Support for psych::omega models #1115
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Summary of Changes
Hello @strengejacke, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!
This pull request significantly enhances the parameters R package by integrating robust support for psych::omega models. It involves a substantial refactoring of the model_parameters.omega function to provide a richer, more structured output, alongside improvements to the display and printing functionalities, ensuring that users can easily access and interpret the key results from their omega analyses.
Highlights
- New Feature:
psych::omegaModel Support: This pull request introduces comprehensive support forpsych::omegamodels, allowing their parameters to be processed and displayed within theparameterspackage. - Refactored
model_parameters.omegaFunction: The coremodel_parameters.omegafunction has been significantly refactored. It now returns factor loadings as its primary output, while storing omega coefficients and variance summaries as attributes. This new structure provides a more flexible and detailed representation of the model results. - Enhanced Display and Print Methods: Dedicated
displayandprintmethods have been added or updated forparameters_omegaandparameters_omega_summaryclasses. These methods now leverage existing EFA (Exploratory Factor Analysis) display logic for loadings and provide a more structured and informative output for omega coefficients and variance summaries, improving readability for users.
Using Gemini Code Assist
The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.
Invoking Gemini
You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.
| Feature | Command | Description |
|---|---|---|
| Code Review | /gemini review |
Performs a code review for the current pull request in its current state. |
| Pull Request Summary | /gemini summary |
Provides a summary of the current pull request in its current state. |
| Comment | @gemini-code-assist | Responds in comments when explicitly tagged, both in issue comments and review comments. |
| Help | /gemini help |
Displays a list of available commands. |
Customization
To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.
Limitations & Feedback
Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.
You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.
Footnotes
-
Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩
There was a problem hiding this comment.
Code Review
This pull request introduces support for psych::omega models, including a new model_parameters.omega method and S3 methods for printing and summarizing. I recommend removing the unused verbose argument and initializing caption1 and caption2 to NULL before the if blocks.
|
This is the current state. Maybe we can discuss the parameters related stuff for omega here. Anything important missing in this output? Other things will go into model <- psych::omega(mtcars, nfactors = 3, plot = FALSE)
#> Loading required namespace: GPArotation
parameters::model_parameters(model)
#> # Rotated loadings from Omega (oblimin-rotation)
#>
#> Variable | g | F1* | F2* | F3* | h2 | u2 | p2 | Complexity
#> ----------------------------------------------------------------------------------------
#> mpg- | 0.58 | -0.67 | 0.09 | 0.29 | 0.88 | 0.12 | 0.38 | 2.40
#> cyl | 0.70 | -0.61 | 0.28 | 0.07 | 0.96 | 0.04 | 0.52 | 2.33
#> disp | 0.59 | -0.71 | 0.18 | 0.11 | 0.89 | 0.11 | 0.39 | 2.13
#> hp | 0.77 | -0.31 | 0.23 | 0.36 | 0.87 | 0.13 | 0.68 | 2.00
#> drat- | 0.27 | -0.79 | 0.06 | -0.07 | 0.71 | 0.29 | 0.10 | 1.26
#> wt | 0.43 | -0.79 | -0.04 | 0.31 | 0.91 | 0.09 | 0.20 | 1.87
#> qsec- | 0.81 | 0.19 | 0.50 | 0.06 | 0.95 | 0.05 | 0.70 | 1.81
#> vs- | 0.74 | -0.27 | 0.38 | 0.05 | 0.77 | 0.23 | 0.71 | 1.81
#> am- | 8.38e-03 | -0.89 | -0.15 | -9.51e-03 | 0.81 | 0.19 | 8.63e-05 | 1.06
#> gear | 0.03 | 0.87 | 9.01e-03 | 0.32 | 0.87 | 0.13 | 9.03e-04 | 1.27
#> carb | 0.68 | 0.06 | 0.10 | 0.63 | 0.87 | 0.13 | 0.53 | 2.06
parameters::model_parameters(model) |> summary()
#> # Omega Statistics
#>
#> Statistic | Coefficient
#> ----------------------------------
#> Alpha | 0.88
#> G.6 | 0.97
#> Omega (hierachical) | 0.57
#> Omega (asymptotic H) | 0.58
#> Omega (total) | 0.97
#>
#> # Omega Coefficients
#>
#> Composite | Omega (total) | Omega (hierarchical) | Omega (group)
#> ----------------------------------------------------------------
#> g | 0.97 | 0.57 | 0.26
#> F1* | 0.90 | 0.31 | 0.59
#> F2* | 0.91 | 0.69 | 0.22
#> F3* | 0.87 | 0.60 | 0.28
#>
#> # Variances
#>
#> Composite | Total (%) | General Factor (%) | Group Factor (%)
#> -------------------------------------------------------------
#> g | 97.28 | 56.64 | 26.42
#> F1* | 90.12 | 31.07 | 59.05
#> F2* | 91.37 | 69.32 | 22.04
#> F3* | 87.36 | 59.65 | 27.71Created on 2025-06-19 with reprex v2.1.1 |
|
