Linear Regression - "Argument is of length zero"

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)
Post Reply
SierraTango
Posts: 1
Joined: Thu Aug 06, 2020 7:50 pm

Linear Regression - "Argument is of length zero"

Post by SierraTango »

Hello alltogether,

I am trying to calculate some multiple regressions of two continuous covariates and one nominal factor. The dependent variables are continuous, too.
Therefore I am using the "linear regression"- tool of Jamovi.
Sometimes it produces the error: "Argument is of length zero".
I tried changing around variables, it works out fine most of the time, but with one specific nominal factor it always produces the aforementioned error.

Any ideas how to fix this problem?

Greetings
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Linear Regression - "Argument is of length zero"

Post by jonathon »

hi,

could you double check this is still a problem in the 1.2.25, and if it is, can you send us an .omv file with the analysis and the error? either attach it here, or send it to contact <at> jamovi.org

with thanks
Hslater246
Posts: 3
Joined: Thu Jul 22, 2021 2:07 pm

Re: Linear Regression - "Argument is of length zero"

Post by Hslater246 »

I am struggling with the same issue. I am trying to do a moderation multiple regression, with an interaction term which includes a categorial variable and two predictor continuous variables but it keeps coming up with 'argument is of length zero' when i try and do an interaction with all three things. If interaction with just the categorical and a predictor is works. Please help as not sure why it isn't working??
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Linear Regression - "Argument is of length zero"

Post by jonathon »

i think the latest version of jamovi improves the error message here. this is usually caused by "aliased coefficients".

take a look here:

https://stats.stackexchange.com/questions/112442/what-are-aliased-coefficients

if that doesn't appear to be the problem, update to the latest version of jamovi, and if it's still an issue, email the file to contact <at> jamovi.org

cheers

jonathon
Hslater246
Posts: 3
Joined: Thu Jul 22, 2021 2:07 pm

Re: Linear Regression - "Argument is of length zero"

Post by Hslater246 »

Thank you for your quick reply Jonathan. I am also trying to test the simple slopes but in jamovi they all have to be continuous variables, is there anyway you can do this with a categorical variable and with two predictors?
Hslater246
Posts: 3
Joined: Thu Jul 22, 2021 2:07 pm

Re: Linear Regression - "Argument is of length zero"

Post by Hslater246 »

Hi Jonathon,

Also when I download the newest version it keeps coming up as the old version when I open it up, not sure what to do about this?

Thanks,
Hannah
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Linear Regression - "Argument is of length zero"

Post by jonathon »

> I am also trying to test the simple slopes but in jamovi they all have to be continuous variables, is there anyway you can do this with a categorical variable and with two predictors?

i'm not sure. maybe maurizio will chime in.

> Also when I download the newest version it keeps coming up as the old version when I open it up, not sure what to do about this?

after downloading the new version, you need to install it and replace the old version.

cheers

jonathon
User avatar
MAgojam
Posts: 421
Joined: Thu Jun 08, 2017 2:33 pm
Location: Parma (Italy)

Re: Linear Regression - "Argument is of length zero"

Post by MAgojam »

Hi, Jonathon.
I was around here and I saw you mentioned me.

I would suggest @Hslater246 to take a look at Marcello Gallucci's jAMM module for jamovi.
You can also read here: https://jamovi-amm.github.io/glm.html

I hypothesize that there may be missing data in the file, which is responsible for the reported error, but if for various reasons it is not possible to share the file, it is not easy to help.

Cheers,
Maurizio
Post Reply