Dr.
Ajau Danis
BHS570: Research Methodology
Explain Distinguish Understand
Explain the role Distinguish Understand the
of sampling in between steps in
the research probability and developing a
process. non-probability sampling plan.
sampling.
Sampling is the process of
selecting a small number of
elements from a larger defined
(N) target group of elements such
that the information gathered
from the small group will allow
(n) judgments to be made about
the larger groups.
Target
population or
reference
population
Study population :
sampling unit/
accessible
population
Sample
Involves two basic issues:
◦ Making the right decisions in the selection
of items (people, products, or services)
◦ Feeling confident that the data generated
by the sample can be transformed into
accurate information about the overall
population.
Define the target population
◦ Decide who should (inclusion) and should not
(exclusion) be in the sample.
◦ Use a list or set of directions for identifying the
elements of the target population (e.g. Geographic
Area, Demographics, Usage/Lifestyle, Awareness).
◦ Entirely depending on Research Problems and
Research Design.
Determine sampling frame
◦ Sampling frame is a list of elements from which a
sample may be drawn.
E.g. list of registered Physiotherapist/Health Science
students.
◦ Sample frame error occurs when certain elements of
the population are accidentally omitted or not
included on the list.
Select sampling technique
• Decide whether to use probability or non-probability
sampling technique.
Select sampling units
• Sampling unit is a single element or group of elements
subject to selection in the sample.
• E.g. A researcher wishes to sample women from two out of
five villages who comes for the mammography. The
researcher may decide to take all women who attended to
Hospital A within the last three months.
Enough samples produce statistically
significant results.
If your sample size is too small, you could
miss important research findings; statistically
Determine sample size inconsistence results.
If it’s too large, you could waste valuable time
and resources.
Consult your statistician to get the right
sample size.
Conducting field work The process of collecting data.
Sampling error is any type of bias
that is attributable to mistakes
in either drawing a sample or
determining the sample size.
Non-
Probability probability
Simple random Convenience sampling
sampling
Systematic random Purposive sampling
sampling
Stratified random Quota sampling
sampling
Cluster sampling Snowball sampling
Probability sampling refers to the selection
of a sample from a population, when this
selection is based on the principle of
randomization, that is, random selection or
chance. Probability sampling is more
complex, more time-consuming and
usually more costly than non-probability
sampling.
Probability sampling allows researchers to
create a sample that is accurately
representative of the real-life population of
interest.
Simple random sampling – randomly pick
individuals to be included in the sample.
• All individuals must have an equal
chance of being selected.
• As sample size increases, sample
becomes more and more representative
of population.
• Sampling is generally without
replacement
• Problem: can be very costly if population
is large.
Systematic random sampling is a method of
probability sampling in which the defined target
population is ordered, and the sample is
selected according to position of a skip
interval, e.g., every fifth person is chosen
Problems: same as simple random. Rule must
not lead to bias.
1: Obtain a list of units that contains an acceptable frame of the
target population
2: Determine the number of units in the list and the desired
sample size
3: Compute the skip interval
4: Determine a random start point
5: Beginning at the start point, select the units by choosing each
unit that corresponds to the skip interval
Sampling frame There are 50 students in a
1 14 26 39 class and you want to select
10 students. The first step is
2 15 26 40
Interval
determine the width of the
3 16 28 41 interval (50/10=5). This mean
4 17 29 42 from every 5 you need to
select one. Using the SRS
5 18 30 43 techniques, from the first
6 19 31 44 interval (1-5 elements), select
7 20 32 45 one of the elements. Example
you select the 3rd element.
8 21 33 46 From the rest element select
9 22 34 46 every 3rd element.
10 23 35 48
11 24 36 49
Sample selected:
12 25 37 50 3,8,13,18,23,28,33,38,43,48
13 38
Stratified random sampling is a method of
probability sampling in which the population is
divided into different subgroups and samples are
selected from each subgroups.
Stratified sampling – break
the sample into various
subgroups or strata and
get the sample from them.
Must have good knowledge
of strata.
