Our answers to the latest question in our “ask us anything” series. Question has been paraphrased, click through to see the full original question.
Jeremy: Look at this picture and ask yourself, “Do I really want this guy’s advice on field study design?”😉
Brian: There are two good books on field methods: Henderson & Southwood and Krebs I can recommend both first hand (they both sit in prime places on my bookshelf). There are plenty of others if you look that seem fine but I can’t personally recommend them. Then of course there is the more broad subject of experimental design. There’s a ton of books on that topic and I don’t have a specific one to recommend (although Box’s book is a classic). But you can do worse than just go read Hurlbert’s paper on pseudoreplication (Meg’s post on this paper). There is a whole section (3 chapters) in Gotelli & Ellison’s Primer of Ecological Statistics that is also very good.
I take it as a given that you are already accessing colleagues and mentors as a key resource as well. Experimental design is an art with a lot of peculiarities particular to a system (you mention being a forest ecologist) and most of this is uncapturable in a book.
But I would also just give you kudos for asking the right question. Many people want to ask what is the right statistical method. But asking what are the right field methods and experimental designs are actually much more important (and have to be decided much earlier and ignored by more people).
Jeremy adds: I actually thought of suggesting Southwood [who was the sole author of the earlier edition that I own], but didn’t because I wasn’t sure if it’s still regarded as the best reference. So now having seen Brian’s recommendation and having total confidence in anything Brian recommends, I’m unreasonably proud of having almost given advice that would’ve been good if I had given it.🙂