Partial matching

[chainsawriot]

Describe the bug

Partial matching is used in .anova_es.anova() (and probably many other places).

To Reproduce

options(
  warnPartialMatchArgs = TRUE,
  warnPartialMatchAttr = TRUE,
  warnPartialMatchDollar = TRUE
)

data(mtcars)
mtcars$am_f <- factor(mtcars$am)
mtcars$cyl_f <- factor(mtcars$cyl)

model <- aov(mpg ~ am_f * cyl_f, data = mtcars)

library(effectsize)
eta2 <- effectsize::eta_squared(model)
#> Warning in model_parameters.aov(model, verbose = verbose, effects = "fixed"):
#> partial argument match of 'effects' to 'effectsize_type'

sessionInfo()
#> R version 4.4.0 (2024-04-24)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 22.04.4 LTS
#> 
#> Matrix products: default
#> BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.10.0 
#> LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=de_DE.UTF-8        LC_COLLATE=en_US.UTF-8    
#>  [5] LC_MONETARY=de_DE.UTF-8    LC_MESSAGES=en_US.UTF-8   
#>  [7] LC_PAPER=de_DE.UTF-8       LC_NAME=C                 
#>  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
#> [11] LC_MEASUREMENT=de_DE.UTF-8 LC_IDENTIFICATION=C       
#> 
#> time zone: Europe/Berlin
#> tzcode source: system (glibc)
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#> [1] effectsize_0.8.8
#> 
#> loaded via a namespace (and not attached):
#>  [1] vctrs_0.6.5       cli_3.6.2         knitr_1.46        rlang_1.1.3      
#>  [5] xfun_0.43         estimability_1.5  purrr_1.0.2       styler_1.10.3    
#>  [9] xtable_1.8-4      glue_1.7.0        htmltools_0.5.8.1 parameters_0.21.6
#> [13] datawizard_0.10.0 rmarkdown_2.26    grid_4.4.0        R.cache_0.16.0   
#> [17] evaluate_0.23     fastmap_1.1.1     mvtnorm_1.2-4     yaml_2.3.8       
#> [21] lifecycle_1.0.4   insight_0.19.10   compiler_4.4.0    coda_0.19-4.1    
#> [25] fs_1.6.3          emmeans_1.10.1    R.oo_1.26.0       lattice_0.22-5   
#> [29] R.utils_2.12.3    digest_0.6.35     reprex_2.1.0      magrittr_2.0.3   
#> [33] R.methodsS3_1.8.2 tools_4.4.0       withr_3.0.0       bayestestR_0.13.2

^{Created on 2024-05-24 with reprex v2.1.0}

And exactly this line:

https://github.com/easystats/effectsize/blob/43c6d0c163a56e35bbe32ed00b18491f6b864260/R/eta_squared-main.R#L836

[strengejacke]

Yes, there's no effects argument for the anova-method of model_parameters(), only effectsize_type.

[mattansb]

Sorry, why is this a bug report?

[chainsawriot]

It can generate a lot of warning messages for some methods (e.g. eta_squared_posterior() of a brms model, which is where I found this) when the above options are set. I know it is not default, but some R developers set them to default to prevent partial matching.

[mattansb]

I see.

The effect = argument actually shouldn't be partially matched, it is referring to cases where there are multiple effect types, but we only want the fixed effects (not relevant for aov or aovList models, where I thought it was just "absorbed" into ...). So although your PR fixes this issue, it causes another issue.

I will try to think of a solution.

[strengejacke]

Easiest is probably to rename in parameters.

[mattansb]

I have a solution - will post tomorrow.

[mattansb]

Fixed in #641

@chainsawriot if you come across any more of these, let us know.