Skip to contents

Summarize Dataset

Source: R/wid_make_summaries.R
wid_make_summaries.Rd

Calculate summaries of the reformatted dataset.

Usage

wid_make_summaries(reformat_dir, manifest_dir = NULL, by_sex = FALSE,
  cores = parallel::detectCores() - 5)

Arguments

reformat_dir

Directory containing the reformatted dataset.

manifest_dir

Directory to write summaries to.

by_sex

Logical; if TRUE, will further break down summaries by sex.

cores

Number of CPU cores to use during processing.

Value

A list with entries for each variable:

  • type Value type.

  • missing Number of missing values across the entire dataset.

  • value_summary Summary of the values, depending on type. For categorical variables, this will be observation counts for each level. For continuous variables, this will be the min, mean, standard deviation, median, and max. For identifier-like variables (character vectors with many unique values), this will be the alphabetically first and last value, and the number of unique values.

  • by_survey Value summaries within each survey and year which have any observations.

Examples

if (FALSE) { # \dontrun{

  wid_make_summaries("../gender_growth_gap", "explore/public", by_sex = TRUE)
} # }