SURVEYS

SAMPLING TECHNIQUES
SURVEY
• A survey is an examination of an aggregate of units.
• A group of students is an example of an aggregate.
• The examination usually involves counting members of the aggregate and characteristics of the members.
• Inepidemiological surveys, characteristics might include the presence of particular diseases or death (discrete), or
production parameters (continuous) suchas weight or height.

• Surveys can be undertaken on a sample of the population.

• Examples:
• A cross-sectional survey records events occurring at a particular point in time.
• An important application is estimation of prevalence of clinical disease, infection, or
seropositive animals from samples of animal population
• Alongitudinal survey records eventsover a period of time. Theseeventsmay be recorded
prospectively from the present into the future; or may be a retrospective record of past events.
• A particular type of diagnostic survey is screening.
CENSUS

a) The survey in which we investigate all animals/humans of a population is known as

census.
b) Through census, we measure the exact distribution of a variable in a population.
c) Censuses are expensive and difficult to conduct.
• For Example:

• Jasper et al. (1979) surveyed 2400 out of 2800 farms in California to determine the
prevalence of mycoplasmal mastitis in dairy cows.
Basic Terminologies
◦ Population
A set of all individuals or objects having some common observable characteristics.
◦ Sample
A sample is subset of population with which we can draw valid inferences regarding
population parameters.
◦ Parameter
Parameters are the true values of the population under study
◦ Statistics
Statistics are true values of samples.
◦ Sampling:
A subset or part of the population is a sample from that population. The process of
selecting the sample is called sampling.
Basic Terminologies
◦ Target population:
 It is the total population about which information is required or it is population
at risk.
◦ Study Population
 It is the population from which sample is drawn. It is the actual sampling
frame, from which we randomly drew our sample.
◦ Sampling frame:
 Complete list of all sampling units called sampling frame. e.g. Complete list
of men, women, children, younger, elders called sampling frame.
◦ Sampling Units
 This forms the basis of sampling procedure. They are the smaller parts of
the population which are distinct, unambiguous and non-overlapping e.g. children
less than 5 years of age, households, schools etc. Each member of sampling frame
is sampling unit.
◦ Elementary unit:
 The study population consists of elementary units which cannot be divided
further e.g individual
Basic Terminologies
◦ Stratum:
◦ A collection of elementary units, grouped according to a common characteristic, is a stratum.

◦ Sampling fraction
◦ The sampling fraction is the ratio of sample size to study population size. Thus, if 10
individuals were chosen from 1000, the sampling fraction would be 1 %.
Sampling theory
◦ An aggregate of units can be
divided into representative
subunits, and that characteristics
of the aggregate can be
estimated from the subunits.
Sampling in Epidemiology
◦ Why?
◦ Unable to study all members of a population
◦ Reduce bias
◦ Save time and money
◦ Measurements may be better in sample than in entire population
◦ Feasibility
 When and Where sampling technique is
appropriate

◦ Vast data
– No. of units is very large-Sampling economizes money, time &
effort
◦ When utmost accuracy is not required
– suitable in those situations where 100% accuracy is not
required
◦ Where census is impossible
-- not enumerating all individuals
◦ Homogeneity
– if all the units are alike. Sampling is very easy to use
OBJECTIVE OF SAMPLING

• The objective of sampling is to provide an unbiased estimate of the

variable that is being measured in the population.
THINGS THAT CAN LEAD TO BIASED ESTIMATES

• lists of members of the frame are incomplete;

• information is obsolescent;
• segments of the frame are untraceable;
• there is lack of co-operation by some members of the frame;
• sampling procedures are not random.
 TYPES OF SAMPLING

NON-PROBABILITY PROBABILITY SAMPLING

SAMPLING • in which selection of sample is made
using a deliberate, unbiased process, so
• in which the choice of the sample that each sampling unit in a group has
is left to the investigator. an equal probability of being selected.
• The results will be unbiased and the
• The main disadvantage is that effect of sampling variation will be
‘representativeness’ cannot be minimal.
quantified.
NON-PROBABILITY
SAMPLING
NON-PROBABILITY SAMPLING
METHODS
•In non-probability sampling, researchers choose samples
according to his on choice.

• Researcher select study population according to his own will.

