We can specify the aggregation functions by use of function names in quotes (e.g., 'sum'), as bare functions (e.g., sum), or as one-sided R formulas using a leading ~.

We can provide our preferred names by naming the functions within the list (e.g, list(average = 'mean', total = 'sum', SD = 'sd')).

So, each summary row will receive a summary row label. Each function provided will result in a summary row for each group.īecause each function will yield a row, we need to be able to label that row. We can provide base functions such as mean(), sum(), min(), max(), and more, within a list(). To make any sort of summary, we need to use functions that will perform the aggregation. Since each call to summary_rows() only performs one set of aggregation functions, we may want specific aggregations for different subsets of columns. Aside from the selection of groups, there is control over which columns are to used for the summary. We can provide a vector group names, as in c('grp_a', 'grp_b'), or, use TRUE to signify that all groups should receive summary rows. Groupwise summary rows can be generated by using the summary_rows() function and, importantly, through specifying the groups that will receive summary rows. To make the examples a bit easier to follow, some of the columns in exibble will first be dropped through a dplyr::select() statement. The end result will be a table organized with labeled rows that are grouped together (the row group labels identify each of the row groups). We’ll create a table stub with both row labels (using the row column) and row groups (using the group column).

Summary r studio