Chapter 4
the Census of Canada
1666 originally a door-to-door enumeration a series of wars between France and England ended with France ceding Canada to England in 1763, and the British undertook census on an irregular but fairly consistent basis censuses were then made regular in 1867 Canada began using sampling in 1941 in 1971, Canada made it so a census would be taken every five years the National Household Survey was sent out to 30% of households in 2011
Demographic Surveys in the US
American Community Survey (ACS) Current Population Survey (CPS) Survey on Income and Program Participation (SIPP) American Housing Survey National Survey of Family Growth (NSFG) National Health Interview Survey (NHIS)
American Community Survey
An ongoing "continuous measurement" survey conducted by the U.S. Census Bureau to track the detailed population characteristics of every American community designed to allow the long form to be dropped from the decennial census in 2010 modeled after the *Current Population Survey (CPS)*
European Surveys
Family and Fertility Surveys (FSS) Generations and Gender Program European Social Survey (ESS)
Canadian Surveys
Labor Force Survey (LFS) General Social Survey
Mexican Surveys
National Survey of Occupation and Employment (ENOE)
First Censuses
US: 1790 - censuses had been around in colonial times before US independence was established England: 1801 India: 1881 Canada: 1661 most developing countries don't conduct censuses in the past, taking a census has sometimes been seen as a violation of rights due to taking statistics on age, sex, religion, etc., so it was frequently met with resistance racial questions were met with particular hostility and frustration in the US, especially as the population becomes more and more interracial
Demographic and Health Surveys (DHS)
USAID. The Demographic and Health Surveys (DHS) program has collected, analyzed, and disseminated accurate and representative data on population, health, HIV, and nutrition through more than 300 surveys in over 90 countries a complementary set of surveys has been conducted in poorer countries
European Social Survey (ESS)
a cross-national survey that has been conducted every two years across Europe since 2001 measures attitudes, beliefs, and behavior patterns also measures the demographics of populations in more than thirty European nations
Population Registers
a list of all people in a country on which are recorded all vital events for each individual, typically birth, death, marriage, divorce, and change of residence
Generations and Gender Program
a longitudinal survey of 18-79-year-olds in 19 countries gathering data on a broad array of topics including fertility, partnership, the transition to adulthood, economic activity, care duties, and attitudes
Sample Surveys
a method of collecting data by obtaining information from a sample of the total population, rather than by a complete census *key:* need a representative sample (random sample) *random sample* each individual has the same chance to be selected small sample can represent large population (1 - 2,000 for most political polls) low cost can be conducted very often
Demographic Analysis (DA)
a method of evaluating the accuracy of a census by estimating the demographic components of change since the previous census and comparing it with the new census count uses the *demographic balancing equation* to estimate what the population at the latest census should have been, and then compares that number to the actual count
Current Population Survey (CPS)
a monthly survey of a sample of US households done by the US Census Bureau it measures employment, unemployment, the labor force, and other characteristics of the US population since 1943
Georeferenced
a piece of information that includes some form of geographic identification such as precise latitude-longitude coordinates, a street address, ZIP code, census tract, county, state, or country
Public Use of Microdata Sample (PUMS) Data
a random sample of individual census records that have been stripped of personally identifying information this makes national census data of North American countries (US, Canada, and Mexico) available to the public through the Internet
GIS and the Census
after WWII, the Census Bureau sponsored the development of the first computer designed for mass data processing (UNIVAC-I) therefore, a device was then needed to tabulate data from census forms: FOSIDC (film optical sensing device for inputer to computers) - first used for the 1960 census *Dual Independent Map Encoding (DIME)* - first step towards computer mapping - each piece of data was coded in a way that could be matched electronically to a place on a map - 1980 *Topologically Integrated Geographic Encoding and Referencing (TIGER)* - early 1990s - Census Bureau reconfigured its geographic coding of data - computers were advancing and improving - the TIGER files helped spawn the now-booming GIS industry
Nonsampling Error
an error that occurs in the enumeration process as a result of missing people who should be counted, counting people more than once, respondents providing inaccurate information, or recording or processing information inaccurately
Content Error
an inaccuracy in the data obtained in a census; possibly an error in reporting, editing, or tabulating people might not understand what's being asked, information might not be accurately recorded, glitches in processing, etc.
