Population and Society

Lecture 2

Dr Enock Ngome

GEC 278: Sources of Demographic Data

 Introduction
• In order to understand sources of Demographic data, we
need to revisit the definition of demography/
• What is demography?
• The scientific study of human population
• Demography is the scientific study of the size,
composition, and spatial distribution of human
populations
• It investigates changes in population size,
composition, and distribution, resulting from fertility,
mortality, and migration
Introduction
• How does demography connect the dots?

• Nearly everything is connected to demography

• The relationship of population to resources

• The relationship of population to political and

social dynamics
 Concerns of demography

• Population size
• Population growth or decline
• Population processes/components
• Population distribution
• Population structure
• Population characteristics
Primary demographic questions

• How large (or small) is the population (Size)?

• How is the population composed, in terms of age, sex,
race, marital status, and so forth?
• What are the characteristics of the population?
• How is the population distributed spatially?
• Populations are not randomly distributed in space
• How population changes happen over time (trends)?
 Primary demographic questions

Answers to these questions

• These demographic questions are answered in terms

of the three demographic processes (components of
demographic change)
• Fertility
• Mortality
• Migration
 Demographic Data

• Before reviewing various sources of demographic

data, we need to explain certain basic concepts used
to explain data such as:

• Variables
• Observations
 Variables and observations
• Variables
• Characteristics that can change values from case to case.
For example: gender, age, income, political party
affiliation...
• Observations (cases)
• Refers to the entity from which data are collected – Also
known as ”unit of analysis”. For example: individuals,
households, states, countries...
 Variables
• Variable: a characteristic/phenomenon whose
value varies (changes) from case to case, and is
empirically quantifiable
• Dependent variable: a variable whose variation
depends on another variable
• Independent variable: a variable whose variation
produces (“causes”) variation in another variable
 Observations
• Observations (cases) are collected information used
to test hypotheses
• Decide how variables will be measured and how
cases will be selected and tested
• Measure social reality: collect numerical data
• Information can be organized in databases
• Variables as columns
• Observations as rows
Demographic models

We may divide Demographic models into TWO;

depending on how we use various types of variables.
• Formal demography
• Population studies (Has two parts)
 Demographic models

Demographic Model Independent Dependent Variable

Variable

Formal Demography Demographic Demographic

Population Studies I Non-Demographic Demographic

Population Studies II Demographic Non-Demographic

Introduction to Demographic Data Sources

• Demographic, socio-economic and any other vital

statistics are useful to social development:
• Social security, programmes, epidemiology, public health
and community services data are available from
numerous sources.
• Some sources are more useful than others, thus
applicability of data needs consideration
 Purposes of Data Collection
• The population size and composition have important
social, economic, political and environmental
implications for the future.

• The population size is defined by the number of

individuals living in designated regions of the world
(such as a particular country for example) at a certain
time while the population composition represents
their statistical distribution according to age, sex,
marital status, education, occupation, etc.
 Purposes of Data Collection Cont...

• There are two main ways for collecting information

about population:
• (i) by enumeration at a point of time, i. e. information is
obtained from population censuses and demographic
surveys

• (ii) by recording vital events (such as births, deaths,

marriages, migration, etc.) as they occur over a period of
time, i.e. information is collected through what is known
as the vital registration system. In social research you may
record observations as they occur.
Purposes of Data Collection Cont...

• Theoretically with the current advance in

information technologies, it should be possible to
maintain an up-to-day record of population size and
composition.
• In fact, because of human fallibility and delays in
recording of vital events it is almost impossible at
this stage to ensure any complete counts of
population.
Purpose of Review of Sources of Data

• To make you understand the way data are collected,

the errors and biases the data may likely to contain.
• Very few data are collected primarily for demographic or
any specific discipline purpose.
• Most of the data are produced for, or as a by product of,
administrative exercises carried out by or under the
control of governments or international agencies.
• Scientists such as Epidemiologists, demographers, sociologists
etc. normally has very little control over the precise way in which
epidemiologic and demographic data are collected, aggregated
and presented.
Administrative Exercises

• Examples of administrative exercises

• Vital registration systems

• Migration controls
• Public Health campaigns
• Censuses
• Population control programmes
• Special surveys etc.
Administrative Exercises Cont...
• The existence of the vital registration and migration
controls are usually for legal purposes.

• They assist in the production of births and deaths

certificates, passports, identity cards (Omang),
work permits, residence permits, residence
qualifications and citizenship.
Population Control Programmes or
Health Programmes
• Such programmes collect data to monitor progress as well
as evaluate the impact of their programmes.

