Remove outliers from trait data (IQR rule)

Finds outliers with find_outliers() and sets those values to NA, so the plots stay in the design but the values are left out of the analysis. The plots and values removed are printed and returned with the data.

Usage

remove_outliers(
  data,
  traits,
  multiplier = 1.5,
  by = if ("location" %in% names(data)) "location" else NULL,
  id_cols = c("location", "plot_name", "plots", "plot_number", "all_entries",
    "accession", "row", "col"),
  quiet = FALSE
)

Arguments

data

A data frame with one row per plot.

traits

Names of the numeric trait columns.

multiplier

The IQR multiplier \(k\).

by

Optional grouping column(s); by default "location" when the data have it.

id_cols

Columns that identify the plot in the report (those present in data).

quiet

Logical. If FALSE (default), print what was removed.

Value

data with the outlying values set to NA. The removed values are in attr(result, "outliers") (as from find_outliers()).

Examples

d <- data.frame(plots = 1:20, all_entries = paste0("G", 1:20),
                yield = c(rnorm(18, 50, 2), 80, 10))
clean <- remove_outliers(d, "yield")
#> Removed 2 of 20 values as outliers (outside Q1 - 1.5 IQR and Q3 + 1.5 IQR):
#>  plots all_entries trait value lower upper side
#>     19         G19 yield    80 43.13 54.86 high
#>     20         G20 yield    10 43.13 54.86  low
attr(clean, "outliers")
#> # A tibble: 2 × 7
#>   plots all_entries trait value lower upper side 
#>   <int> <chr>       <chr> <dbl> <dbl> <dbl> <chr>
#> 1    19 G19         yield    80  43.1  54.9 high 
#> 2    20 G20         yield    10  43.1  54.9 low