Demography in Sociology

Abstract

Briefly defined, demography is the study of human population change. Many of our oldest documents make reference to counting or detailing populations. Populations are dynamic, constantly changing. This constant change in populations and their characteristics is the focus of demography, yet also presents challenges to demographers. The study of human society, or sociology, developed in tandem with demographic studies. A vast and formidable field of study in and of itself, demography is considered an essential sub-field of sociology as well as a number of other human and biological sciences. Dynamic models of population change processes not only answer questions about current populations but also give us a glimpse of possible future scenarios for population growth and change. This article explains some basic demographic concepts and gives examples of their use in demographic studies.

Overview

Briefly defined, demography is the study of human population change. The term derives from the Greek words demos, or "people," and graph, which means "cipher" or "write." Population change has been a human fascination since the beginning of recorded history. Many of our oldest documents make reference to counting or detailing populations through a census, which is an enumeration and description of people in a given area at a given point in time. Even these early counts detailed the characteristics of a population, such as a person's origin, dwelling place, and property owned. These accounts were often used for purposes of assessing taxes.

One of the oldest surviving census rolls is known as the Domesday Book, a work commissioned in 1085 by William the Conqueror after he invaded British lands in 1066. The Domesday Book contains records of 13,418 settlements in the southern part of England. William needed an accounting of the new people and property under his dominion so that he could command tax payments to support his rule and conscript men for service (Maitland, 1987). Housed at the National Archives in Kew, London, this document can be viewed online at the Open Domesday website.

However, references to census taking predate the Domesday Book and are found even in the oldest scriptures of the world's most enduring religions, including Judaism, Christianity, and Islam. A Muslim scholar, Ibn Khaldûn, is often credited with first using demographic methods in his study of social and economic systems in the mid-fourteenth century (Boulakia, 1971). Populations are dynamic and constantly changing. Certainly the world's population is much changed from the time of William the Conquerer, or even Ibn Khaldûn. This constant change in populations and their characteristics is the focus of demography. The evolution of contemporary demography coincided with the development of the new techniques of statistics, arithmetical and mathematical methods extended to assemble, classify, and tabulate numerically based data. The study of human society, or sociology, grew alongside demographic studies. Indeed, demographic techniques allowed sociologists to assert the empirical--that is, observable--nature of their formal discipline, giving rise to early studies of society such as those of Auguste Comte, who first coined the term "sociology." A vast and formidable field of study in and of itself, demography is considered an essential subfield of sociology and a number of other human and biological sciences as well.

Demographic Concepts: The Basics. Vital statistics, such as births and deaths, are direct building blocks used in demographic analysis. They are primary data, assembled through direct observations of life events, typically collected continuously and summarized on an annual basis by governmental units. Another source of primary data is censuses, which are usually conducted by a national government or intergovernmental organization that attempts to enumerate every person in a given location. A census often does more than just count people; it also gathers information about families or households, as well as individual characteristics such as age, sex, marital status, literacy, education, employment status, occupation, and geographical location. Censuses are usually conducted only every few years, often once per decade or less. The study of population change would not exist without the human ability to assemble and tabulate these fundamental data points.

While a census provides a snapshot of a population at one point in time, the most important task of demographers is to document population dynamics, or the processes by which populations change. The three most basic processes of population dynamics are those that affect population size:

• Fertility,
• Mortality, and
• Migration (Weinstein & Pillai, 2000).

These three essential population processes underlie all other changes in population and are used in the most basic model of population dynamics. Models of population processes are developed through theories, or sets of hypotheses, pertaining to how population processes operate together to dynamically produce change. These hypotheses are ultimately expressed as sets of equations called models, which must then be tested against actual observed data points.

Fertility. A logical starting point in the study of population dynamics is the assembly and tabulation of fertility data. Fertility is the variable frequency of childbearing in a given population. The crude birth rate, or the annual number of live births per 1,000 people in a population, is used to represent fertility.

Mortality. Even as children are born, people die. Mortality, the variable frequency of death in a given population, is represented by the crude death rate, or the annual number of deaths per 1,000 people in a population.

Migration. The size of a given population is also affected by migration, the variable movement of people into and out of the population. Demographers must consider both immigration, in which people move into a specified population, and emigration or out-migration, in which people leave a specified population. The combination of the two is the net migration rate, which is the difference between persons gained through immigration and those lost through emigration, expressed as a ratio per 1,000 people in the population.