• Useful for identification of an epidemic

Censuses
• The largest single data gathering exercise carried out
in a country.
• Involves which will be responsible for setting up of a
large administrative body the whole exercise of
planning and organising the whole exercise of
processing completed questionnaires, tabulating and
publication of results. The body is also involved in the
collection and formulation questionnaires.
• It is a very costly exercise which involves great
administrative effort and disruption. Although it is a
very costly exercise, it’s advantages far outweigh its
disadvantages. The advantaged spread over a wide
variety of activities both in the government and the
private sector.
Purpose of a Census
• Primarily meant to aid planning for the provision of
social infrastructure and basic services for the
communities, i.e. for the provision of schools,
health facilities, power, water and roads.
• In Botswana, it is also used to determine the level of
electoral representation.
• It also provides data for rational distribution of
resources, and or, the pattern of resource allocation
from the central government to the local councils as in
Botswana.
 Characteristics of a Census
• It involves a complete recording of the entire population
within a geographical well defined area (no sampling)

• Each person must be enumerated separately

• It must have a legal basis making it compulsory to be

included and to provide information requested.

• It must relate to a single point in time.

• It should be held at least at ten year intervals (U.N.

recommendation)
Design of a Census
• Key Concepts:
1.De Facto vs. De Jure Enumeration:
1. De Facto Enumeration:
2. De Jure Enumeration:
3. Choosing the Appropriate Method:
2.Questionnaire Design and Content:
1. Clarity and Simplicity:
2. Cultural Sensitivity:
3. Relevance and Necessity:
3.Training and Deployment of Enumerators:
1. Comprehensive Training:
2. Enumerator Selection:
3. Supervision and Support:
4.Use of Technology in Data Collection and Processing:
1. Digital Data Collection:
2. Data Security and Privacy:
3. Real-Time Monitoring and Analysis:
5.Public Awareness and Engagement:
1. Awareness Campaigns:
2. Community Involvement:
3. Addressing Concerns and Misinformation:
 Design of a Census Cont….
• The design must take into consideration certain crucial points;

• The literacy level of the population to be enumerated

• Whether to be completed by enumerators or by respondents
themselves
• And the number of questions to be included.

• The number of questions to be included also depends on the

country.
• Most of the developing countries do have a poor or a non-existent
vital registration system and so it would be wise to include questions
on deaths and births in the census for such countries.
• Indirect questions on the deaths and births are crucial to the
demographers since they may be used to estimate the levels of
fertility and mortality for countries that have poor or non-existent
vital registration.
Errors in a census
• The larger the questionnaires the more likely there
would be mistakes. This would also contribute to the
administrative complexity of the census exercise.
• Errors in census taking are inevitably very great.
• The large part of the organisation and design is
concerned with the minimisation of these errors which
are grossly underestimated especially in
underdeveloped countries.

• Errors of this nature fall into two main categories

namely incomplete coverage and response errors.
• Incomplete coverage
• Response errors
Types of Errors Cont...
• Incomplete Coverage
• Some areas may be missed probably due to an outdated
frame that was used, or by an enumerator.
• Certain groups may be missed such as the nomads, the
homeless (Bo bashi), fishermen, students, infants etc.
 Types of Errors Cont...
• Response Errors
• The enumerator recording the answer given to him
wrongly
• The respondent misunderstanding the question (not
knowing the question being inappropriate in the cultural
context)
• Deliberate wrong answer may be due to pride, shame,
fear or mischievousness.
• Data processing errors
• Data Processing of the census consumes a lot of time.
• It involves the labourious coding and can introduce substantial
errors and biases if not carried out carefully and meticulously.
• Errors can be introduced during data entry but these can be
minimised by double entering and automatic verification
procedures.
Types of Errors Cont...
• A properly run and well financed census would
hopefully have final tabulations published after two
years, though publication of all volumes take
longer, sometimes after five or six years.

