Skip to contents

Get list of unique raw variables for a database/cycle

Source: R/mockdata_helpers.R
get_raw_variables.Rd

Returns a data frame of unique raw (source) variables that should be generated for a specific database/cycle. This is the correct approach for generating mock data, as we want to create the raw source data, not the harmonized variables.

Usage

get_raw_variables(cycle, variables, variable_details, include_derived = FALSE)

Arguments

cycle

Character string specifying the database/cycle (e.g., "cycle1", "cycle1_meds" for CHMS; "cchs2001" for CCHS).

variables

Data frame from variables.csv containing variable metadata.

variable_details

Data frame from variable_details.csv containing detailed specifications.

include_derived

Logical. Should derived variables be included? Default is FALSE (since derived variables are computed from other variables, not in raw data).

Value

Data frame with columns:

  • variable_raw - Raw source variable name (unique)

  • variableType - "Categorical" or "Continuous"

  • harmonized_vars - Comma-separated list of harmonized variables that use this raw variable

  • n_harmonized - Count of how many harmonized variables use this raw variable

Details

This function:

  1. Gets all variables available in the database/cycle using get_cycle_variables

  2. Extracts unique raw variable names

  3. Groups harmonized variables by their raw source

  4. Returns one row per unique raw variable

This is the correct approach because:

  • Mock data should represent raw source data (before harmonization)

  • Each raw variable should appear exactly once

  • Multiple harmonized variables can derive from the same raw variable

See also

get_cycle_variables, parse_variable_start

Other mockdata-helpers: apply_missing_codes(), apply_rtype_defaults(), extract_distribution_params(), extract_proportions(), generate_garbage_values(), get_cycle_variables(), get_variable_details(), has_garbage(), make_garbage(), sample_with_proportions()

Examples

if (FALSE) { # \dontrun{
# Load metadata
variables <- read.csv("inst/extdata/variables.csv")
variable_details <- read.csv("inst/extdata/variable-details.csv")

# CHMS example
raw_vars <- get_raw_variables("cycle1", variables, variable_details)

# CCHS example
raw_vars_cchs <- get_raw_variables("cchs2001", variables, variable_details)

# Generate mock data from raw variables
for (i in 1:nrow(raw_vars)) {
  var_raw <- raw_vars$variable_raw[i]
  var_type <- raw_vars$variableType[i]
  # Generate the raw variable...
}
} # }