Population Dynamics: The Most Basic Demographic Formula. The most basic formula in demography uses these basic changes in the size of a population to provide an overall rate of population growth or decline. In its simplest form, this formula can be expressed as:

P_t2 = P_t1 + (CBR - CDR) + NMR where:

P_t2 = total number of persons in a population at time t2

P_t1 = total number of persons in a population at time t1

CBR = crude birth rate

CDR = crude death rate

NMR = net migration rate

With this formula and the data needed to calculate a basic population growth rate (or rate of decline), we now have the foundation from which to begin investigating other population characteristics and the parts these characteristics play in population dynamics. Also called demographic variables, these include any characteristics, attributes, or properties of people or collectives of people that can vary between persons or collectives. Many of these characteristics affect or are otherwise related to the three basic population processes.

Widely Used Demographic Variables. There are so many variations between people and collectives of people that we cannot begin to detail them all in this context. Age, sex, pregnancies, births, siblings, sexual orientation, gender orientation, ethnicity, race, national citizenship, religious preference, occupation, income, and wealth are just a few of the variables most frequently used in the study of human societies. Models are assembled to represent the interrelation of these variables. We can look at an example of how such characteristics are related to basic population processes, beginning with fertility. Obviously, sex and sexual orientation affect sexual reproduction and fertility. Age is also closely related to sexual reproduction, as it affects a woman's ability to bear children. The widest range of the childbearing age for women is from onset of menarche, around 12 years of age, to onset of menopause, usually in the late forties to mid-fifties. All of these factors affect fecundity, the maximum possible number of births per woman in the population.

Fecundity. While it is possible, indeed documented, that women can give birth to 30 or more children--usually including multiple births, such as twins or triplets (Pearl, 1939)--these are rare cases. Typically, fecundity is estimated at around 12 to 15 births maximum per woman. However, a number of other life circumstances affect the actual number of children born (Kent & Haub, 2005). Cultural norms, occupation, income, and the availability of contraception all affect the choices women and couples make about having children (Koenig, et al., 2006; Bollen, Glanville, & Stecklov, 2007).

Furthermore, recall that the crude birth rate only counts live births. Differing standards of living between locations, in addition to variations in income within locations, can affect a woman's access to nutrition, clean water, and health care, which in turn affects the viability of any pregnancy. The availability of certain methods of birth control also affects fecundity in a population.

Such differences in life circumstances also affect how well a child fares after birth, thus influencing a population's infant mortality rate (Kent & Haub, 2005). This is the number of annual deaths of children under one year of age per every 1,000 live births that year.

Applications

Demographic Models. We begin to see the dynamic interrelation of demographic variables. Complex statistical methods have been developed to assemble, classify, and tabulate data to describe and model the extent and type of relationships between demographic variables. Dynamic modeling of variables provides estimates and projections of population change (see for example Haupt & Kane, 2004; Korotayev, Malkov, & Khaltourina, 2006).

Modeling also helps in the development of theory about how numerous variables combine to form patterns of characteristics among human populations. Figure 1 is a graphic representation of variables in a model developed by researchers (Bollen, Glanville, & Stecklov, 2007) investigating the relationship between socioeconomic status, income, and fertility in Ghana and Peru. In the model, Ɛ represents random measurement error, ζ represents equation disturbance, and "Head" refers to the male head of household.

Demographers conduct many different types of studies, not all of them based in dynamic modeling. Three essential types of studies are necessary to the overall enterprise of demographic research:

• Description
• Estimation
• Projection

Description. Demographic research based in census data is a good example of population studies that use descriptive statistics. Descriptive studies seek only to describe the characteristics of a population at a given point in time or how those characteristics changed between two points in time. Such demographic profiles are static; they are the population snapshots mentioned earlier. Descriptive studies are not dynamic and therefore do not require detailed modeling of the interaction of the variables, although these studies often use statistical equations to provide additional information about possible relatedness between variables. Descriptive demographic studies usually include large tables of statistical data detailing the characteristics of a given population. Though not as rich in detail, visual representations of the data are usually more effective in illustrating an overall pattern in a population. One very effective, and therefore common, graphic illustration used to present demographic profiles is called a population pyramid (Weinstein & Pillai, 2000).

