Hi Alex,
This is a known bug when trying to do use the CLIPPER tool with the North Atlantic polygon data layer in QGIS 2.18.2. The exercises were specifically written for QGIS 2.8.3 (which is an earlier version than the one you're using), and this is the recommended version for doing the exercises.
There are two possible solutions here.
The first is to install QGIS 2.8.3, and use it to complete the exercise. You can do this from the links provided on the GIS For Biologists support page (
http://www.GISinEcology.com/GFB.htm).
The second is to continue the exercise without the CLIPPER step. This will give you the correct species richness values for the North Atlantic, but it will incorrect value for the small areas of the North Pacific that are covered by the raster data layer (since distributional information on all North Pacific beaked whale distributions was not included in the species richness calculation).
There is also a third option, where you can convert the North Atlantic polygon data layer into a raster data layer, and then use the RASTER CALCULATOR tool to apply it as a mask, but this is a more complicated work-around. If you'd like to try this option, post back on this thread, and I'll post some specific instructions on how to do this.
All the best,
Colin