Sample Surveys

A census is a complete enumeration of the population: data are collected from every unit in the population. In a survey, a subset of the population, called a sample, is taken. Census are taken at regular but infrequent intervals, e.g. every 5 or 10 years. In between, surveys are used to update results. The selection and estimation procedures of official surveys are almost always based on previous census of that population. A probability sample is such that one can calculate the probability of inclusion in the sample for each unit in the population. Then one can apply probability theory to estimate sampling error and find confidence intervals. Two types of non-probability samples are purposive sampling (units are chosen subjectively to be “representative” of the population) and quota sampling (enumerators are instructed to select units to fill a quota, i.e. a prescribed number of units of either sex, in certain age-groups, social classes, etc.) We shall consider only probability sampling. Advantages of surveying: