A directory ("folder", if you like) is a file like all the others, just a somewhat "special" one. It's own size is rather small, but if you'd like to know the size of all of it's contents (recursively), you can use 'du':
for more information about the command line switches available.
I'll add an example:
First let's create a directory and a few files:
dd if=/dev/urandom of=testdir/testfile1 bs=2M count=1
dd if=/dev/urandom of=testdir/testfile2 bs=3M count=1
dd if=/dev/urandom of=testdir/testfile3 bs=5M count=1
Then what does the directory information look like?
It prints this:
drwxr-xr-x 2 username username 4096 2008-05-06 12:03 testdir
It's size seems to be 4K. Ok, that doesn't count the file sizes:
-rw-r--r-- 1 username username 2097152 2008-05-06 12:03 testfile1
-rw-r--r-- 1 username username 3145728 2008-05-06 12:03 testfile2
-rw-r--r-- 1 username username 5242880 2008-05-06 12:03 testfile3
What if we use 'du' to calculate the disk usage, then?
or in human-readable form:
As a note: the size of the content of a file isn't necessarily exactly the same size it takes up in disk space. For example if you create a file smaller than 4K, it's size still might show up as 4K (rounded up); this is because the filesystems typically don't (can't) reserve space exactly as much as is needed, but instead they divide the disk space into "blocks" of certain size, and the real data then uses as many blocks as it needs - the final block then either fills up, or more often is left partially empty. To get more information, read about filesystems; these things might sound stupid, minor or both, but actually they do have quite a lot to do with efficiency of a filesystem. Also note that in "human readable form" the file sizes are again rounded a little..