Repeated mesures with a different number of items: correct analysis?

#1
Hello,
I want to compare two groups, a group with a disease and a control group that look to photographs of humans and birds. For the group of humans I grouped the photos into 6 conditions and for the group of humans I did the same, meaning the the same 6 conditions. My natural thought was to do a Repeated measures ANOVA with disease group and control group as between subjects and two levels of within subjects, first level human/bird and the other level with the 6 conditions. However, I have 60 photos of humans and 21 photos of birds. So, the sample is a bit unbalance. In this case, is the Repeated measures ANOVA the correct analysis? If not, what is the best analyses to compare the performance between two groups and within to different levels? Thank you for your help.
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
Will there be multiple people reviewing the images? There being difference numbers doesn't seem to problematic. Can you provide a sample of your data (can be made up), so we can see the structure. Multilevel models may be a better approach.
 
#3
Thank you for your kind reply.
Each group has 50 subjects (two groups, 100 subjects). Each subject view a total of 81 photographs, 18 birds and 60 humans. The photos of birds have 6 conditions (A, B, C , D, E, F) and 3 photos for each condition. The photos of humans have the same 6 conditions (A, B, C, D, E, F) and 10 photos for each condition. They are asked to classified the photos and each answer is rated as correct (0) or incorrect (1). So the organization goes like this for the subject X:

1) 18 photos of birds
6 conditions

A - 1; 0; 1
B - 0; 0;1
C - 1;0;0
D- 1; 1; 1
E - 0; 0; 1
F - 1;1;1

2) 60 photos of humans
6 conditions

A - 1; 1; 1; 0; 0; 1; 1; 1; 0; 0
B - 1; 1; 0; 0; 0; 1; 1; 1; 0; 0
C - 0; 0; 1; 1; 1; 1; 1; 1; 1; 1
D - 1; 1; 1; 1; 1; 1; 1; 1; 1; 1
E - 1; 1; 0; 0; 0; 1; 1; 1; 1; 1
F - 0; 0; 0; 1; 1; 1; 1; 0; 0; 0
 

hlsmith

Less is more. Stay pure. Stay poor.
#4
Why are there two groups? This wasn't my favorite project, but it dealt with having providers view dermatologic slides and try to classify them into two groups. Slides had different characteristics and accompanying information. I did multilevel modeling and reclustered them a couple of ways. Though, like you there were multiple comparisons, so I did a lot of corrections for false discovery (e.g. Bonferroni adjusted alpha level).

https://onlinelibrary.wiley.com/doi/abs/10.1111/cup.12495
 
#5
Thank you very much for your help.
Unfortunately, I can't have access to the article. I am not very familiar with the idea of multilevel model. Would the idea be to make two levels, one level to human and bird photo and the second level to six conditions? I put two groups because I aim to see if there are differences between the group with a disease and a healthy group when evaluating the photos birds and human photos and within the six conditions.
I also desire to evaluate if factors such empathy and experience have an effect in the results. What should I do? Thank you very much