NON-PROBABILITY SAMPLING

Convenience
Purposive Sampling
Sampling

Judgement Snow ball

Quota
sampling sampling
CONVENIENCE SAMPLING
•It is the collection of easily accessible sampling units. In convenience sampling, it
is very unlikely that the sample will be the truly representative of the study
population.
• A non-random collection of sampling units from an undefined sampling frame
• Advantage: Convenient and easy to perform
• Disadvantage: Not statistical justification for sample
• For Example

•A university researcher wants to survey students’ opinions about campus

facilities. Instead of reaching out to a random sample of students from all
faculties, they decide to survey students in the campus library because they’re
easily accessible.
PURPOSIVE SAMPLING:
➢ Judgmental sampling:
It is one that is selected based on the knowledge of a population and the
purpose of the study. The subjects are selected because of some
characteristic. It is sampling for some specific purpose.

For example:
A researcher who is undertaking a tuberculosis test on several
individuals may be asked to take blood samples from selected samples
for titration of antibodies against bacteria and viruses. This is also error
based.
 Snowball sampling is often used for hard-to-reach or
niche populations. Initial participants help the
researcher connect with additional participants,
creating a “snowball” effect. This method is valuable for
studying specific networks but may lead to non-
SNOW BALL representative samples.

SAMPLING Example: A sociologist is studying the experiences of

migrant workers in a particular city. Since this group is
difficult to locate, the researcher interviews a few
migrant workers and then asks them to refer others they
know.
QUOTA SAMPLING

Quota sampling involves dividing the population into subgroups and

ensuring that each subgroup is represented in the sample. This method
helps make the sample more representative than pure convenience
sampling but can still introduce biases.

For example:

Equal quota of young and old people.

PROBABILITY
SAMPLING METHODS
 PROBABILITY
SAMPLING

Simple Random Systematic Stratified Cluster

Sampling Sampling Sampling Sampling
SIMPLE RANDOM SAMPLING
In which samples are selected randomly from study population.

• For Example: Through lucky draw, random table and computer etc.
• Simple Random Sampling: n=1.962Pexp (1-Pexp)
d2

• Where:
n= Required sample size

P exp = expected prevalence

d= desired absolute precision

 SIMPLE RANDOM SAMPLING
• Simple random sampling
• Advantages
• Simple process and easy to understand
• Easy calculation of means and variance

• Disadvantages
• Not most efficient method, that is, not the most precise estimate for the cost
• Requires knowledge of the complete sampling frame
• Cannot always be certain that there is an equal chance of selection
• Non respondents or refusals
SYSTEMATIC SAMPLING

•It involves selection of sampling at equal intervals, the first

individual being selected randomly.

• For Example:
• If one person in every 100 were required, then the first animal
would be selected randomly from first 100.
 SYSTEMATIC SAMPLING
• The sampling units are spaced regularly
throughout the sampling frame, e.g., every 3rd
unit would be selected

• May be used as either probability sample or not

• Not a probability sample unless the starting point is
randomly selected
• Non-random sample if the starting point is
determined by some other mechanism than chance
SYSTEMATIC SAMPLING
• Advantages
• Sampling frame does not need to be defined in advance
• Easier to implement in the field
• If there are unrecognized trends in the sample frame,
systematic sample ensure coverage of the spectrum of
units
• Disadvantages
• Variance cannot be estimated unless assumptions are
made
 SYSTEMATIC SAMPLING

• Estimate HIV prevalence in children born

during a specified period at a hospital
1. Impossible to construct sampling frame in advance
2. Select a random number between some pre-specified bounds
3. Beginning with the random number chosen, take every 5th birth
and measure for HIV infection
STRATIFIED SAMPLING

•In which sample is obtained by dividing the study population

into exclusive groups, then sampling units selected randomly from
those groups.

• For Example:
• Different ranges of herds or flock size, or different geographical
regions.
Example of stratified random
sampling
 STRATIFIED RANDOM SAMPLING

•The sampling frame comprises groups, or

strata, with certain characteristics
•A sample of units are selected from each
group or stratum
Stratified Random selection for drug trail in hypertension

Mild Moderate Severe

STRATIFIED RANDOM SAMPLING

• Advantages
• Assures that certain subgroups are represented in a sample
• Allows investigator to estimate parameters in different strata
• More precise estimates of the parameters because strata are more
homogeneous, e.g., smaller variance within strata
• Strata of interest can be sampled most intensively, e.g., groups with
greatest variance
• Administrative advantages
• Disadvantages
• Lossof precision if small number of units is sampled from strata
STRATIFIED RANDOM SAMPLING

• Assess dietary intake in adolescents

1. Define three age groups: 11-13, 14-16, 17-19
2. Stratify age groups by sex
3. Obtain list of children in this age range from schools
4. Randomly select children from each of the 6 strata until
sample size is obtained
5. Measure dietary intake
CLUSTER SAMPLING

•Cluster sampling is a sampling technique where the entire

population is divided into groups, or clusters and a random sample
of these clusters are selected.

