A comparison of statistical and participatory typologies from a case study in northern Ghana. Are they complementary? Should you go for one or the other – depending on your purpose? Or are they best applied together?