Functions for computing item summary, i.e. some summary measurements (of arbitrary nature) of item (one or more columns) present in data frame.
summarise_item(tbl, item, ..., .prefix = "")
summarise_game(tbl, ..., .prefix = "")
summarise_player(tbl, ..., .prefix = "")
summarize_item(tbl, item, ..., .prefix = "")
summarize_game(tbl, ..., .prefix = "")
summarize_player(tbl, ..., .prefix = "")| tbl | Data frame. |
|---|---|
| item | Character vector of columns to group by. |
| ... | Name-value pairs of summary functions (as in dplyr::summarise). |
| .prefix | A string to be added to all summary functions' names. |
Output of summarise() as not grouped tibble.
Basically, summarise_item() performs the following steps:
Group tbl by columns stored in item. Note that starting from 0.8.0
version of dplyr this might give a warning in case of implicit NAs in
factor columns (NA present in column values but not in its levels)
suggesting to add NA to levels.
Apply dplyr's summarise().
Ungroup result.
Convert to tibble.
Add .prefix to names of summary functions.
summarise_game() and summarise_player() are wrappers for
summarise_item() using item = "game" and item = "player" respectively.
Common item summary functions for competition results.
ncaa2005 %>%
dplyr::mutate(game_type = game %% 2) %>%
summarise_item(c("game_type", "player"), mean_score = mean(score))
#> # A tibble: 10 x 3
#> game_type player mean_score
#> <dbl> <chr> <dbl>
#> 1 0 Duke 10.5
#> 2 0 Miami 25
#> 3 0 UNC 15.5
#> 4 0 UVA 12
#> 5 0 VT 48.5
#> 6 1 Duke 7
#> 7 1 Miami 37.7
#> 8 1 UNC 9.5
#> 9 1 UVA 38
#> 10 1 VT 18.5#> # A tibble: 10 x 3
#> game mean_score min_score
#> <int> <dbl> <int>
#> 1 1 29.5 7
#> 2 2 22.5 21
#> 3 3 22.5 7
#> 4 4 22.5 0
#> 5 5 25 16
#> 6 6 21 17
#> 7 7 17 7
#> 8 8 6 5
#> 9 9 16.5 3
#> 10 10 33 14