head and tail, for those familiar with the unix command line, are two very handy utilities for looking at data frames. Along with str, which displays the structure of a data frame, they help you look at your data:

`

> d <- data.frame(mean=rep(1:10,5), val = rnorm(n=50, mean=rep(1:10,5)))
> d <- d[ order(d$mean), ]
> 
> str(d)
'data.frame':   50 obs. of  2 variables:
 $ mean: int  1 1 1 1 1 2 2 2 2 2 ...
 $ val : num  2.303 2.222 -1.153 1.795 -0.232 ...
>
> head(d)
   mean        val
1     1  2.3026422
11    1  2.2216371
21    1 -1.1533163
31    1  1.7945563
41    1 -0.2318763
2     2 -0.4994239
> 
> tail(d)
   mean       val
49    9  8.462525
10   10 10.437314
20   10 10.815264
30   10 10.218853
40   10  9.754245
50   10  9.596825
> 

`

If you are familiar with data frames, you’ll know that head(d) is no different than displaying the first 6 rows via subsetting, eg, d[1:6, ] but it saves some typing. tail, on the other hand, saves us from either asking first how many rows a data frame has with nrow, or typing a mess: d[ (nrow(d)-5):nrow(d), ] `

> nrow(d)
[1] 50
> d[45:50,]
   mean       val
49    9  8.462525
10   10 10.437314
20   10 10.815264
30   10 10.218853
40   10  9.754245
50   10  9.596825
> d[ (nrow(d)-5):nrow(d), ]
   mean       val
49    9  8.462525
10   10 10.437314
20   10 10.815264
30   10 10.218853
40   10  9.754245
50   10  9.596825
>