R/check_miss.R

In BERENZ/MiscFuns: Package contains functions that I use during my analysis

#' Function to count missing data in input dataset
#' 
#' Function that returns information bout missing values, 
#' characters that have 0 length as well as number of 
#' unique values and numer of observations
#' 
#' @name check_miss
#' @param data inputdataset.
#' @param ... arguments passing variabele names to dplyr.
#' @return data.frame with number of NA (N_miss), number of 0 length characters (N_0_char), number of unique values (N_unique) and number of observations (N).
#' @examples
#' check_miss(mtcars)

check_miss <- function(data,...) {
  p <- match.call(expand.dots=FALSE)
  if (length(p) <= 2) {
    tab <- data %>% 
      summarise_each(funs(.na_sum,
                          .nchar_sum,
                          .n_unique,
                          n())) %>%
      matrix(ncol=4,
             dimnames=list(
               names(data),
               c('N_miss','N_0_char','N_unique','N')
             )) %>%
      as.data.frame()
  } else {
    v_name <- data %>% 
                select(...) %>%
                names()
    
    tab <- data %>% 
      select(...) %>%
      summarise_each(funs(.na_sum,
                          .nchar_sum,
                          .n_unique,
                          n())) %>%
      
      matrix(ncol=4,
             dimnames=list(
               v_name,
               c('N_miss','N_0_char','N_unique','N')
             )) %>%
      as.data.frame()
  }
  return(tab)
}

.na_sum <- function(x) {
  return(sum(is.na(x)))
}

.nchar_sum <- function(x) {
  return(sum(str_length(x)==0))
}


.n_unique <- function(x) {
  return(length(unique(x)))
}