In ddsjoberg/bstfun: Functions for the MSK Biostatistics Department

knitr::opts_chunk$set(echo = TRUE)

set.seed({{format(Sys.Date(), format = "%Y%m%d")}})

Raw Data Import

library(biostatR)
library(tidyverse)
library(here)
library(styler)
library(labelled)
library(xlsx)

{{(ifelse(tolower(Sys.info()[["user"]]) %in% eval(as.list(bstfun::theme_gtsummary_msk)$name), glue::glue("
# Set custom gtsummary theme if exists in the bstfun::theme_gtsummary_msk
bstfun::theme_gtsummary_msk('{tolower(Sys.info()[['user']])}')
"), ""))}}
# example code for importing excel file
df_raw_data <-
  readxl::read_excel(
    path = here_data("Raw Data from PI.xlsx")
  ) %>%
  janitor::clean_names(case = "all_caps") %>% # uppercase col names
  mutate(across(where(lubridate::is.POSIXt), lubridate::as_date)) %>% # use lubridates
  mutate_if(is.logical, as.numeric)

Create Master Analytic Data Set

Check Variables

Label analytic dataset

# Import data dictionary
df_dict <-
  readxl::read_excel(
    path = here::here("secure_data", "_data-dictionary.xlsx")
  ) 

# Only include variables in the datasets that I want to label
vars_tolabel <- colnames(df_drvd)
df_dict_tolabel <- df_dict %>% 
  filter(variable %in% vars_tolabel)

# Create label lists
label_list <- as.list(df_dict_tolabel$label)
names(label_list) <- as.list(df_dict_tolabel$variable)

# Apply labels to analytic datasets
labelled::var_label(df_drvd) <- label_list

Save Analytic Data

# overview of master analytic dataset
skimr::skim(df_master)

# saving master analytic data set
saveRDS(
  object = df_drvd,
  file = here::here(path_data, "df_master.Rds")
)