Census of Population
an official enumeration of an entire population, usually with details as to age, sex, occupation, and other population characteristics; defined by the United Nations 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" census = "assessing" or "taxing" originally, just to count males for purposes of taxes changed so that everyone had to record the number of people in their household statistic = "facts about a state" first US census: 1790 by the latter part of the nineteenth century, the statistical approach to understanding business and government affairs had started to take root in the Western world
Geodemographics
analysis of demographic data that have been georeferenced to specific locations
Sources of Demographic Data
census of population vital statistics population registers sample surveys administrative data historical data *for class* there's three types of data: - census - vital statistics - surveys
Survey on Income and Program Participation (SIPP)
companion of the CPS gathers detailed data on sources of income and wealth, disability, and the extent to which household members participate in government assistance programs uses a rotating panel of more than 40,000 household that are queried several times over a two- to four-year period
Geographic Information Systems (GIS)
computer-based system that allows the user to combine maps with data that refer to particular places on those maps and then to analyze those data and display the results as thematic maps or some other graphic format
Family and Fertility Surveys (FSS)
conducted in 23 European nations during the 1990s
Historical Data
data derived from sources such as early censuses, genealogies, family reconstitution, grave sites, and archaeological findings
Vital Statistics
data referring to the so-called vital events of life, especially birth and death, but usually also including marriage, divorce, and sometimes abortion these records are usually kept by public health agencies like the NCHS major source of information on the population processes of births and deaths accomplished instead by *population registers* in some countries, and by *sample surveys* in most developing countries originally began as a chore of the church; priests often recorded baptisms, marriages, and deaths people were originally skeptical about what the government would do with this information, so it wasn't until the middle of the nineteenth century that civil registration of births and deaths became compulsory in European and North American government--not until 1900 for the US although most nations have a system of birth and death registration that is separate from census activities, dozens of countries (mostly in Europe), maintain *population registers*
Two Principal Methods to Measure Coverage Error
demographic analysis (DA) dual-system estimation (DSE)
Administrative Data
demographic information derived from administrative records, including tax returns, utility records, school enrollment, and participation in government programs
No Census is Perfect
error of coverage misreporting
Sampling Error
error that occurs in sampling due to the fact that a sample is rarely identical in every way to the population from which it was drawn
National Survey of Family Growth
generates data about fertility and reproductive health every five years or so
American Housing Survey
generates important data on mobility and migration patterns in the US
Other Useful Data Resources
immigration statistics published by the INS Bureau of Labor Statistics to understand the causes and consequences of an demographic fact, we usually not only need demographic data on birth, death, migration, race, sex, education, etc. of a population, we also need to look at other data such as *labor market data, economic data*
Four Features of Census
individual universal simultaneous periodic
Labor Force Survey (LFS)
initiated in 1945 produces data on the labor force also gathers data on most of the core sociodemographic characteristics of people in each sampled household; provides a continuous measure of population trends in Canada on a rotating panel of 56,000 households
Can the enumerator collect many detailed information on each individual?
kind of a "middle" form cannot reach everybody and ask every question
National Survey of Occupation of Employment (ENOE)
large sample of household (120,000) undertaken three times a year designed to be representative of the entire country goal is to provide a way of regularly measuring and monitoring the social and economic characteristics of the population beyond just data on current employment
Possible Errors in the Enumeration Process
nonsampling error - coverage error - content error sampling error
National Health Interview Survey (NHIS)
obtains data on health and disability
the Census of Mexico
recorded census-like data goes back to 1116, and the Aztec empire also kept count of the population for tax purposes when Mexico was under Spain, Spain began conducting a census of Mexico in 1790 when Mexico gained independence of Spain in 1821, it began conducting a census in 1895 every ten years somewhat different set of questions and sampling strategy than the US and Canada
Spatial Demography
represents the application of spatial concepts and statistics to demographic phenomena any analysis of population data that takes location into account by using georeferenced information demography is evolving from being a primarily spatially *aware* science to an increasingly more spatially *analytic* science
General Social Survey
sample of about 25,000 voluntary respondents each survey has a different set of in-depth topics designed to elicit detailed data about various aspects of life in Canada
Who is included in the census?
