When using random sampling techniques, large numbers of samples/records are taken from different random positions within a habitat. A quadrat (a square frame) is most often used for random sampling. It is used to obtain samples from areas of consistent size. The actual size of the quadrat is determined by the habitat being sampled and by the purpose of the survey.

In the simplest form of random sampling, the quadrat is thrown to fall at ‘random’ within the site. However, this is unsatisfactory because a personal element enters into the throwing and it is never completely random. True randomness is an important factor, because many of the common statistical techniques used to process results are only valid on data that is truly randomly collected. This technique would also only be possible where quadrats of small size are being used. It would not be possible to throw anything larger than a 1m quadrat and even this might pose difficulties.

A better method of random sampling is to map the area and then to lay a numbered grid over the map. A (computer generated) random number table is then used to select which grids to sample in.

In this survey, a numbered grid was drawn on paper to represent the 20m square to be randomly sampled. Each square on the grid represents a 2m square quadrat in the sampling square. Ten consecutive random numbers (less than 100) were chosen from a random number table. The quadrats in the grid bearing these numbers were then sampled.