R/epi_clean_fct_to_na.R

In AntonioJBT/episcout: Quickly Clean, Explore and Visualise Large Epidemiological Datasets

# Loop through to recode NAs but preserve attributes:
# Function to convert specified factor values to NA
epi_clean_fct_to_na <- function(factor_var, values_to_na) {
    # Ensure the input is a factor:
    if (!is.factor(factor_var)) {
        stop("The input variable is not a factor.")
    }

    # Convert specified values to NA:
    levels_to_na <- levels(factor_var) %in% values_to_na
    factor_var[as.character(factor_var) %in% values_to_na] <- NA

    # Preserve the factor structure
    levels(factor_var)[levels_to_na] <- NA

    return(factor_var)
}

# Apply the function to the factor variable in the data frame
vec_test <- epi_clean_fct_to_na(data_f$SEXO, NA_labels)
summary(vec_test)

# Specify columns to modify:
cols_mod <- colnames(df_ord)[7:40]
str(df_ord)

for (i in cols_mod) {
    df_ord[[i]] <- epi_clean_fct_to_na(df_ord[[i]], NA_labels)
}