several ways to answer; the two extremes are: *de facto population* *de jure population* most countries (including the US, Canada, and Mexico) have adopted a concept that lies somewhere between these extremes and include . people on the basis of *usual residence*
Misreporting
some people get counted twice, some aren't counted, etc.
Redistricting
spatially redefining US congressional districts (geographic areas) represented by each seat in Congress
Demographic Surveillance System (DSS)
the INDEPTH Network was created in 1998 to provide a way of tracking the lives of people in specific "sentinel" areas of sub-Saharan Africa (and to a lesser extent South Asia) by working with individual countries to select one or two defined geographic regions that are representative of a larger population a census is conducted in that region, and then subsequent demographic changes are continuously measured by keeping track of all births, deaths, migration, and related characteristics of the population
Coverage Error
the combination of undercount (the percentage of a particular group or total population that is inadvertently not . counted in a census) and overcount (people who are counted more than once in the census) also called *net census undercount* there are several ways to measure and adjust for undercount, but it becomes more complicated (and political) when is a *differential undercount*
Usual Residence
the concept of including people in the census on the basis of where they usually reside
Spatial Autocorrelation
the concept that everything is related to everything else, but near things are more related than distant things based on Waldo Tobler's First Law of Geography
Demographic Balancing Equation
the formula that shows that the population at time 2 is equal to the population at time 1, plus the births between time 1 and 2, minus the deaths between time 1 and 2, plus the in-migrants between time 1 and 2, minus the out-migrants between time 1 and 2
Historical Sources
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 family genealogies the results of these labors can be of considerable importance in testing our notions about how the world used to work by quantifying our knowledge of past patterns of demographic events, we are also better able to interpret historical events in a meaningful fashion
De Facto Population
the people actually in a given territory on the census day *de facto residence* the place him/her is found at the "census moment"
De Jure Population
the people who legally "belong" in a given area whether or not they are there on census day *de jure residence* usual place of residence
Intercensal Year
the period between the taking of censuses by combining vital statistics with census data, population estimates between censuses can be made
Differential Undercount
the situation that occurs in a census when some groups of people are more likely to be underenumerated than other groups usually this means that racial/ethnic minorities are less likely to be included in the census count than whites
Apportionment
the use of census data to determine the number of seats in the US Congress that will be allocated to each state
Two Major Difficulties with Using Data Collected in the Census, by the Vital Statistics Registration System or from Administrative Records
they are usually collected for purposes other than demographic analysis and thus do not necessarily reflect the theoretical concerns of demography *AND* they are collected by many different people using many different methods and may be prone to numerous kinds of error the principle limitation is that they provide less extensive geographic coverage than a census or system of vital registration
Census
to get size, we need *count* to get age, sex, or racial composition, we need *information* (?) to get distribution, we need *location* we need to collect all these data at one specified moment *("census moment")* - population is very dynamic and changes every second April 15, 2019 is the Census day in the US
What information should be collected on characteristics of individuals?
we only need the most important characteristics - age - sex - marital status - ethnicity mostly everyone gets the "short" form of the census there used to be 15% that would get the "long" form that includes employment, income, housing, immigration background, education, etc. but that has since been replaced by the *American Community Survey (ACS)*
Administrative Records
with respect to migration, this refers to forms filled out for each person entering the U.S. from abroad that are then collected and tabulated by the U.S. Citizenship and Immigration Service