• For Example:
• In a city of population over 2 million, we form clusters of the
target population, than select randomly from these clusters.
CLUSTER SAMPLING
• Advantages
• The entire sampling frame need not be enumerated
in advance, just the clusters once identified
• More economical in terms of resources than simple
random sampling
• Disadvantages
• Loss of precision, i.e., wider variance, but can be
accounted for with larger number of clusters
 CLUSTER SAMPLING

• Cluster sampling
• Estimate the prevalence of dental caries in
school children
1. Among the schools in the catchments area, list all of the
classrooms in each school
2. Take a simple random sample of classrooms, or cluster of
children
3. Examine all children in a cluster for dental caries
4. Estimate prevalence of caries within clusters than combine in
overall estimate, with variance
 Cluster sampling

one-stage two-stage multistage

cluster cluster cluster
sampling sampling sampling
 ONE-STAGE CLUSTERSAMPLING
• It includes selecting a few clusters, and sampling the individuals only in these
Clusters

• For example,
• Animals in a few villages or herds could be sampled. This is cluster sampling.
• Commonly, all animals in each selected cluster are sampled; this is one-stage cluster
sampling.
 TWO-STAGE CLUSTER SAMPLING

•A sample of clusters can be selected, followed by sub-sampling of

some individuals in the clusters (in contrast to all in one-stag
cluster sampling).
• This procedure is therefore called two-stage cluster
sampling;

• Clusters are the primary units

• Selected members of the sub-samples are the secondary units.
MULTISTAG CLUSTER SAMPLING

• If the secondary units are the individual members of the study population, there
is no point in going further.
• However, if secondary units from two stage clustering consist of groups of
population members further stages of sampling could be undertaken,
corresponding to progressively higher levels of sub-sampling. This is called
multistage cluster sampling.
• For example, sampling regions, then dairy farms in each selected region, then
cows on each selected farm.
• The sampling technique at each stage is usually simple random sampling.
 MULTISTAGE CLUSTER SAMPLING

• Multistage sampling
• Similar to cluster sampling except that there are two sampling
events, instead of one
• Primary units are randomly selected
• Individual units within primary units randomly selected for measurement
 MULTISTAGE CLUSTER SAMPLING

• Multistage sampling
• Estimate the prevalence of dental caries in school children
1.Randomly select a few districts or towns from the region.
2.From each selected district, randomly choose several schools.
3.Within each selected school, randomly pick certain classes (e.g., grades 1-5)
4.In each selected class, randomly sample a number of students to assess for
dental caries.
 Random sampling:
Availability sampling:
every combination of a given
selecting on the basis of
size has an equal chance of
convenience.
being chosen.

Cluster sampling:
dividing the population into Snowball sampling:
clusters, typically on the basis asking individuals studied to
of geography, and taking a provide references to others.
sample of the clusters.

Stratified sampling:
Multi-stage sampling: dividing the population into
sampling subunits within groups on the basis of some
sampled units. characteristic and then
sampling each group.

Quota sampling: Systematic sampling:

selecting fixed numbers of choosing every nth item from a
units in each of a number of list, beginning at a random
categories. point.
 IMPERFECT TESTS

• Frequently,surveys utilize diagnostic tests to identify disease, and such tests

are not 'perfect'; that is, they have a diagnostic sensitivity and
diagnostic specificity less than 100%.
• Therefore, generate false-negative and false-positive results,respectively,
and thus producing estimates of test prevalence, rather than true
prevalence.
WHAT SAMPLE SIZE SHOULD BE
SELECTED ?

• The question that shouldbe answered in all sample surveys is

'How many individuals should bechosen for the survey?’

• The choosing of sample size depends on:

• Non-statistical considerations: It include the availability of manpower and
sampling frames.
• Statisticalconsiderations: These include the precision of the estimate of
prevalence and the expected prevalence of the disease
ANY QUESTIONS