Skip to contents

Associate multiple variables with columns

Source: R/colby_constructors.R
split_cols_by_multivar.Rd

In some cases, the variable to be ultimately analyzed is most naturally defined on a column, not a row, basis. When we need columns to reflect different variables entirely, rather than different levels of a single variable, we use split_cols_by_multivar.

Usage 

split_cols_by_multivar(
  lyt,
  vars,
  split_fun = NULL,
  varlabels = vars,
  varnames = NULL,
  nested = TRUE,
  extra_args = list(),
  show_colcounts = FALSE,
  colcount_format = NULL
)

Arguments

lyt

(PreDataTableLayouts)
layout object pre-data used for tabulation.

vars

(character)
vector of variable names.

split_fun

(function or NULL)
custom splitting function. See custom_split_funs.

varlabels

(character)
vector of labels for vars.

varnames

(character)
vector of names for vars which will appear in pathing. When vars are all unique this will be the variable names. If not, these will be variable names with suffixes as necessary to enforce uniqueness.

nested

(logical)
whether this layout instruction should be applied within the existing layout structure if possible (TRUE, the default) or as a new top-level element (FALSE). Ignored if it would nest a split underneath analyses, which is not allowed.

extra_args

(list)
extra arguments to be passed to the tabulation function. Element position in the list corresponds to the children of this split. Named elements in the child-specific lists are ignored if they do not match a formal argument of the tabulation function.

show_colcounts

(logical(1))
should column counts be displayed at the level facets created by this split. Defaults to FALSE.

colcount_format

(character(1))
if show_colcounts is TRUE, the format which should be used to display column counts for facets generated by this split. Defaults to "(N=xx)".

Value

A PreDataTableLayouts object suitable for passing to further layouting functions, and to build_table().

See also

analyze_colvars()

Author

Gabriel Becker

Examples 

library(dplyr)

ANL <- DM %>% mutate(value = rnorm(n()), pctdiff = runif(n()))

## toy example where we take the mean of the first variable and the
## count of >.5 for the second.
colfuns <- list(
  function(x) in_rows(mean = mean(x), .formats = "xx.x"),
  function(x) in_rows("# x > 5" = sum(x > .5), .formats = "xx")
)

lyt <- basic_table() %>%
  split_cols_by("ARM") %>%
  split_cols_by_multivar(c("value", "pctdiff")) %>%
  split_rows_by("RACE",
    split_label = "ethnicity",
    split_fun = drop_split_levels
  ) %>%
  summarize_row_groups() %>%
  analyze_colvars(afun = colfuns)
lyt
#> A Pre-data Table Layout
#> 
#> Column-Split Structure:
#> ARM (lvls) -> value:pctdiff (vars) 
#> 
#> Row-Split Structure:
#> RACE (lvls) -> NA (** col-var analysis **) 
#> 

tbl <- build_table(lyt, ANL)
tbl
#>                                    A: Drug X                B: Placebo              C: Combination     
#>                               value       pctdiff       value       pctdiff       value       pctdiff  
#> ———————————————————————————————————————————————————————————————————————————————————————————————————————
#> ASIAN                       79 (65.3%)   79 (65.3%)   68 (64.2%)   68 (64.2%)   84 (65.1%)   84 (65.1%)
#>   mean                         0.0           39          -0.1          31          -0.0          33    
#> BLACK OR AFRICAN AMERICAN   28 (23.1%)   28 (23.1%)   24 (22.6%)   24 (22.6%)   27 (20.9%)   27 (20.9%)
#>   mean                         0.2           11          0.1           10          -0.3          10    
#> WHITE                       14 (11.6%)   14 (11.6%)   14 (13.2%)   14 (13.2%)   18 (14.0%)   18 (14.0%)
#>   mean                         0.3           8           0.4           7           -0.2          9