By the end of this Module, you should be able to;
1. Identify and explain the types of study designs in epidemiology
2. Choose a suitable sampling technique based on your research question
3. Apply these methods to your projects and manuscripts
1. INTRODUCTION TO RESEARCH WRITING:
METHODS AND TOOLS (An Online Course)
ORGANIZED BY
TORASIF, IN COLLABORATION WITH FGMELSA, GHF, SAS AFRICA
MODULE 2 : RESEARCH METHODS AND DESIGNS
Stephen Opoku
Research and Project Coordinator, FG-MELSA and COHES Foundation
(Immunologist)
INTRODUCTION TO RESEARCH WRITING Organized by
TORASIF, FGMELSA, GHF and SAS AFRICA
2. Roadmap
•By the end of this Module, you should be
able to;
•Identify and explain the types of study
designs in epidemiology
•Choose a suitable sampling technique based
on your research question
•Apply these methods to your projects and
manuscripts
INTRODUCTION TO RESEARCH WRITING Organized by
TORASIF, FGMELSA, GHF and SAS AFRICA
3. STUDY DESIGN
Research design refers to the overall strategy utilized to
carry out research that defines a succinct and logical
plan to tackle established research question(s)
INTRODUCTION TO RESEARCH
4. STUDY DESIGNS IN EPIDEMIOLOGY
STUDY DESIGNS IN
EPIDERMIOLOGY
Observational
B. Descriptive
1.Case series
2. Case Report
B. Analytical
1. Cross sectional
2. Case control
3.Cohort
Experimental
RCT’s
Community trials
Field trials
INTRODUCTION TO RESEARCH WRITING Organized by
TORASIF, FGMELSA, GHF and SAS AFRICA
10. Population and sample
• Samples are drawn to support both descriptive
studies (often called surveys) and analytic
studies (often called observational studies).
• A descriptive study (or survey) aims to describe
population attributes (frequency of disease, level
of production).
• An analytic study is done to test a hypothesis
about an associationbetween outcomes and
exposure factors in the population
Dr Michael Owusu
11. Population and sample
• The external population is the population to
which it might be possible to extrapolate results
from a study.
• The target population is the immediate
population to which the study results will be
extrapolated. The individuals included in a study
would be derived from the target population.
Dr Michael Owusu
12. Population and sample
• The study population is the population of
individuals selected to participate in the study.
Dr Michael Owusu
13. Census and Sample
• Census: Every individual in the population is
evaluated.
• Sample: data are only collected from a subset of
the population.
• In a census, the only source of error is the
measurement itself.
• With a sample, there are both measurement and
sampling error to contend with.
Dr Michael Owusu
15. Non-probability sampling
• These are samples that are drawn without an
explicit method for determining an individual's
probability of selection.
• Judgement sample: This type of sample is
chosen because, in the judgement of the
investigator, it is 'representative' of the target
population
Dr Michael Owusu
16. Non-probability sampling
• Convenience sample: A convenience sample is
chosen because it is easy to obtain.
• A purposive sample is a non-probability sample
that is selected based on characteristics of a
population and the objective of the study.
• If a random sample is drawn from the sampling
units meeting the study criteria, then it becomes
a probability sample from the subset of the
target population.
Dr Michael Owusu
17. Non-probability sampling
• Referral or Snowball Sampling: This sampling
strategy is based on referrals to subjects of
interest.
Dr Michael Owusu
18. Probability sampling
• A probability sample is one in which every element in the
population has a known non zero probability of being
included in the sample.
• In a simple random sample, every element in the target
population has an equal probability of being included.
• A complete list of the target population is required and a
formal random process is used.
• Random sampling can be based on drawing numbers
from a hat, using computer-generatedor flipping a coin
or throwing dice.
Dr Michael Owusu
20. Probability sampling
• Systematic random sample: The sampling interval
(j) is computed as the study population size
divided by the required sample size.
• The first element is chosen at random from among
the first j elements, then every Jth element after
that is included in the sample.
Dr Michael Owusu
21. Probability sampling
The Kth element can be selected using the formula:
푁
K = , where k = order of selected sample
푛
N = Population size
n = Sample size Dr Michael Owusu
22. Probability sampling
• Stratified random sampling: Prior to sampling,
the population is divided into mutually exclusive
strata based on factors likely to affect the
outcome. Then, within each stratum, a simple or
systematic random sample is chosen.
• The simplest form of stratified random sampling
is called proportional.
Dr Michael Owusu
23. Probability sampling
Types of Stratified Sampling
- Proportional Stratified Sampling: In this type of
sampling elements are selected from each
subgroup—stratum—according to their proportion
in the population of interest.
Eg: Assuming one wants to select 1000 individuals
according to religious affiliation?
Muslims = 40% ;
Christians = 30%
Traditional = 30%
How can this be achieved?
Dr Michael Owusu
25. Probability sampling
Types of Stratified Sampling:
- Disproportional Stratified Sampling: In this
method of sampling, random elements are
selected according to the will of the investigator.
Dr Michael Owusu
26. Probability sampling
• There are three advantages of stratified random
sampling.
• It ensures that all strata are represented in the
sample.
• The precision of overall estimates might be
greater than those derived from a simple
random sample.
Dr Michael Owusu
27. Probability sampling
• Cluster sampling: A cluster is a natural or
convenient collection of elements with one or
more characteristics in common.
• Cluster sampling is done because it might be
easier to get a list of clusters common objects
than it would be to get a list of individuals
objects
Dr Michael Owusu
29. Probability sampling
• Cluster sampling can be done in the following
two ways:
• Single Stage Cluster Sampling: In single stage
cluster sampling all the elements of a cluster are
selected as a sample.
• In two stage cluster sampling: first, random
selection of some clusters from the given
population is performed, then some elements
from each cluster are randomly selected.
Dr Michael Owusu
32. Probability sampling
• Multistagesampling: This is similar to cluster
sampling except that, after the first level has
been chosen, then a sample of secondary units is
selected.
• If you want to ensure that all subjects in the
population have the same probability of being
selected then a probability proportional to their
size must be used
Dr Michael Owusu