Validation checks

suppressPackageStartupMessages({
  suppressWarnings({
    library(assertr)
    library(assertive)
    library(assertthat)
    library(palmerpenguins)
    library(janitor)
  })
})

Validation checks

Data validation is performed using the assertr package, with extra checks provided by the assertive package. The assertthat package can also be used as an alternative to stopifnot(). All examples perform checks on the palmerpenguins dataset and return TRUE if validation check passes and stops with an error if the validation check fails

The file path to the validation checks markdown document exists

assert_that(file.exists("validation_checks.Rmd"), 
            msg = "Path to validation check markdown document is UNSUCCESSFUL")
#> [1] TRUE

The penguin dataset is not empty

see_if(not_empty(penguins), 
       msg = "Penguins dataset is EMPTY")
#> [1] TRUE

The penguins dataset contains 344 rows

penguins %>% 
  verify(nrow(.) == 344,
         success_fun = success_logical, error_fun = error_stop)
#> [1] TRUE

The penguins dataset contains eight named columns. This uses a functional sequence to store verify functions for each column name and number of columns.

check_penguins_column_names <- . %>% 
  chain_start() %>% 
  verify(has_all_names("species")) %>%
  verify(has_all_names("island")) %>%
  verify(has_all_names("bill_length_mm")) %>%
  verify(has_all_names("bill_depth_mm")) %>%
  verify(has_all_names("flipper_length_mm")) %>%
  verify(has_all_names("body_mass_g")) %>%
  verify(has_all_names("sex")) %>%
  verify(has_all_names("year")) %>%
  verify(ncol(.) == 8) %>% 
  chain_end(success_fun = success_logical, error_fun = error_stop)
  
penguins %>% 
  check_penguins_column_names
#> [1] TRUE

Penguins raw dataset date_egg column is a date type

penguins_raw %>% 
  clean_names() %>% 
  assert(assertive::is_date, date_egg,
         success_fun = success_logical, error_fun = error_stop)
#> [1] TRUE

Combination of sample number and species name in penguins raw dataset are unique

penguins_raw %>% 
  clean_names() %>%  
  assert_rows(col_concat, is_uniq, sample_number, species,
              success_fun = success_logical, error_fun = error_stop)
#> [1] TRUE

All penguins are assigned to a species and an island (they is no missing data in the species or island columns)

penguins %>% 
  assert(not_na, species, island,  
         success_fun = success_logical, error_fun = error_stop)
#> [1] TRUE

All penguins are either male or female (set parameters allow.na = TRUE to exclude NAs and inverse = TRUE to check if all penguins do not contain either male or female)

penguins %>% 
  assert(in_set("male", "female"), sex,
         success_fun = success_logical, error_fun = error_stop)
#> [1] TRUE

All penguins weight are positive values

penguins %>% 
  dplyr::filter(!is.na(body_mass_g)) %>% 
  assert(assertive::assert_all_are_positive, body_mass_g,
         success_fun = success_logical, error_fun = error_stop)
#> [1] TRUE

All penguins weigh between 2700 and 6300 grams, inclusively

penguins %>% 
  assert(within_bounds(2700, 6300), body_mass_g,
         success_fun = success_logical, error_fun = error_stop)
#> [1] TRUE

Data exploration

Find the rows in the penguins dataset that are missing all the four bill, flipper and body mass measurements. Row numbers are displayed in the index column

penguins %>% 
  assert_rows(num_row_NAs, function(.x){.x != 4}, bill_length_mm:body_mass_g,
               success_fun = success_continue, error_fun = just_warn)
#> Data frame row reduction 'num_row_NAs' violates predicate 'function(.x) {
#>     .x != 4
#> }' 2 times
#>          verb    redux_fn                      predicate
#> 1 assert_rows num_row_NAs function(.x) {\n    .x != 4\n}
#> 2 assert_rows num_row_NAs function(.x) {\n    .x != 4\n}
#>                        column index value
#> 1 ~bill_length_mm:body_mass_g     4     4
#> 2 ~bill_length_mm:body_mass_g   272     4
#> Warning: assertr encountered errors
#> # A tibble: 344 x 8
#>    species island    bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
#>    <fct>   <fct>              <dbl>         <dbl>             <int>       <int>
#>  1 Adelie  Torgersen           39.1          18.7               181        3750
#>  2 Adelie  Torgersen           39.5          17.4               186        3800
#>  3 Adelie  Torgersen           40.3          18                 195        3250
#>  4 Adelie  Torgersen           NA            NA                  NA          NA
#>  5 Adelie  Torgersen           36.7          19.3               193        3450
#>  6 Adelie  Torgersen           39.3          20.6               190        3650
#>  7 Adelie  Torgersen           38.9          17.8               181        3625
#>  8 Adelie  Torgersen           39.2          19.6               195        4675
#>  9 Adelie  Torgersen           34.1          18.1               193        3475
#> 10 Adelie  Torgersen           42            20.2               190        4250
#> # ... with 334 more rows, and 2 more variables: sex <fct>, year <int>