bucket tDCS_anova

Any thoughts on what I’m doing wrong?

2

I think 3dANOVA3 requires a complete input table. So with 2 conditions x 2 time points x 14 subjects, there should be 56 = 2x2x14 -dset entries, where each subject has 4 entries. Or are the conditions separate between 2 groups of 7 subjects? If so, then perhaps clevels should be 7.

  • rick
3

oooh, thanks! I was confused what C level meant. I assumed a level would group them into their ‘groups’ where c level signified if it was independent or a paired test (i.e. like values means same participant)

It works with your suggestion.