Zuur AF, Ieno EN, Elphick CS (2010) A protocol for data exploration to avoid common statistical problems. Methods in Ecol. Evol. 1: 3-14.
If you are a biologist who practices statistics or teaches statistics this is a nice article and a good place to start any course. Provides some nice examples of 8 areas that should be considered. Some of the ol' standbys such as normality and heterogeneity, and some other less often considered ones like collinearity or issues of "double zeros". Worth a read if you have some time.
I also highly recommend Zuur's book, Mixed Effects Models and Extensions in Ecology with R. Its dense, but provides great examples and lots of good advice. I've shared it with many students as well, so its approachable given a little time.