There are two books I'd highly recommend for someone starting to dabble in Epidemiology. The first is Kenneth Rothman's Epidemiology: An Introduction, which I'm affectionately going to refer to as "Baby Rothman" from hereon out for reasons that will become obvious.
Baby Rothman is an excellent "getting your feet wet" book. It's simplistic, and clearly designed for someone with no background in Epidemiology, so it starts from things that, if you have a statistical background, you might find a bit dull. But it accomplishes two key tasks:
- A basic treatment of the subject from a pre-eminent methodologist and advocate for the field that doesn't make some of the sloppy slight-of-hand mistakes many introductory books do.
- An introduction to the "vocabulary" of Epidemiology as a field - what we call things, how we thing about effect estimates, etc.
The second book I would buy would be Modern Epidemiology: 3rd Edition by Rothman, Greenland and Lash. This is the definitive reference book for most epidemiological methods currently used in the field. As I said on a related CrossValidated question, ME3 is the only epidemiology book that has always come with me, regardless of what project I'm doing, and which I have routinely found invaluable.
The other benefit of the book is that it's "platform agnostic". It's not going to try to teach you a statistics package - its going to teach you the science and philosophy of Epidemiology, the actual doing of the science, rather than the toolkit. Without that kind of methodological rigor, you can be a wizard at your language of choice and still do bad science.
If I were only going to buy one book, it would be ME3.
No comments:
Post a Comment