• It is easy to make a mess of things in the census

and even the bets run census would still contain a
myriad of errors and biases.
Vital Registration System (VRS)
• Main objectives of the vital registration system are
essentially administrative and legal.
• Provides the main source of information on births,
marriages and deaths, and sometimes also on
migration.
• VRSs are either non-existent or more frequently
seriously incomplete in developing countries.
(sometimes NOT not even used for demographic
purposes)
What does it involve?
• It involves only a few, simple questions such:
• Date of event,
• age,
• sex,
• marital status,
• occupation,
• cause of death in case of death etc.
Errors in the VRSs
• VRSs are very effective only in a few countries due
to many reasons.
• The combination of non-registration and incomplete
aggregation in the system produces deficient
demographic data.
Errors in the VRSs
• Reasons for such errors may be that:
• If the need to register for individuals is not there, then few
will register regardless of the legal requirements.
• Even if the need is there, there may other issues that make
it difficult for individuals to register (lets discuss this in class)
• In addition, even if the need is there, it does not mean that
demographic tabulations produced by the system are
complete as well.
• The administrative structure reaching right down to the local
level is usually complex and difficult to manage.
• A poor structure and poorly motivated, poorly paid,
overstretched administrative system will lead to errors and
omissions in the data collection system.
Distribution of errors
• Omissions are usually not randomly distributed
and tend to concentrate amongst certain groups,
• i.e. registration tends to be better in urban areas than
rural areas, why? Discuss this in class.
• Is often sex biased, why? Discuss this in class.
• registration of people with very better socio-economic
status tends to be better than those with very low socio-
economic status etc., why? Discuss this in class.
• Generally, information covered in vital registration
is likely to be poorer than in a census.
Sample Surveys

• What is a sample survey:

• This involves taking a sample from of the population to
collect data about the population from which the
sample is chosen.
• The advantage with sampling is that it reduces
effort required, and hence the cost compared to
the census. By doing this, it introduces the problem
of sampling error. To reduce this problem, sampling
has to be carried out randomly from the suitable
‘frame’.
Advantages of a sample survey
• Cheap to run
• It reduces effort required, and hence the cost compared
to the census.
• They can organised and executed relatively quickly
• They can gather more detail than the census
including attitudinal questions,
• Interviewers can be intensively trained
Disadvantage of a sample survey
• By reducing effort through less coverage, it introduces the
problem of sampling error.
• How do you reduce the problem of sampling error? When
carrying out sampling, randomization should be
introduced
• This is simply sampling (or selecting the sample form the
population) from the suitable ‘frame’ has to be carried out.
• This ensures that everyone in the population is given a chance to
be included in the sample.
Other errors in a survey
• Other errors are just similar to those that occur in
the census undertaking.
• Response errors occur just like in the census and these
are:
• Errors that have to do with the enumerator
• Errors that have to do with the respondents
• And data processing errors
Sampling bias
• Occurs when a sample is collected in such a way
that some members of the population are less likely
to be included than others.
• For example: A voluntary television poll is an example of
a biased sample. Since it is voluntary, only those with
strong views are likely to call or text in to vote.
Furthermore, only those watching the particular station
at the time the poll is given will participate. In this case,
the entire segment of the population who do not watch
that particular station will be left out of the sample.
Types of Sample Surveys

• Census type enquiries

• Retrospective surveys
• Prospective surveys
• Dual record surveys
Census type enquiries
• May replace a full census.
• Commonly used to supplement the census by
asking a wider range of questions of a sample of
the enumerated population.
• Sometimes carried out to evaluate the accuracy of
the census data
• To determine magnitude of errors in coverage etc.
Retrospective surveys
• In such surveys, events which happened in the past
are asked.
• In underdeveloped countries, these are primarily
designed to gather information that the
registration system would normally provide.
• For demographers, these are important and the answers
from this type of survey provide ground for estimation
of mortality and fertility levels.
• The idea behind this survey is that it contains
simple yet specific questions which can be
answered accurately by almost everyone.
Prospective surveys
• This involves first carrying out a normal census type
survey which acts as a baseline, followed by
repeated visits to check and record subsequent
events.
• It can be costly.
• A cohort survey are an example of a prospective
survey
 Dual record surveys

• This is largely fallen from favour in the contemporary

world
• It involves running two statistically independent
surveys covering the same population, and
matching results.
• The disadvantage with this type of survey is;
• the impossibility of running two surveys which are truly
independent and
• the difficulties involved in the matching process on which
the method crucially depends.
Population Registers
• Its a mechanism for the continuous recording of selected
information pertaining to each member of the resident
population of a country or area, making it possible to
determine up-to-date information about the size and
characteristics of the population at selected points in time.

• Because of the nature of a population register, its organization,

as well as its operation, should have a legal basis.

• From the beginning, they start with a base consisting of an

inventory of the inhabitants of an area and their
characteristics, such as date of birth, sex, marital status, place
of birth, place of residence, citizenship and language.
 Population Registers Cont…
• An identification number is provided for each entity (such
as Omang….Population Register does not exist in
Botswana) to assist in locating a record for that particular
person, household or family.

• May or can (Depending on need) contain other socio-

economic data, such as occupation or education.

• It is or should be updated by births, deaths, marriages

and divorces, which are part of the civil registration
system of the country.

• Should also be updated by migration.

