pretty_names = TRUE do not fully work for "repeated rows"

[iago-pssjd]

In the table outputs of the next reprex, we should see Indirect Effect (ACME) in each file where ACME appears and the same for Direct Effect (ADE) with ADE, however, for each table they only appear in the first ACME/ADE rows. In particular I added the line attr(m, "pretty_names") <- c(attr(m20, "pretty_names"), attr(m70, "pretty_names")) to be sure the pretty_names are for all the involved rows.

library(mediation)
#> Loading required package: MASS
#> Loading required package: Matrix
#> Loading required package: mvtnorm
#> Loading required package: sandwich
#> mediation: Causal Mediation Analysis
#> Version: 4.5.0
library(parameters)

data(jobs)
# Allowing ``moderated mediation'' with respect to age
b.int <- lm(job_seek ~ treat*age + econ_hard + sex, data=jobs)
d.int <- lm(depress2 ~ treat*job_seek*age + econ_hard + sex, data=jobs)
contcont.age20 <- mediate(b.int, d.int, sims=50, treat="treat", mediator="job_seek",
            covariates = list(age = 20))
contcont.age70 <- mediate(b.int, d.int, sims=50, treat="treat", mediator="job_seek",
            covariates = list(age = 70))
m20 <- model_parameters(contcont.age20)
m70 <- model_parameters(contcont.age70)
m20[["age level"]] <-  20
m70[["age level"]] <-  70
m <- rbind(m20, m70)
attr(m, "pretty_names") <- c(attr(m20, "pretty_names"), attr(m70, "pretty_names"))
str(m)
#> Classes 'parameters_model', 'see_parameters_model' and 'data.frame': 20 obs. of  8 variables:
#>  $ Parameter: chr  "ACME (control)" "ACME (treated)" "ADE (control)" "ADE (treated)" ...
#>  $ Estimate : num  -0.0459 -0.0274 -0.0502 -0.0318 -0.0776 ...
#>  $ CI       : num  0.95 0.95 0.95 0.95 0.95 0.95 0.95 0.95 0.95 0.95 ...
#>  $ CI_low   : num  -0.1078 -0.0656 -0.1868 -0.1682 -0.2264 ...
#>  $ CI_high  : num  0.00768 0.00874 0.06707 0.10573 0.06468 ...
#>  $ p        : num  0.12 0.12 0.48 0.68 0.36 0.32 0.32 0.12 0.56 0.32 ...
#>  $ Component: chr  "control" "treated" "control" "treated" ...
#>  $ age level: num  20 20 20 20 20 20 20 20 20 20 ...
#>  - attr(*, "object_name")= chr "contcont.age20"
#>  - attr(*, "pretty_names")= Named chr [1:20] "Indirect Effect (ACME)" "Indirect Effect (ACME)" "Direct Effect (ADE)" "Direct Effect (ADE)" ...
#>   ..- attr(*, "names")= chr [1:20] "ACME" "ACME" "ADE" "ADE" ...
#>  - attr(*, "ci")= num 0.95
#>  - attr(*, "verbose")= logi TRUE
#>  - attr(*, "exponentiate")= logi FALSE
#>  - attr(*, "ordinal_model")= logi FALSE
#>  - attr(*, "linear_model")= logi FALSE
#>  - attr(*, "mixed_model")= logi FALSE
#>  - attr(*, "model_class")= chr "mediate"
#>  - attr(*, "bootstrap")= logi FALSE
#>  - attr(*, "iterations")= num 1000
#>  - attr(*, "robust_vcov")= logi FALSE
#>  - attr(*, "ignore_group")= logi TRUE
#>  - attr(*, "ran_pars")= logi TRUE
#>  - attr(*, "show_summary")= logi FALSE
#>  - attr(*, "weighted_nobs")= num 899
#>  - attr(*, "coefficient_name")= chr "Coefficient"
#>  - attr(*, "zi_coefficient_name")= chr "Log-Odds"
#>  - attr(*, "digits")= num 2
#>  - attr(*, "ci_digits")= num 2
#>  - attr(*, "p_digits")= num 3
#>  - attr(*, "footer_digits")= num 3
print_md(m)

Parameter	Estimate	95% CI	p	age level
Indirect Effect (ACME)	-0.05	(-0.11, 7.68e-03)	0.120	20.00
Direct Effect (ADE)	-0.05	(-0.19, 0.07)	0.480	20.00
Prop. Mediated	0.33	(-4.73, 2.78)	0.320	20.00
ACME	0.03	(-0.06, 0.13)	0.720	70.00
ADE	-0.08	(-0.34, 0.18)	0.600	70.00
Prop. Mediated	0.04	(-3.08, 5.61)	0.880	70.00

control

Parameter	Estimate	95% CI	p	age level
Indirect Effect (ACME)	-0.03	(-0.07, 8.74e-03)	0.120	20.00
Direct Effect (ADE)	-0.03	(-0.17, 0.11)	0.680	20.00
Prop. Mediated	0.25	(-2.45, 1.33)	0.320	20.00
ACME	0.02	(-0.05, 0.10)	0.600	70.00
ADE	-0.09	(-0.34, 0.16)	0.640	70.00
Prop. Mediated	0.05	(-1.54, 3.22)	0.920	70.00

treated

Parameter	Estimate	95% CI	p	age level
Indirect Effect (ACME)	-0.04	(-0.09, 8.21e-03)	0.120	20.00
Direct Effect (ADE)	-0.04	(-0.18, 0.08)	0.560	20.00
Prop. Mediated	0.29	(-3.55, 1.88)	0.320	20.00
ACME	0.02	(-0.05, 0.11)	0.600	70.00
ADE	-0.08	(-0.34, 0.17)	0.680	70.00
Prop. Mediated	0.04	(-2.04, 4.03)	0.920	70.00

average

Parameter	Estimate	95% CI	p	age level
Total Effect	-0.08	(-0.23, 0.06)	0.360	20.00
Total Effect	-0.06	(-0.38, 0.22)	0.800	70.00

