[R] [PS] Re: a more elegant way to get percentages? (now R books)

[hadley wickham]

>  There has been a virtual population explosion of R books in recent years
>  and we all have our favorites.  You may wish to pick one oriented toward
>  your specialty, but the absolute minimum lowest common denominator (by
>  which I mean that it has the ground zero essential information that all
>  users must share, not that it is minimal or incomplete) is the manual
>  "An Introduction to R," available by download from the Cran website.

I don't mean to pick on you in particular, or on the authors of "An
introduction to R", but I really don't see how anyone in good
conscience can recommend this to a new user of R.  I think it does a
great job of covering the basics, and is probably a good read after
you've been using R for a year or so, but in goes into a lot of depth
into things that you really don't need to know for doing practical,
day-to-day data analysis.

For example, you don't find out how to actually load data into R until
page 30, while you get 3 page on the mode and length of objects at
page 12.  Do we really need to know that an empty (zero length) vector
still has a mode?

These comments are based on my experience teaching R to undergrad stat
majors, and so may not apply to your audience.  If you teach R in the
same order as "an introduction to R" it takes you about 4 weeks before
you can actually do anything useful with R, by which time the students
are bored to tears.  If you start with getting data into R and
displaying the data with graphics, you can do useful things very
quickly, providing interest and motivation, and then you can gradually
introduce a more rigourous description of the components as needed.

Hadley

-- 
http://had.co.nz/