INTRODUCTION

The objective of a sample survey is to make inferences about a population parameter by observing a portion or sample from it. It is natural to expect that this phenomenon of interpreting about a wider group (a population) on the basis of a sample from it will have some margin of error in the result. How the samples are drawn or selected, while designing a sample survey, is important to ensure that the inference drawn is both valid and reliable. A proper design not only helps obtain a valid idea about the wider population parameters, but also provides a margin of error in an estimate. In addition, the theory can also guide in choosing an alternative design so that the margin of error can be minimized. Different designs that can be employed in a sample survey are discussed in the subsequent chapters. This chapter elaborates few terminologies that will facilitate further discussion on the specific methodologies.

POPULATION, UNITS AND SAMPLING UNITS

Aggregates of units or elements comprise a population. A study unit is a unit or a member of a study population, as in the case of a study concerning human beings, it can be individual persons, families or households in a given geographical area. A study unit, however, can be any living or non-living subject in a study. For example, if the interest is to estimate the number of fruits in an orchard, the study units will be the fruits in the area. Further, if the interest is to estimate the number of words in a book, words of the book will form the study unit.

A distinction, however, needs to be made between a study unit and a sampling unit. Considering an example of sampling from a human population, it is often preferred to select households before selection of individuals – the study units in a population. In this case, the households, consisting of individuals, which facilitate the process of the sample selection are called sampling units. However, if our aim is also to estimate household income, then household will be both the study unit and the sampling unit. Therefore, a sampling unit can also be the study unit in a study. Generally, sampling units are a higher level or a group of units.