Figure 2 shows a population pyramid illustrating the age and sex composition of the population of China in the year 2000 (US Census Bureau, 2008). The pyramid is built using horizontal bar graphs to represent the number of persons of each sex that fall within in each age range.

Estimation. An estimate is a number calculated through statistical inference, not direct observation of people or events. Population estimates are based on past observations of the same population and are necessary when direct measurement is not possible at a specific point in time. For example, the number of people connected to the Internet right at this very moment can be estimated based on actual counts of Internet users taken at this time of day on previous days, going back months or even years. Population estimates are often calculated between census years. Estimation models are derived from many past direct observations, but they do not predict future events (see Bickel et al., 1993). Estimates can be checked for accuracy after the fact. They can also be used to check the accuracy of current direct observation by a process of comparison, a method often used with census data.

Projection. A projection is based on a conditional if-then statement about the future. For example, if a city grows by 10 percent between 2000 and 2010, from 10,000 residents to 11,000, and a continued ten-year growth rate of 10 percent is assumed, then the city will have a population of 12,100 in the year 2020.

A projection is an assumption about the future based on the past. As long as such if-then assumptions are made explicit, another knowledgeable person can decide the extent to which a projection might be accurate. Projection models based in if-then scenarios are effectively saying, "If the following happens, then the future will look like this." If the assumed conditional does not eventually happen, the result will usually be different from the projected result.

Population projection models can be checked for accuracy after the fact by comparing the estimated numbers with actual, directly observed numbers at a future date. Figure 3 is a graph of just such a comparison of predicted versus observed values, involving a researcher's (Kitov, 2005) model of the relationship between relative mean income and work experience in the United States. The red squares represent the data points predicted by the model, while the blue circles represent the actual data points.

Some projection models are more experimental, in that predicted values have not been repeatedly checked for accuracy against actual data after the fact. Models that involve several variables that have not been checked in this way in the past have a higher likelihood of being inaccurate due to this lack of data, representing an incomplete understanding of the interactive relationships of all variables included in the set of equations used in the modeling. Such population-change models represent well-educated "best guesses" about the future. These models are researchers' best determinations of the most likely future scenarios, given their expertise and many years of experience working with demographic data. Demographers must be explicit, describing in detail the nature of the models they use. Knowledgeable readers can judge for themselves the expertise of the modeler and the accuracy of the resulting projections.

Issues

Some Challenges in Data Collection & Definition. As in any research, especially studies involving human beings, the collection and classification of data presents a number of challenges. What is it we are observing? How often and how well are we observing it? How can we measure what we are observing? These are the issues surrounding any research problem.

Precisely defining what we are measuring and how we are measuring it is the single most important task in research. We often find, during the task of exact definition, that our assumptions about the world are continually challenged. In demographic research, for example, one of the most controversial issues is the concept of "race" and its definition. In many countries today, many persons are of mixed ancestry.

A number of years ago, in the United States and other countries, census officials began allowing respondents to "self-identify" their race rather than having a census taker identify their race on the basis of ambiguous characteristics, such as skin color. Census researchers devised categories of race that they thought should be mutually exclusive, meaning that one category did not overlap with another category. However, people of mixed ancestry repeatedly self-identified with one or more racial categories, or they failed to respond to the question at all. Thus, census researchers began investigating current definitions and data-collection methods and considering what adjustments could be made to count persons who self-identify as multiracial.

Table 1 below, from the US Census Quality Survey released in 2003, demonstrates the complexities inherent in asking people to self-identify their race. The respondents were asked to self-identify twice. The first time, they were allowed to choose more than one race; the second time, they could only choose one (Snipp & Lott, 2003).

The study by Snipp and Lott (2003) focused on American Indians and Alaska natives (AIAN) and their responses to questions about race. However, their research showed that the complex issue of racial categories affects not only those of American Indian descent but others as well. This study demonstrates how a shift in the definition of a basic demographic characteristic, such as viewing race in multiracial rather than monoracial terms, can have profound implications for fundamental methods of classification and tabulation of demographic data.

