How can I convert the summary run on a data.frame into a data.frame itself? I need a data.frame as an output to knitr::kable in RMarkdown.
In particular I have this dataframe
d <- data.frame(a=c(1,2,3), b=c(4,5,6))
ds <- summary(d)
class(ds)
# returns "table"
And I need ds in a data.frame format.
The output I would like would be a data.frame with "Min.", "1st Qu.", "Median", etc. as row names, "a" and "b" as column names, and the numbers in the cells.
as.data.frame doesn't work:
ds.df <- as.data.frame(ds)
print(ds.df)
# Output is messed up
The code in this related question doesn't work either:
df.df2 <- data.frame(unclass(summary(ds.df)), check.names = FALSE, stringsAsFactors = FALSE)
print(df.df2)
# Output equally messed up
broom::tidy on a table is deprecated and in anyway returns an error:
df.df3 <- broom::tidy(ds)
# Returns error
# Error: Columns 1 and 2 must be named.
# Moreover
# 'tidy.table' is deprecated.
The as.data.frame.matrix puts "Min" and the other names of the statistics inside each cell, instead of them being row names:
ds.df3 <- as.data.frame.matrix(ds)
print(ds.df3)
# Returns "Min" and "1sd Qu." inside the cell
# instead of them being row names
We could use the matrix route
out <- as.data.frame.matrix(ds)
row.names(out) <- NULL
-output
out
a b
1 Min. :1.0 Min. :4.0
2 1st Qu.:1.5 1st Qu.:4.5
3 Median :2.0 Median :5.0
4 Mean :2.0 Mean :5.0
5 3rd Qu.:2.5 3rd Qu.:5.5
6 Max. :3.0 Max. :6.0
If we need the min etc as row names, loop over the columns with sapply and apply the summary
as.data.frame(sapply(d, summary))
-output
a b
Min. 1.0 4.0
1st Qu. 1.5 4.5
Median 2.0 5.0
Mean 2.0 5.0
3rd Qu. 2.5 5.5
Max. 3.0 6.0