I have to say that the two posters (or more) who recommended both Edward Tufte and Hans Rosling are right on.

I would start by reading Tuftes first book – ‘The Visual Display of Quantitative Information’. Even better, attend one of his seminars. He does a few each year around the country and with admission you get all of his works. I attended last year and learned a ton. For Hans, check out both of his TED Talks (search iTunes podcasts)

I am active in biotech research and routinely need to present reams of data and conclude with a nice succinct point. Combining the theories of both these guys has changed my life (really like suddenly becoming a Buddhist or naturist).