Multi stage or Cluster Random sampling
Under this method, the random selection is made of primary, intermediate and final units from a given population. The area of investigation is scientifically restricted to a small number of ultimate units which are representative of the whole. This will reduce the cost compared with a simple sampling from the whole area with a number of straight random selections. For example, from a population of a climatic region of the eastern states the two and south eastern from within each of these primary sampling areas a certain number of blocks May e selected are random or ultimate selection of the village for complete enumeration of households. At each stage there is a random selection and the size of sample may be proportional or disproportional depending on the size variations relevant to the purpose of inquiry.
Merits:
(1) It introduces flexibility in the sampling method which is lacking in other methods.
(2) This method is very helpful in large scale investigations where the preparation of list of all units of the universe is very difficult and expensive. For example in a socio economic survey certain families are to be selected from different village different villages are to be selected from the list of districts of a state which are given ,This is a case of three stage sampling. It will consume time and money in the selection of items from the population.
(3) This method is useful in such cases where sub division into second stage units be carried out only for such first stage units which are included in the sample.
Limitations:
A multi stage sample is usually less accurate than a sample containing the same number of final stage units selected by means of suitable multi stage process.