 Population Registers Cont…
• It is the result of a continuous process, in which
notifications of certain events, which may have been
recorded originally in different administrative
systems, are automatically linked to a population
register on a current basis.

• The method and sources of updating should cover all

changes so that the characteristics of individuals in
the register remain current.
 Population Registers Cont…
• Main Uses
• The main administrative functions of population registers
are to provide reliable information for the various
purposes of government: particularly for:
• programme planning,
• budgeting and taxation;
• for issuing unique personal identification numbers;
• for establishing the eligibility of individuals for voting, education,
health, military service, social insurance and welfare and the
pension system;
• and for police and judicial references.
Population Registers Cont…
• Main Uses Cont…
• Useful for population estimation, census planning,
census evaluation and for sampling frame of household
surveys.
• Used to produce census-typed tables every five or 10
years in place of conducting regular census operations.
• If complete, they can produce data on both internal
and international migration (recording of changes of
residence as well as the recording of international
arrivals and departures).
Population Registers Cont…
• Main Uses Cont…
• They represent one of the independent sources of data
with which the population census results can be
compared as part of the process of evaluating the
accuracy of the latter.
• Comparison can be made between aggregates compiled
from the two sources or by one-to-one matching of the
corresponding records of the individuals so as to correct
either the census or the population register.
 Population Registers Cont…
• Coordination with other Systems
• Some countries have separate agencies for the population
register for civil registration and for vital statistics.
• Recommendation: In such a situation, births, deaths,
marriages, divorces and other vital events recorded by the
civil registration system be used as the base for updating
the population register.
• Opportunity: both programmes can share and compare
information while meeting their own separate objectives.
• Information on vital events should be transmitted to the
agency responsible for vital statistics.
Population Registers Cont…
• Coordination with other Systems:
• The production of vital statistics is the responsibility of the
population registration agency in some countries.
• In such situations: this agency is concerned not only with
the registration of various vital events and their changes
but also with the updating of the register and the
compilation of vital statistics.
• Examples: The Norwegian and Bulgarian population
registers are examples of this type of arrangement and are
under the administration of the respective statistical
authorities of the countries.
• Preferred situation: Where one agency is responsible for
civil registration, the maintenance of the population
register and the production of vital statistics.
Population Registers Cont…
• Coordination with other Systems
• What happens when different agencies are
responsible for different functions?:
• Might result in the production of different series of vital
statistics which are inconsistent.
• The coordination of the production of vital statistics is
possible at two levels: the data-collection level and the
data-processing level.
• At the data-collection level, one form, such as a multi-part
form, is used to record the data and copies are sent to
each organization for entry into its system. Thus, the
same source documents serve as input into the respective
systems (Duplication of efforts)
 Population Registers Cont…
• Coordination with other Systems

• Experience of some countries has shown that when a

single record is used for both population register and vital
statistics purposes, the most difficult task is handling
confidential medical data on births, deaths and foetal
deaths.
• A discussion of the problem of using the same form for
registration and statistics can be found in the Handbook on
Vital Statistics Systems and Methods, using the Norway
population register as an example.
 Population Registers Cont…
• Main Uses Cont…
• Both the population register and the civil registration
system contain common data elements (Both have
personal identifying information in the database, such as
name, age or birth date, sex or place of residence)

• Record linkage between the population register and the

civil registration databases is thus a requirement.
Population Registers Cont…
• Main Uses Continued…

• The linkage then becomes a task for computer matching

since the volume of records would make any manual
approach very difficult.

• Use of unique personal identifiers simplifies the

matching process.
 Population Registers
References:
1. United Nations. (1969). Methodology and Evaluation of
Population Registers and Similar Systems in 1969 …… can
be accessed on the following
link:https://unstats.un.org/unsd/demographic/standmet
h/handbooks/Series_F84/Series_F84en.pdf
2. Handbook on Vital Statistics Systems and Methods, vol. 1,
Legal, Organizational and Technical Aspects. (United
Nations publication, Sales No.E.91.XVII.5).
Health records
• The terms medical record and health record are used
somewhat interchangeably to describe the systematic
documentation of a single patient’s medical history and care
generated by one or more encounters in any care delivery
setting. Included in this information are patient
demographics, progress notes, problems, medications, vital
signs, past medical history, immunizations, laboratory data,
and radiology reports.
• Health records form an essential part of a patient’s present
and future health care. As a written collection of
information about a patient’s health and treatment, they
are used essentially for the present and continuing care of
the patient. In addition, medical records are used in the
management and planning of health care facilities and
services, FOR MEDICAL AND SOCIAL RESEARCH AND THE
PRODUCTION OF HEALTH CARE STATISTICS.