Total Effect

sessionInfo()
#> R version 4.2.0 (2022-04-22)
#> Platform: x86_64-pc-linux-gnu (64-bit)
#> Running under: Debian GNU/Linux 11 (bullseye)
#> 
#> Matrix products: default
#> BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
#> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.13.so
#> 
#> locale:
#>  [1] LC_CTYPE=en_GB.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=en_GB.UTF-8        LC_COLLATE=en_GB.UTF-8    
#>  [5] LC_MONETARY=en_GB.UTF-8    LC_MESSAGES=en_GB.UTF-8   
#>  [7] LC_PAPER=en_GB.UTF-8       LC_NAME=C                 
#>  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
#> [11] LC_MEASUREMENT=en_GB.UTF-8 LC_IDENTIFICATION=C       
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#> [1] parameters_0.18.1 mediation_4.5.0   sandwich_3.0-1    mvtnorm_1.1-3    
#> [5] Matrix_1.4-1      MASS_7.3-57      
#> 
#> loaded via a namespace (and not attached):
#>  [1] splines_4.2.0       R.utils_2.11.0      datawizard_0.4.1   
#>  [4] Formula_1.2-4       assertthat_0.2.1    latticeExtra_0.6-29
#>  [7] yaml_2.3.5          bayestestR_0.12.1   pillar_1.7.0       
#> [10] backports_1.4.1     lattice_0.20-45     glue_1.6.2         
#> [13] digest_0.6.29       RColorBrewer_1.1-3  checkmate_2.1.0    
#> [16] minqa_1.2.4         colorspace_2.0-3    htmltools_0.5.2    
#> [19] R.oo_1.24.0         lpSolve_5.6.15      pkgconfig_2.0.3    
#> [22] xtable_1.8-4        purrr_0.3.4         scales_1.2.0       
#> [25] jpeg_0.1-9          lme4_1.1-29         emmeans_1.7.4-1    
#> [28] tibble_3.1.7        htmlTable_2.4.0     styler_1.7.0       
#> [31] generics_0.1.2      ggplot2_3.3.6       ellipsis_0.3.2     
#> [34] TH.data_1.1-1       withr_2.5.0         nnet_7.3-17        
#> [37] cli_3.3.0           survival_3.3-1      magrittr_2.0.3     
#> [40] crayon_1.5.1        estimability_1.3    evaluate_0.15      
#> [43] R.methodsS3_1.8.1   fs_1.5.2            fansi_1.0.3        
#> [46] R.cache_0.15.0      nlme_3.1-157        foreign_0.8-82     
#> [49] tools_4.2.0         data.table_1.14.2   multcomp_1.4-19    
#> [52] lifecycle_1.0.1     stringr_1.4.0       munsell_0.5.0      
#> [55] reprex_2.0.1        cluster_2.1.3       compiler_4.2.0     
#> [58] rlang_1.0.2         grid_4.2.0          nloptr_2.0.3       
#> [61] rstudioapi_0.13     htmlwidgets_1.5.4   base64enc_0.1-3    
#> [64] rmarkdown_2.14      boot_1.3-28         codetools_0.2-18   
#> [67] gtable_0.3.0        DBI_1.1.2           R6_2.5.1           
#> [70] gridExtra_2.3       zoo_1.8-10          knitr_1.39         
#> [73] dplyr_1.0.9         fastmap_1.1.0       utf8_1.2.2         
#> [76] Hmisc_4.7-0         insight_0.17.1      stringi_1.7.6      
#> [79] Rcpp_1.0.8.3        vctrs_0.4.1         rpart_4.1.16       
#> [82] png_0.1-7           coda_0.19-4         tidyselect_1.1.2   
#> [85] xfun_0.31

^{Created on 2022-06-02 by the reprex package (v2.0.1)}

[iago-pssjd]

A question on pretty_names. I see they can be used for the model_parameters of certain objects/classes, like regressions, while cannot be used with the model_parameters of others, like htest.

Indeed, I see that

model <- t.test(iris$Sepal.Width, iris$Sepal.Length)
mmp <- model_parameters(model)
attr(mmp, "pretty_names") <- "mpg"
mmp |> print_md()
Error in `$<-.data.frame`(`*tmp*`, "Parameter", value = character(0)) : 
  replacement has 0 rows, data has 1

as the model_parameters output does not include Parameter, however

model <- t.test(mtcars$mpg ~ mtcars$vs)
mmp <- model_parameters(model)
attr(mmp, "pretty_names") <- "mpg"
mmp |> print_md()


Table: Welch Two Sample t-test

|Parameter  |     Group | mtcars$vs = 0 | mtcars$vs = 1 | Difference |          95% CI | t(22.72) |      p |
|:----------|:---------:|:-------------:|:-------------:|:----------:|:---------------:|:--------:|:------:|
|mtcars$mpg | mtcars$vs |         16.62 |         24.56 |      -7.94 | (-11.46, -4.42) |    -4.67 | < .001 |

has no effect.

Then, my question is if is it some place where it is specified for which models pretty_names is used.