R/blindDataRegression_ScramblePredictors.R

In rjbderooij/blindData_Git: Blind Data

#' Suitable for regression:
#' This function reassigns the predictor names at random.
#' @param df_original # original dataframe
#' @param y # name of dependent variable
#' @param predictors # name of predictors, for example c("expert", "conflict)
#' @param update_labels # if TRUE update labels to BLIND_[abbrevation]_[name]
#' @keywords Scramble predictors, Regression
#' @importFrom stats lm runif resid
#' @importFrom tidyselect all_of
#'


blindDataRegression_ScramblePredictors <- function(df_original,
                                                   y,
                                                   predictors,
                                                   update_labels = TRUE){


  ### Input:
  # df_original = original dataset (i.e., unblinded)
  # y = dependent variable
  # predictors = predictor variables (a.k.a. independent variables)
  # update_labels = TRUE updates labels from [name] to blind_AB_[name]
  #                 FALSE retains original labels
  #                 defaults to true to clearly indicate a changes to the dataset



  # Step 1: Check that there is more than 1 predictor (else no scrambling)
  if(length(predictors) == 1){
    print("Cannot scramble predictor variables if there is only one predictor")
    stop()
  }

  # Step 2: Reorder original dataframe with Y in first column
  df_original <- df_original[,c(y, predictors)]

  # Step 3: Create new dataset of predictors without outcome variable
  df_blindScramblePredictors <- df_original[,predictors]

  # Step 4: Sample without replacmeent from the index-number of columns
  # (e.g., 1,2,3,4) a new order of columns (e.g., 4,2,3,1)
  df_blindScramblePredictors =
    df_blindScramblePredictors[,
                               sample(
                                 ncol(df_blindScramblePredictors),
                                 replace = F)
                               ]

# Step 5: cbind the original value of Y to the blinded predictor df
df_blindScramblePredictors = data.frame(df_original[, y],
                                        df_blindScramblePredictors)

# Step 6: Replace names of random-column df with original variable names
# (that still haave the original ordering)
# NOTE that y was on same position [1]
if(update_labels){
  names(df_blindScramblePredictors) =
    c(y, # Y
      paste0("BLIND_SP_", names(df_original)[-1])) # X (-1 was Y)
}

# Return
return(df_blindScramblePredictors)
} # End blindDataRegression_ScramblePredictors