Thresholds: model <- psych::omega(mtcars, nfactors = 3, plot = FALSE)
#> Loading required namespace: GPArotation
parameters::model_parameters(model)
#> # Rotated loadings from Omega (oblimin-rotation)
#>
#> Variable | g | F1* | F2* | F3* | h2 | u2 | p2 | Complexity
#> ----------------------------------------------------------------------------------------
#> mpg- | 0.58 | -0.67 | 0.09 | 0.29 | 0.88 | 0.12 | 0.38 | 2.40
#> cyl | 0.70 | -0.61 | 0.28 | 0.07 | 0.96 | 0.04 | 0.52 | 2.33
#> disp | 0.59 | -0.71 | 0.18 | 0.11 | 0.89 | 0.11 | 0.39 | 2.13
#> hp | 0.77 | -0.31 | 0.23 | 0.36 | 0.87 | 0.13 | 0.68 | 2.00
#> drat- | 0.27 | -0.79 | 0.06 | -0.07 | 0.71 | 0.29 | 0.10 | 1.26
#> wt | 0.43 | -0.79 | -0.04 | 0.31 | 0.91 | 0.09 | 0.20 | 1.87
#> qsec- | 0.81 | 0.19 | 0.50 | 0.06 | 0.95 | 0.05 | 0.70 | 1.81
#> vs- | 0.74 | -0.27 | 0.38 | 0.05 | 0.77 | 0.23 | 0.71 | 1.81
#> am- | 8.38e-03 | -0.89 | -0.15 | -9.51e-03 | 0.81 | 0.19 | 8.63e-05 | 1.06
#> gear | 0.03 | 0.87 | 9.01e-03 | 0.32 | 0.87 | 0.13 | 9.03e-04 | 1.27
#> carb | 0.68 | 0.06 | 0.10 | 0.63 | 0.87 | 0.13 | 0.53 | 2.06
parameters::model_parameters(model, threshold = 0.3)
#> # Rotated loadings from Omega (oblimin-rotation)
#>
#> Variable | g | F1* | F2* | F3* | h2 | u2 | p2 | Complexity
#> -------------------------------------------------------------------------
#> mpg- | 0.58 | -0.67 | | | 0.88 | | 0.38 | 2.40
#> cyl | 0.70 | -0.61 | | | 0.96 | | 0.52 | 2.33
#> disp | 0.59 | -0.71 | | | 0.89 | | 0.39 | 2.13
#> hp | 0.77 | -0.31 | | 0.36 | 0.87 | | 0.68 | 2.00
#> drat- | | -0.79 | | | 0.71 | | 0.10 | 1.26
#> wt | 0.43 | -0.79 | | 0.31 | 0.91 | | 0.20 | 1.87
#> qsec- | 0.81 | | 0.50 | | 0.95 | | 0.70 | 1.81
#> vs- | 0.74 | | 0.38 | | 0.77 | | 0.71 | 1.81
#> am- | | -0.89 | | | 0.81 | | 8.63e-05 | 1.06
#> gear | | 0.87 | | 0.32 | 0.87 | | 9.03e-04 | 1.27
#> carb | 0.68 | | | 0.63 | 0.87 | | 0.53 | 2.06Created on 2025-06-19 with reprex v2.1.1 |
|
And in markdown: model <- psych::omega(mtcars, nfactors = 3, plot = FALSE)
#> Loading required namespace: GPArotation
parameters::model_parameters(model) |> insight::display()
Rotated loadings from Omega (oblimin-rotation) parameters::model_parameters(model, threshold = 0.3) |> insight::display()
Rotated loadings from Omega (oblimin-rotation) parameters::model_parameters(model) |> insight::display(threshold = 0.3)
Rotated loadings from Omega (oblimin-rotation) Created on 2025-06-19 with reprex v2.1.1 |
|
nothing important missing that I can see .... looks great! |
|
Great! We can still add / modify things, if needed. Fit indices go into performance. |
|
@gemini-code-assist review |
There was a problem hiding this comment.
Pull Request Overview
This PR adds support for psych::omega models by extending functionality across tests, documentation, and S3 methods for model parameters and factor scores.
- Introduces new tests for omega models in testthat and updates snapshot files.
- Updates function documentation, print methods, and S3 registrations to incorporate psych::omega support.
- Bumps the package version to 0.26.0.5.
Reviewed Changes
Copilot reviewed 12 out of 12 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
| tests/testthat/test-factor_analysis.R | Added tests for omega models using psych::omega |
| tests/testthat/_snaps/factor_analysis.md | Updated snapshot tests to include omega model outputs |
| man/model_parameters.principal.Rd | Revised documentation to mention psych::omega support |
| man/factor_scores.Rd | Updated documentation for factor_scores to include omega models |
| R/utils_pca_efa.R | Added and modified functions to handle omega summary formatting |
| R/print_md.R | Added printing methods for omega summaries |
| R/methods_psych.R | Updated model_parameters.omega to support omega, corrected test usage |
| R/factor_scores.R | Added new S3 methods for omega model scores |
| R/display.R | Registered built-in S3 display functions for omega |
| NEWS.md | Updated news to reflect added support for Omega models |
| NAMESPACE | Registered new S3 methods for omega support |
| DESCRIPTION | Bumped package version to 0.26.0.5 |
Comments suppressed due to low confidence (1)
man/model_parameters.principal.Rd:100
- The term 'hierachical' appears in the documentation; please update the spelling to 'hierarchical' to match standard terminology.
of \code{0.20} suggests that \verb{20\%} or that variable's variance is not shared
There was a problem hiding this comment.
Code Review
This pull request introduces support for psych::omega models, aligning their handling with existing EFA and PCA model support within the parameters package. The changes are comprehensive, including refactoring model_parameters.omega for richer output, adding and updating print and summary methods, and ensuring documentation and tests are current. A minor typo was found in a user-facing string within the summary output.
Fixes #136