What are the Sources of Data in Demography

I have offered you a variety of facts thus far as I described the history of population growth and provided you with an overview of the world’s population situation. To analyze the demography of a particular society, we need to know how many people live there, how they are distributed geographically, how many are being born, how many are dying, how many are moving in, and how many are moving out. Let me begin the discussion, however, with sources of basic information about the numbers of living people, births, deaths, and migrants.

Sources of Demographic Data

We collect all information or data mainly two sources-

1. Primary Sources: Through census or sample we collect data through a particular variable is called primary sources of data.
2. Secondary Sources: UNDP, UNISEF, BBS, IMF, HDR, BD all of its reports are called  secondary sources of data

In demography there are three sources of demographic data collection-

1. Census of population.
2. Registration of vital statistics 
3. Sample survey

Other sources of demographic data-

• Administrative data.
• Historical data.

(Sources: weeks, 2008)

These types of sources of data collection are described in the below:

A. Census of population 

For centuries, governments have wanted to know how many people were under their rule, who the taxpayers were, or they wanted to identify potential laborers and soldiers. The most direct way to find out how many people there are is to count them, and when you do that you are conducting a population census. The United Nations defines a census of population more specifically as 

“the total process of collecting, compiling and publishing demographic, economic and social data pertaining, at a specified time or times, to all persons in a country or delimited territory” (United Nations 1958:3).

Who is included in the Census?

There are several ways to answer that question, and each produces a potentially different total number of people. At one extreme is the concept of the de facto population, which counts people who are in a given territory on the census day. At the other extreme is the de jure population, which represents people who legally “belong” to a given area in some way or another, regardless of whether they were there on the day of the census.

Most countries (including the United States, Canada, and Mexico) have now adopted a concept that lies somewhere between the extremes of de facto and de jure, and they include people in the census on the basis of usual residence, which is roughly defined as the place where a person usually sleeps. College students who live away from home, for example, are included at their college address rather than being counted in their parents’ household. People with no usual residence (the homeless, including migratory workers, vagrants, and “street people”) are counted where they are found. On the other hand, visitors and tourists from other countries who “belong” somewhere else are not included, even though they may be in the country when the census is being conducted. At the same time, the concept of usual residence means that undocumented immigrants (who legally do not “belong” where they are found) will be included in the census along with everyone else.

Errors in census

There are several possible errors that can creep into the enumeration process. We can divide these into the two broad categories of non-sampling error (which includes coverage error and content error) and sampling error.

Non-sampling errors: The two most common sources of error in a census are coverage error and content error. 

• Coverage errors: A census is designed to count everyone, but there are always people who are missed, as well as a few who are counted more than once. The combination of the undercount and the over count is called coverage error, or net census undercount.
• Content Error: Although coverage error is a concern in any census, there can also be problems with the accuracy of the data obtained in the census (content error). Content error includes non-responses to particular questions on the census or inaccurate responses if people do not understand the question. Errors can also occur if information is inaccurately recorded on the form or if there is some glitch in the processing (coding, data entry, or editing) of the census return.

Sampling errors: If any of the data in a census are collected on a sample basis (as is done in the United States, Canada, and Mexico), then sampling error is introduced into the results. With any sample, scientifically selected or not, differences are likely to exist between the characteristics of the sampled population and the larger group from which the sample was chosen. However, in a scientific sample, such as that used in most census operations, sampling error is readily measured based on the mathematics of probability. To a certain extent, sampling error can be controlled— samples can be designed to ensure comparable levels of error across groups or across geographic areas (U.S. Census Bureau 1997). Non-sampling error and the biases it introduces throughout the census process probably reduce the quality of results more than sampling error (Schneider 2003).

B. Registration of Vital Events

When you were born, a birth certificate was filled out for you, probably by a clerk or volunteer staff person in the hospital where you were born. When you die, someone (again, typically a hospital clerk) will fill out a death certificate on your behalf. Births and deaths, as well as marriages, divorces, and abortions, are known as vital events, and when they are recorded by the government and compiled for use they become vital statistics. These statistics are the major source of data on births and deaths in most countries, and they are most useful when combined with census data.

Registration of vital events in Europe actually began as a chore of the church. Priests often recorded baptisms, marriages, and deaths, and historical demographers have used the surviving records to reconstruct the demographic history of parts of Europe (Landers 1993; Wall, Robin, and Laslett 1983; Wrigley 1974; Wrigley and Schofield 1981).

C. Sample Surveys

There are two major difficulties with using data collected in the census, by the vital statistics registration system, or derived from administrative records: 

1. They are usually collected for purposes other than demographic analysis and thus do not necessarily reflect the theoretical concerns of demography, and 
2. They are collected by many different people using many different methods and may be prone to numerous kinds of error. 

For these two reasons, sample surveys are frequently used to gather demographic data. Sample surveys may provide the social, psychological, economic, and even physical data I referred to earlier as being necessary to an understanding of why things are as they are. Their principal limitation is that they provide less extensive geographic coverage than a census or system of vital registration. By using a carefully selected sample of even a few thousand people, demographers have been able to ask questions about births, deaths, migration, and other subjects that reveal aspects of the “why” of demographic events rather than just the “what.” In some poor or remote areas of the world, sample surveys can also provide good estimates of the levels of fertility, mortality, and migration in the absence of census or vital registration data.

D. Administrative Data

Knowing that censuses and the collection of vital statistics were not originally designed to provide data for demographic analysis has alerted demographers everywhere to keep their collective eyes open for any data source that might yield important information. For example, an important source of information about immigration to the United States is the compilation of administrative records filled out for each person entering the country from abroad. Of course, we need other means to estimate the number of people who enter without documents and avoid detection by the government. At the local level, a variety of administrative data can be appointed to determine demographic patterns. School enrollment data provide clues to patterns of population growth and migration. 

E. Historical Sources

Our understanding of population processes is shaped not only by our perception of current trends but also by our understanding of historical events. Historical demography requires that we almost literally dig up information about the patterns of mortality, fertility, and migration in past generations. Historical sources of demographic information include censuses and vital statistics, but the general lack of good historical vital statistics is what typically necessitates special detective work to locate birth records in church registers and death records in graveyards. Even in the absence of a census, a complete set of good local records for a small village may allow a researcher to reconstruct the demographic profile of families by matching entries of births, marriages, and deaths in the community over a period of several years. Yet another source of such information is family genealogies, the compilation of which has become increasingly common in recent years. For example, small family units were quite common throughout Europe for several centuries before the Industrial Revolution and may actually have contributed to the process of industrialization by allowing the family more flexibility to meet the needs of the changing economy. In the United States, conversely, extended families may have been more common prior to the nineteenth century than has generally been thought (Ruggles 1994).