In cluster sampling there is a successive random . sampling of units. The first unit to be sampled is large grouping or clusters. For example, in drawing a sample of nursing students, researcher might first draw a random sample of nursing schools of the country. To draw the schools of nursing, the random selection of the State have to be done. So the usual procedure for selecting general sample of students is first to select states, cities, districts, then schools and students successively. Although cluster sampling may be necessary for large scale survey studies, the chance of sampling errors increases with each stage of sampling. Despite the disadvantage cluster sampling is considered more economical and practical than other types of probability sampling, particularly when population is large and widely disperse.