There are many examples of challenges in developing methods that group and aggregate data, exactly because populations are constantly changing. Yet data need to be compatible across years. This is particularly problematic in cohort grouping. A cohort is a group of people who have common characteristics, but the term is used almost exclusively among demographers to indicate groups of persons around the same age. Conversions or translations of data between differing data sets are often necessary to group age cohorts for investigation. This is particularly problematic as lifespans become longer on a global basis, as there exist few data from the past that detail the life course of people of advanced age. Thus, new demographic models are being developed to better aggregate age-related data to build age-specific models of population mortality (see for example Goldstein, & Wachter, 2006).

Conclusion

Given the pace of population growth and change, demography is a vital field of research that is changing as rapidly as the population itself. Demographers collect statistical data to analyze the size and composition of populations, studying how, why, and under what circumstances people are born, die, and move from place to place. Dynamic models of population-change processes not only answer questions about what is happening today but also give us a glimpse of possible future scenarios for population growth and change. Population change and social change are so interrelated that it seems quite impossible to disassociate the two.

Terms & Concepts

Census: An enumeration and description of people in a given area at a given point in time.

Crude Birth Rate: The annual number of live births expressed as a ratio per 1,000 persons in a specified population.

Crude Death Rate: The annual number of deaths expressed as a ratio per 1,000 persons in a specified population.

Demography: The study of human population change.

Demographic Variables: Any characteristic, attribute, or property of people or collectives of people that can vary between persons or collectives.

Descriptive Statistics: The use of arithmetical summary equations to describe demographic categories of the characteristics, attributes, or properties of specified populations.

Fecundity: The maximum possible number of births per woman in a specified population.

Fertility: The variable frequency of childbearing in a specified population.

Infant Mortality Rate: The number of annual deaths of children under one year of age expressed as a ratio per 1,000 live births per year.

Migration: The variable geographic movement of persons into and out of a specified population.

Mortality: The variable frequency of death in a specified population.

Net Migration Rate: The difference between the number of persons gained due to immigration and those lost due emigration, expressed as a ratio per 1,000 persons in the population.

Population Estimates: Estimated population numbers calculated using statistical inference equations.

Population Growth Rate: The net gain (or loss) of numbers of persons in a specified population due to changes in population processes such as birth, death, and migration.

Population Projections: The projected future numbers and characteristics of persons in a specified population, based on observed data over past periods of time and calculated using mathematical theory and equations.

Population Pyramid: A graphic representation of characteristics in a specified population, using horizontal bar graphs to represent different age groups.

Statistics: An extension of arithmetical and mathematical methods to assemble, classify, and tabulate numerically based data.

Bibliography

Bickel, P., Klaassen, C., Ritov, Y, & Wellner, J. (1993.) Efficient and adaptive estimation for semiparametric models. Baltimore, MD: Johns Hopkins University Press.

Bollen, K., Glanville, J., & Stecklov, G. (2007). Socioeconomic status, permanent income, and fertility: A latent-variable approach. Population Studies, 61(1), 15–34. Retrieved August 3, 2008 from EBSCO Online Database SocINDEX with Full Text. http://search.ebscohost.com/login.aspx?direct=true&db=sih&AN=24078160&site=ehost-live

Boulakia, J. (1971). Ibn Khaldûn: A fourteenth-century economist. The Journal of Political Economy, 79(5), 1105-1118.

Goldstein, J., & Wachter, K. (2006). Relationships between period and cohort life expectancy: Gaps and lags. Population Studies, 60(3) , 257-269. Retrieved August 3, 2008 from EBSCO Online Database SocINDEX with Full Text: http:// search.ebscohost.com/login.aspx?direct=true&db=sih&A N=22651109&site=ehost-live

Haunsperger, D., Hodge, J., Jones, M., & Saari, D. (2017). Mathematics for social and behavioral sciences. Journal of Commerce & Management Thought, 8(1), 10–17. Retrieved October 23, 2018 from EBSCO Online Database Business Source Ultimate http://search.ebscohost.com/login.aspx?direct=true&db=bsu&AN=120648532&site=ehost-live&scope=site

Haupt, A. & Kane, T. (2004). Population Handbook (5th ed.). Washington, DC: Population Reference Bureau.

Kent, M. & Haub, C. (2005). Global demographic divide. Population Bulletin, 60(4) , 1-23.

