Make a table showing the 10 largest values of a variable in R?

Question

I want to make a simple table that showcases the largest 10 values for a given variable in my dataset, as well as 4 other variables for each observation, so basically a small su

Answer 1

You can do this using arrange from dplyr. This should also work if there are grouping variables. Just add group_by before the arrange. We filter the first 10 observations using slice.

 library(dplyr)
 df1 %>%
    arrange(desc(Score)) %>%
    slice(1:10) 

Or another option is ?top_n (commented by @docendodiscimus) from dplyr which is a wrapper that uses filter and min_rank to select the top n (i.e. 10) entries for 'Score'.

 top_n(df1, 10, Score)    

Or we use filter by creating a logical condition with row_number which is equivalent to rank(ties.method='first') (contributed by @Steven Beaupre)

 filter(df1, row_number(desc(Score)) <= 10)

Or a data.table option (by @David Arenburg). We convert the 'data.frame' to 'data.table' (setDT(df1)), order (decreasing) the 'Score' variable, and select the first 10 observations. .SD means Subset of DataTable.

 library(data.table)
 setDT(df1)[order(-Score), .SD[1:10]]