Thank you for help! This sure is usefull information and knowledge I can use for my thesis.

But concluding: my graph is not reliable because the small sample size in the subgroups? So I can’t conclude that autism is a safeguard just because my graph says so. I need to stick with my statistics, concluding a parallel age-decline in visual memory for both groups. However because of the sample size problem, my statistics might me biased as well. So I would need to discuss this point in my thesis, is that correct?

Is there a non parametric method for a multiple regression or MANOVA? I already did run the mann whitney, but my supervisor didn’t agree with it, because he said it wasn’t a replacement for the MANOVA (I used the MANOVA for groupcomparisons). Also, I can’t choose a larger subgroup, I got this dataset from my supervisor and I’m already in the last phase of my thesis. The only thing I have to do is to write a discussion and conclusion. But what you’re saying might be an idea for future research. Do you agree with that?

Also, in my graph you can see that autism and control show a parallel decline over time till age 65, after age 65 the lines are still parallel but the line for autism lays way above the line for controls. Is this because of that small sample size for the 65+ subgroup?

Thank you for your comment! Now that you point it out I definitely believe that my graph is manipulated by a small group of points. My controlgroup (N = 129) is indeed bigger than my autism group (N = 76). I was interested in subgroups as well, namely a 65+ subgroup and 30-64 years old subgroup. My 65+ subgroup consists of only 27 people, with N = 5 for autism and N = 22 for my controlgroup. I am aware of the fact that this is a extremely small sample size for my subgroup. So based on that: I believe I can assume my statistics are telling me the truth but not my graph?

Thank you all for your comments! I found out how to do it: using a line graph instead of a scatterplot, and then plotting age and my predicted values instead of age and dependent variable.

Thank you for your comment. My supervisor indeed said that age only as a categorical variable wouldn't be good. But he recommended me to do both. My second supervisor approved my research proposal as well. That's why I don't think I should remove subgroup (age) now, since I'm in the last phase of my master thesis. I did indeed keep age continous while graphing, but as I said before in my post, my graph now only illustrates the groupxage interaction you get when you run a regression seperately.