Page 1 of 2
Cannot run ANOVA, but can perform GLM with the same datat
Posted: Wed Nov 10, 2021 12:17 pm
by IlzeD
Hi everybody,
I absolutely love Jamovi. However, I struggle with some difficulties time by time. I not that difficulties are most likely caused by my not-expertise.
Now I work with data set having 2 factors one of which has only 2 groups (two genotypes). When I try performing ANOVA, I get SS=0 and dF NaN for factor having 2 groups.
At the same time, when I perform GLM with the same data, I get a nice output of ANOVA results.
So far I know that mathematically there is no difference between ANOVA and GLM (isn't it?). However, I am confused why ANOVA doesn't work for my data. Any ideas?
The second thing - is it somehow possible to get single-factor ANOVA output containing sum of squares and Mean square?
Many thanks in advance,
Ilze
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Wed Nov 10, 2021 6:24 pm
by mcfanda@gmail.com
Hi
would you like to share an example of this issue?
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Thu Nov 11, 2021 8:03 am
by IlzeD
Hi, how can I share an example? What shoul I provide?
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Thu Nov 11, 2021 9:27 am
by mcfanda@gmail.com
the .omv file containing the analyses that create the issue
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Thu Nov 11, 2021 11:41 am
by IlzeD
I submit my .omv file. Sorry, factors and variables have names in my language. The variable I struggle with is "skirne" - it has two levels. I want to run factorial ANOVA for "Skirne" and "BarotneTrauks" and get trouble with it for any variable (let's say "FWaugsMg"
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Thu Nov 11, 2021 6:55 pm
by MAgojam
Hi @IlzeD,
the problem you reported is due to the presence of some characters used in your language.
In this case the first two characters in the variable name.
As you can see in the attached screenshot, deleting them from the variable name made it possible to get the correct statistic.

- tempsnip.png (69.51 KiB) Viewed 14977 times
Cheers,
Maurizio
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Fri Nov 12, 2021 8:52 am
by IlzeD
Hi, Maurizio.
Thanks a lot for your help, it really works. However, I remain confused - why it is possible to run one-way ANOVA and GLM with the same factor name (just curious).
What about my second question? is it somehow possible to get single-factor ANOVA output containing sum of squares and Mean square?
Ilze
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Fri Nov 12, 2021 9:37 am
by jonathon
Thanks a lot for your help, it really works. However, I remain confused - why it is possible to run one-way ANOVA and GLM with the same factor name (just curious).
that i am not sure about. it looks to be a bug in R. i'll look into it. i'm reasonably confident you'd only experience this issue under windows.
What about my second question? is it somehow possible to get single-factor ANOVA output containing sum of squares and Mean square?
run a second ANOVA and only specify a single factor? (perhaps i don't understand the question)
cheers
jonathon
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Fri Nov 12, 2021 9:52 am
by IlzeD
Hi, Jonathon.
Actually, my question is very simple

I run one-way Anova (with one factor or grouping variable) and I get output table containing only F value Df1 and Df2 and p value. I wonder if it is possible to get output similar to excel, SPSS and similar ANOVA outputs that contain SS and MS values? Between and within groups?
I apologize if my questions are dumb

I am not a statistician, i just do a routine analysis of my data
Cheers,
Ilze
Re: Cannot run ANOVA, but can perform GLM with the same data
Posted: Fri Nov 12, 2021 9:58 am
by jonathon
use ANOVA, rather than One-way ANOVA, and simply specify a single factor. like this:

- Screen Shot 2021-11-12 at 20.56.26.png (178.75 KiB) Viewed 14848 times