R/dataset_stats.R

In VanRijnLab/SlimStampeRData: Analyze SlimStampen App Data

#' Descriptive specifications about the data set
#'
#' \code{dataset_stats} describes several specifications regarding the data set,
#' these are: Number of participants, number of unique facts, number of lessons,
#' number of unique user-fact combinations, participants per lesson, facts per
#' lesson, unique user-fact combinations per lesson, average RT per lesson,
#' average accuracy per lesson, lessons per participant and sessions per
#' participant.
#'
#' The number of participants, number of unique facts, number of lessons,
#' participants per lesson, average RT per lesson and average accuracy per
#' lesson are always printed in the console. The data frame with the sessions
#' and lessons per participants is not printed, since the amount of participants
#' can be substantial. All information can be found in $general, $lessons and
#' $participants.
#'
#' @param data A data frame
#'
#' @return Three data frames ($general, $lessons, $participants)
#' @export
#'
dataset_stats <- function(data) {
  if(missing(data)){
    stop("No data is provided")
  }
  missingcol <- missing_columns_check(data, c("sessionId", "factId", "userId", "reactionTime", "lessonId", "correct", "lessonTitle", "presentationStartTime"))
  if(length(missingcol) > 0){
    stop("No ", missingcol[[1]] ," column is provided in the data")
  }

  if(-1 %in% data$factId){
    data <- resetremoval(data)
    cat("- There are resets present in the data. Reset data is excluded in this function. - \n")
  }

  participants = unique(data$userId)
  nrparticipants = length(participants)

  nrfacts = length(unique(data$factId))

  lessons = unique(data$lessonTitle)
  nrlessons = length(lessons)

  nrfactsuser = dplyr::n_distinct(paste(data$factId, data$userId))

  partPerLesson = integer(0)
  factsPerLesson = integer(0)
  userfactsPerLesson = integer(0)
  RTPerLesson = integer(0)
  AccPerLesson = integer(0)
  for (i in seq_along(lessons)) {
    dat1 <- dplyr::filter(data, lessonTitle == lessons[i])
    partPerLesson[i] <- length(unique(dat1$userId))
    factsPerLesson[i] <- length(unique(dat1$factId))
    userfactsPerLesson[i] <- dplyr::n_distinct(paste(dat1$factId, dat1$userId))
    RTPerLesson[i] <- mean(dat1$reactionTime, na.rm=TRUE)
    AccPerLesson[i] <- mean(dat1$correct, na.rm=TRUE)
  }
  sessionPerPart = integer(0)
  lessonPerPart = integer(0)
  factsPerPart = integer(0)
  for (i in seq_along(participants)) {
    dat1 <- dplyr::filter(data, userId == participants[i])
    sessionPerPart[i] <- length(unique(dat1$sessionId))
    lessonPerPart[i] <- length(unique(dat1$lessonId))
    factsPerPart[i] <- length(unique(dat1$factId))
  }
  out <- list()
  out$general <- data.frame(descriptives = c("Total unique participants", "Total facts", "Total lessons", "Fact-user combinations"),
                   values = c(nrparticipants, nrfacts, nrlessons, nrfactsuser))
  out$lessons <- data.frame(lessons = lessons,
                    participants = partPerLesson, facts = factsPerLesson, user_facts = userfactsPerLesson, RT = RTPerLesson, accuracy = AccPerLesson)
  out$participants <- data.frame(participant = participants,
                    sessions = sessionPerPart, lessons = lessonPerPart, facts = factsPerPart)
  cat("General stats ($general): \n")
  print(out$general)
  cat("\n Information per lesson ($lessons): \n")
  print(out$lessons)
  cat("\n The information per participant can be found in $participants")

  invisible(out)
}