Moonbook significant digit version function to run each variable individually, determining significant digits then join them into a single file at the end
e_table1_summaries(
dat_tab1 = dat_sub,
list_var_col_names = NULL,
list_var_col_labels = NULL,
list_var_row_names = NULL,
list_var_row_labels = NULL,
fn_root = "tab1_vars",
fn_all = NULL,
label_width = 40,
sw_verbose = FALSE,
moonbook_max_ylev = 2,
moonbook_digits = c(factor = 1, numeric = 3),
moonbook_method = c(1, 2, 3)[2]
)Dataset for table.
Column variable name.
Column variable label.
Variable names to loop over.
Variable labels to label table.
Filename for temporary tables.
Filename for output table, defaults to paste0(fn_root, "_", "all", ".csv").
Width to wrap column and row variable labels, does not affect category labels.
Print status to console during execution?
An integer indicating the maximum number of levels of grouping variable ('y'). If a colummn have unique values less than max.ylev it is treated as a categorical variable. Default value is 2. (?moonBook::mytable_sub)
An integer indicating the number of decimal places (round) or significant digits to be used. Default value is 3. (?moonBook::mytable_sub)
An integer indicating methods for continuous variables. Possible values in methods are: 1 = forces analysis as normal-distributed; 2 = forces analysis as continuous non-normal; 3 = performs a Shapiro-Wilk test to decide between normal or non-normal. (?moonBook::mytable_sub)
NULL, invisibly
Uses moonBook::mytable; default is CatMethod = 2: Numeric summaries are Median and IQR with kruskal.test, Categorical summaries are proportions with chisq.test with continuity correction.
if (FALSE) {
## For Rmd code chunk attractive html table, code chunk option ```{r, results='asis'}
# specify tows to summarize and columns to summarize by
list_var_col_names <- c("cyl", "am")
list_var_row_names <- c("mpg", "cyl", "disp", "hp", "drat", "wt"
, "qsec", "vs", "am", "gear", "carb")
# assigning to new dataset since you may want to filter or subset in some way,
# for example, only baseline measurements in a longitudinal analysis
dat_tab1 <-
dat_mtcars_e %>%
#dplyr::filter(EVENT_ID == "BL") %>%
dplyr::select(all_of(c(list_var_col_names, list_var_row_names)))
# label variables
list_var_col_labels <- labelled::var_label(dat_tab1[, list_var_col_names]) %>% unlist()
list_var_row_labels <- labelled::var_label(dat_tab1[, list_var_row_names]) %>% unlist()
# This loop creates a table summarized by each variable,
# but can also specify multiple columns to summarize conditional on multiple columns.
for (i_col in seq_along(list_var_col_names)) {
# filename root for each separate variable (will write, then read to compile into one)
fn_root <- paste0("tab1_vars_", list_var_col_names[i_col], "_summary")
# run function (additional options are available)
e_table1_summaries(
dat_tab1 = dat_tab1
, list_var_col_names = list_var_col_names[i_col]
, list_var_col_labels = list_var_col_labels[i_col]
, list_var_row_names = list_var_row_names
, list_var_row_labels = list_var_row_labels
, fn_root = fn_root
, sw_verbose = TRUE
)
# read final table
tab1_temp <- utils::read.csv(paste0(fn_root, "_", "all", ".csv"))
# display in Rmd file
knitr::kable(tab1_temp)
}
## For totals, helpful when missing data because has sample size for each row.
# filename root for each separate variable (will write, then read to compile into one)
fn_root <- paste0("tab1_vars_", "total", "_summary")
# run function (additional options are available)
e_table1_summaries(
dat_tab1 = dat_tab1
, list_var_col_names = NULL
, list_var_col_labels = "total"
, list_var_row_names = list_var_row_names
, list_var_row_labels = list_var_row_labels
, fn_root = fn_root
, sw_verbose = TRUE
)
# read final table
tab1_temp <- utils::read.csv(paste0(fn_root, "_", "all", ".csv"))
# display in Rmd file
knitr::kable(tab1_temp)
}