impute_mean: Fills missing values with subgroup mean value

In STAPM/hseclean: Health Survey Data Wrangling

Fills missing values with subgroup mean value

Description

Replaces any values < 0 with NA, calculates the subgroup mean, then replaces missing values with the subgroup mean.

Usage

impute_mean(
  data,
  var_names,
  remove_zeros = FALSE,
  strat_vars = c("year", "sex", "imd_quintile", "age_cat")
)

Arguments

data	Data table - the health survey data
var_names	Character vector - the variable names to be imputed (numeric variables only)
remove_zeros	Logical - should zeros be treated as missing data
strat_vars	Character vector - the variables by which to stratify the subgroup means

Details

If not all NAs can be imputed with the fine scale starting amount of stratification, imputation is attempted again, removing the stratification variable specified last.

Value

Returns an updated version of data in which the variables specified have had their missing values imputed with the subgroup means.

Examples

## Not run: 

data <- read_2001()
data <- clean_age(data)
data <- clean_demographic(data)
data <- impute_mean(data, var_names = c("d7many"))


## End(Not run)

STAPM/hseclean documentation built on June 9, 2025, 4:50 a.m.