How data are collected

Information about population is collected in two main ways: by enumeration at a point of time, and by recording events as they occur over a period. Censuses and social surveys are examples of the first method, and provide ‘stock’ data (see §2.5), while birth registrations and migration records (‘flow’ data) are examples of the second. The information may purport to be complete or it may take the form of a sample. The aim of this chapter is to describe the chief features of the statistics so collected and to indicate the limitations the demographer may need to bear in mind in attempting to interpret the meaning of the published information. In this connexion it is important to consider the relationship the data bear to the biological events in human existence which are the basic objects of study. Although these events are essentially the ‘facts of life’ with which people generally are familiar, what the student of population needs is a precise knowledge of how they are defined for practical purposes and their consequent place in demographic analysis.

Reliance on ‘common sense’ may not be enough to prevent serious misinterpretation of the results of statistical studies. While it is obvious what a ‘birth’ is, the demographer requires to know whether recorded births include or exclude still-births and how still-births are defined. Are the births those which occurred in a given period, or those which were registered during that period? If the latter, what are the rules concerning registration and what delays can occur before registration is effected? Are all births notified to the authorities, or do some escape recording? Such matters can be important when analysing changes in experience, or in measuring differences between populations – more particularly in international comparisons.