R/prepare_table_means.R

In christian-gische/causalSEM: Analyze Causal Effects Using Interventional Distributions

## Changelog:
# MA 0.0.2 2022-03-15: added causal.var.type and inserted NAs for interventional
##                     variables
# MA 0.0.1 2022-02-17: initial programming

## Documentation
#' @title Prepare Tables with Information About Interventional Means
#' @description Internal helper function that creates a formatted table
#' containing various information about the interventional means. Used in
#' print.causalSEM and summary.causalSEM
#' @param x an object of class "causalSEM", usually, a result of a call to
#' intervention_effect.
#' @param digit integer indicating the number of decimal places (round) or
#' significant digits (signif) to be used.
#' @return \code{prepare_table_means} returns a character vector.
#' @references Gische, C., Voelkle, M.C. (2021) Beyond the mean: a flexible
#' framework for studying causal effects using linear models. Psychometrika
#' (advanced online publication). https://doi.org/10.1007/s11336-021-09811-z

## Function definition
prepare_table_means <- function(x, digits = 3){

  # function name
  fun_name <- "prepare_table_means"

  # function version
  fun_version <- "0.0.2 2022-03-15"

  # function name+version
  fun_name_version <- paste0(fun_name, " (", fun_version, ")")

  # console output
  if(x$control$verbose >= 2) {
    cat(paste0("start of function ",fun_name_version, " ", Sys.time(), "\n" ))
  }

  # Prepare first line with names and levels of the interventional variables
  info_variables <- c()
  for (i in 1:x$info_interventions$n_intervention) {
    info_variables[length(info_variables) + 1] <- x$info_interventions$intervention_name[i]
    info_variables[length(info_variables) + 1] <- "="
    info_variables[length(info_variables) + 1] <- x$info_interventions$intervention_level[i]
    if (i < x$info_interventions$n_intervention) {
      info_variables[length(info_variables) + 1] <- ", "
    }
  }
  info_variables <- paste0(info_variables, collapse = "")


  # Collapse infomation about the interventional variables into a single string
  info_variables <- paste0("Intervention: do(", info_variables, ")", collapse = "")

  # Determine causal variable type
  interventions <- rowSums(x$constant_matrices$select_intervention)

  causal_var_type <- ifelse(test = interventions == 1,
                            yes = "interv.",
                            no = "pre/post interv.")

  # data.frame with means
  df_means <- data.frame(
    x$tables$interventional_means$Variable,
    round(x$tables$interventional_means[, 2:3], digits = digits),
    paste0(
      "[",
      round(x$tables$interventional_means$CI_lower, digits = digits),
      ",",
      round(x$tables$interventional_means$CI_upper, digits = digits),
      "]"),
    causal_var_type)

  # Add row with row names
  df_means <- rbind(
    c("Variable", "Est.", "Std. Err.", "95% CI", "Causal. Var. Type"),
    df_means
  )

  # Delete column names
  colnames(df_means)  <- ""

  # Insert NAs for interventions
  df_means[df_means[, 5] == "interv.", c(3, 4)] <- "NA"

  # Transform data.farme into strings
  df_means <- utils::capture.output(print(df_means, row.names = FALSE))
  df_means <- df_means[-1]

  # Return output
  df_means

}