Error in log regression

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
amykah
Posts: 4
Joined: Thu Nov 30, 2023 9:29 pm

Error in log regression

Post by amykah »

I am getting a "Table$setRow(): value 'est' is not atomic" error. I do not know how to figure out which variable is causing the problem, because I think my variables are pretty atomic. I'm sharing my file so that someone might be able to help me proceed. I am brand-new to Jamovi. Let me know if you have any questions I need to answer before you can help.
Attachments
2023.11.30 ASACN Dataset.omv
(156.48 KiB) Downloaded 414 times
amykah
Posts: 4
Joined: Thu Nov 30, 2023 9:29 pm

Re: Error in log regression

Post by amykah »

I've narrowed the problem down to my dummy variables. The error does not come up when I only have my covariates in the model, but adding in any of my factors creates the error. I have five dummy variables for different races, and the reference group (when all 5 variables = 0) is white kids. I have three dummy variables for public insurance, private insurance, and both types of insurance, with a reference group of no insurance.

Does jamovi handle dummy variables differently than SPSS?
User avatar
MAgojam
Posts: 422
Joined: Thu Jun 08, 2017 2:33 pm
Location: Parma (Italy)

Re: Error in log regression

Post by MAgojam »

Hey @amykah,
perhaps you are not using the latest version of jamovi, because if you were, you would receive a different and more understandable error message.
That is this:
Error: One or more coefficients in model '3' could not be estimated due to perfect collinearity.

This error message obviously appears because you have the Collinearity statistics check box active which reminds you to look at Model 3.
In Blok 3 of the Model Builder you have 3 variables "PubIns", "PrivIns" and "Bothins", which you should try to remove individually leaving two and seeing what changes in the Model Fit Measures table.

Cheers,
Maurizio
Post Reply