1 2 3 4
1: Divide the 2: Draw random 3: Combine the Example : age and
target population samples from samples from sex
into each stratum each stratum into
homogeneous a single sample of
subgroups or the target
strata population
Male Female
Age >=35 Age >=35
Age < 35 Age < 35
selection
selection
selection
selection
Random
Random
Random
Random
Schematic representation of proportional stratified sampling, drawing a random
sample of 100 subjects from an accessible of 1,000 people.
Cluster sampling - the
subjects are selected in
groups or clusters rather
than randomly
E.g., interviewing
McDonald’s employees
Clusters would be every
employee at a particular
store.
One-stage cluster sampling – sample all members of the cluster
Two-stage cluster sampling – random sampling within the clusters
Weighting of clusters: probability proportionate to size (PPS) sampling
• Not all clusters are the same size.
• Can weight the clusters to equate the difference.
• Can weight the chances of a cluster being selected
Effectiveness of cluster sampling
• Much more efficient; less costly
• Not quite as effective as random sampling
Example: Two-Stage sampling
Technique involves drawing different samples to
obtain a final sample.
Purpose-the final cost is minimized.
Similar to cluster sampling.
Difference is that there is random sampling again from within each
chosen clusters rather than including all units in the cluster (e.g.: in
previous example of form 4 students, there will be random selection
again in chosen classes).
Multistage Cluster Sampling: Procedure
Used in situations where
probability sampling
Greater chance of being
techniques are either
selected than others
impractical or
unnecessary
Usually much cheaper to Logistically much easier
obtain samples to operational
Sometimes used in the
preliminary stages of
research
•E.g.: pilot study, pretest of the
questionnaire
Purposive sampling - Use judgment and deliberate effort to pick
individuals who meet a specific criteria.
Especially good for exploratory or field research.
Appropriate for at least 3 situations.
1. select cases that are especially informative.
E.g., college coaches and championships
2. desired population for the study is rare or very difficult to locate.
E.g., prostitutes
3. case studies analysis – find important individuals and study them
in depth.
Researcher
Attempts to provide representative samples (but without random sampling)
Guided by some visible characteristics, such as gender, race, age, etc.
Different from stratified sampling is that there is no random selection of
respondent within strata (select any case which fits the prescribed criteria)
Sample collected from a location convenient to the researcher – i.e., entrance to the
premises
Numbers of samples based on quota setting, i.e. 20 males above 50; whenever a
male enters the premises, ask his age until you get 20 males aged 50 years old
Advantages
•Less costly
•Administratively easy
•Quick reply
•Does not need any sampling frame
Disadvantages
•Within quota the sampling may be
unrepresentative
•Widely used social class grouping is subjective
Choose to include anyone who agrees to take part
Recruited quickly and easily with minimal effort
E.g.: “phone-in” polls (survey done by newspaper
Produce quite large samples quickly and cheaply
Usually unrepresentative of underlaying population
Snowball sampling – the process of selecting sample
using networks/can lead to others
An individual or group of individuals are sampled. They
provide other sources to be sampled.
◦ Sampling snowballs into a large selection.
◦ aka. Chain sampling
Useful for hard to identify groups.
◦ E.g., study of criminal organizations
May lead to biased sample
Degree of Research
Resources
accuracy objectives
Knowledge of
Time frame target Research scope
population
Statistical
analysis needs
How many completed questionnaires do we
need to have a representative sample?
Generally, the larger the better, but that
takes more time and money.
How different or dispersed the
population is?
Answer depends on: Desired level of confidence.
Desired degree of accuracy.
Krejcie & Morgan, 1970
Example: Sample size calculator
http://www.raosoft.com/samplesize.html
https://www.scalestatistics.com/sample-size.html
https://www.calculator.net/sample-size-
calculator.html?type=1&cl=95&ci=5&pp=50&ps=1000&
x=41&y=9
https://www.checkmarket.com/sample-size-calculator/
Common Methods:
◦ Budget/time available
◦ Executive decision
◦ Statistical methods
◦ Historical data/guidelines
INCLUSION & EXCLUSION CRITERIA IN SAMPLING
Procedure
Difference between
multistage and cluster Population
sampling
Multistage-
secondary units are Primary sampling
sampled Units (PJ district)
Sampled
Cluster
Secondary unit
All secondary units (City blocks)
are included
Sampled
Tertiary unit
(Household)
Sampled
Further units
(individual)
