Repeated measures with no subject

[palolix]

That was very helpful, thank you so much! So for an ANOVA model to check for the interactions between factors I could use catmod or proc logistic as well for ordinal or non-ordinal data. Both would be appropiate right?

Thanks!
Caroline

[Ksharp · Accepted solution]

Yes. You could use PROC CATMOD to do ANOVA.
Use PROC CATMOD or PROC LOGISTIC to do logistic model (either ordinal or non-ordinal data.).

[palolix]

Great, thank you so much for all your support, it has helped me a lot!

[ShivanshuChobey]

The syntax looks reasonable to me.

Thanks,

[jiltao]

No, I do not think treating Variety as the subject is appropriate. 

You do not seem to have repeated measures data. 

Are you saying that your dependent variable has values 1, 2 ,3, 4 or 5? Then you might fit a multinomial / ordinal regression model. But the excel file you sent seems to have a continuous response variable.

Thanks,

Jill

[palolix]

Thank you so much for your feedback Jiltao.  So I will then use genmod as Ksharp suggested since I only have fixed variables. You are right, the var in the excel file is continious so I will use lognormal for that one, but most of my dependent variables are values from 0-3 or 1-5 with half values sometimes like 1.5. In that case I should then choose multinomial distribution right?

Thank you!

Caroline