Random sampling using quadrats Sampling of the area you are studying must be random. It must show no bias – for instance, choosing to sample where there are lots of plants.