Kitov, I. (2005). A model for microeconomic and macroeconomic development. Society for the Study of Economic Inequality Working Paper Series (ECINEQ WP 2005 - 05). Retrieved August 1, 2008 from the RePEc database: http://ideas.repec.org/s/inq/inqwps.html

Koenig, M., Acharya, R., Singh, S., & Roy, T. (2006). Do current measurement approaches underestimate levels of unwanted childbearing? Evidence from rural India. Population Studies, 60(3), 243-256. Retrieved August 3, 2008 from SocINDEX with Full Text database. http://search.ebscohost.com/login.aspx?direct=true&db=sih&A N=22651107&site=ehost-live

Korotayev, A., Malkov, A., & Khaltourina, D. (2006). Introduction to social macrodynamics: Secular cycles and millennial trends. Moscow, RU: URSS.

Kreager, P. (2011). The challenge of compositional demography. Asian Population Studies, 7(2), 85–88. Retrieved November 8, 2013, from EBSCO Online Database SocINDEX with Full Text. http://search.ebscohost.com/login.aspx?direct=true&db=sih&AN=61205245&site=ehost-live

Maitland, F. (1987). Domesday book and beyond: Three essays in the early history of England. Cambridge: Cambridge University Press.

Matthews, S. A., & Parker, D. M. (2013). Progress in spatial demography. Demographic Research, 28, 271–311. Retrieved November 8, 2013, from EBSCO Online Database SocINDEX with Full Text. http://search.ebscohost. com/login.aspx?direct=true&db=sih&AN=91534653&site=ehost-live

Pearl, R. (1939). The natural history of population. New York, NY: Oxford University Press.

Pool, I. (2016). The seminal relationship between demography and sociology. New Zealand Sociology, 31(3), 146–167. Retrieved October 23, 2018 from EBSCO Online Database Sociology Source Ultimate. http://search.ebscohost.com/login.aspx?direct=true&db=sxi&AN=118447865&site=ehost-live&scope=site

Snipp, C. & Lott, T. (2003, August). Population shifts and demographic methods. Proceedings of the 2003 Joint Statistical Meetings, Section on Government Statistics. Washington, D.C. American Statistical Association, 3970–3981. Retrieved August 1, 2008, from http://www.amstat.org/Sections/Srms/Proceedings/y2003/Files/JSM2003-000683.pdf

Tuljapurkar, S. (2011). Demography as the human story. Population & Development Review, 37(1), 166–171. Retrieved November 8, 2013, from EBSCO Online Database SocINDEX with Full Text. http://search.ebscohost. com/login.aspx?direct=true&db=sih&AN=59205638&site=ehost-live

Weinstein, J. & Pillai, V. (2000). Demography: The science of population. New York, NY: Allyn & Bacon.

Uhlenberg, P. (2013). Demography is not destiny: The challenges and opportunities of global population aging. Generations, 37(1), 12–18. Retrieved November 8, 2013, from EBSCO Online Database SocINDEX with Full Text. http://search.ebscohost.com/login.aspx?direct=true&db=sih&AN=86024530&site=ehost-live

US Census Bureau. (2008, June). World population information. Retrieved July 31, 2008, from http://www.census.gov/ipc/www/idb/worldpopinfo.html

Suggested Reading

Alho, J. & Spencer, B. (2005). Statistical demography and forecasting. New York, NY: Springer.

Caselli,G., Vallin, J., & Wunsch, G. (2006). Demography: Analysis and synthesis. New York, NY: Elsevier.

Porter, J. (2011). Context, location, and space: The continued development of our "geo-sociological" imaginations. American Sociologist, 42(4), 288–302. Retrieved November 8, 2013, from EBSCO Online Database SocINDEX with Full Text. http://search.ebscohost.com/ login.aspx?direct=true&db=sih&AN=67032122&site=ehost-live

Preston, S., Heuveline, M. & Guillot, M. (2001). Demography: Measuring and modeling. London, UK: Blackwell Publishing.

Wang, F., Cai, Y., Shen, K., & Gietel-Basten, S. (2018). Is demography just a numerical exercise? Numbers, politics, and legacies of China’s one-child policy. Demography, 55(2), 693–719. Retrieved October 23, 2018 from EBSCO Online Database Business Source Ultimate. http://search.ebscohost.com/login.aspx?direct=true&db=bsu&AN=129020979&site=ehost-live&scope=site

Weeks, J. (2007). Population: An introduction to concepts and issues (10th ed.). New York, NY: Wadsworth Publishing. [RT1]1 hr 45 min