Mattu University

Department of Psychiatry

Research Methodology

for 4th year psychiatry Students

By: Aman D.
Research population
 Research population is a group of
units that to be investigated for the
problem under study
 It can be persons, households,
institutions, etc.
 The population under consideration

should be clearly defined in terms of

place, time, and other relevant criteria
Population…

 When we can select the population for our

study?
 Investigator will have indirectly chosen
his/her study population when he/she
defined the topic of his investigation by
reason of his/her interest in a specific
community
Population…
 The appropriateness of the study
population refers to its suitability for
the achievement of the objectives
 Suitability usually affects the validity
of subsequent generalizations from
the findings
Population…
 Consider the following conditions while
selecting the population:-
 Select the population that would give
you the required information
 Consider the cooperation of participants
to give a the information
Population…
 Consider the sufficiency of cases to
yield useful conclusions
 Consider the mobility of the
population if a long term 'follow up‘
study is planned
Population…
 Source/Target/ Reference population
 Is the population about which an investigator
wishes to draw a conclusion/generalization
 It is the population from which the sample
will be calculated
E.g.,:- All schizophrenic patients that are
following treatment at MKH in 2016
Population…
 Study population
 Portion of target population that are available
during the study period
 It is the population from which the sample
actually drawn
E.g.,:- All schizophrenic patients that are
available for follow-up treatment during data
collection period
Population…
 Study population is often more limited
than the target population
 In some cases, the target population
and study population can be identical
Population…
 Study unit
 The unit from which the actual data will
be collected
E.g.,: persons (in a study of disease
prevalence) or households (in a study of
family size)
Sampling

 Sampling is the selection of a

number of study units from a
defined study population
Sampling…
 Since it is too large to collect
information from all members of the
population, we select a sample of
individuals hoping that the sample is
representative of that population
Sampling…
 During sampling, the exact number of
the sample needed an the mechanism
to select that much number from the
source population should be clarified
Sampling…
 Sampling unit
 The unit of selection in the sampling process
E.g., - person, household, institution, etc.,
 Sampling frame
 The list of units from which the sample is to be
selected
 The list of all the units that compose the study
population
Sampling Methods/Techniques
 Sampling methods are the techniques
through which we select the calculated
number of samples
 There are two types of sampling methods
 Probability sampling methods
 Non-probability sampling methods
 Probability (Random) samplings
 Probability sampling involves using
random selection procedures to
ensure that each unit of the sample is
chosen on the basis of chance
Probability sampling…

 All units of the study population

should have an equal, or at least a
known (greater than zero) chance of
being included in the sample
Probability sampling…
 Probability sampling requires that a listing
of all study units (sampling frame) exists or
can be compiled
 The following are common types of
probability samplings
Probability sampling…
1. Simple Random sampling
 This is the simplest form of probability
sampling in which we select the sample by
chance (lottery method) from the available
study units
 To use this sampling method, follow the
following procedure:-
Probability sampling…
a. Prepare the numbered list of all the units
in the population from which sample is to be
drawn
b. Decide/calculate your sample size
c. Select the required number of samples
using lottery method or tables of random
numbers
Probability sampling…
 Example

 Assume you are going to select 20 samples

from a total of 150 study units. Then:-
 Assign the number (1-150) to each study
unit
 Write each number (1-150) on a pieces of
paper and fold it
Probability sampling…
 Put the folded pieces of paper in a box
 Shake the box vigorously to ensure
randomization
 Pick 50 pieces of paper by chance and
collect data from units that are belonged to
the selected numbers
Probability sampling…
 If the sampling frame is well prepared,
this sampling is bias free as one
element cannot come twice as a
sample
Probability sampling…
 But the difficulty to use this sampling is that it
needs complete lists of study population
(sampling frame)
 This is difficult to prepare especially if the
population is large and the exact number of
the population is unknown
 Additionally, some small group of interest may
be excluded by chance from the sample
Probability sampling…
2. Systematic Random sampling
 In this sampling, samples are chosen
at regular intervals from the sampling
frame after we calculated for that
interval
Probability sampling…
 Ideally we randomly select a number
to tell us where to start selecting the
samples among the numbers in the
interval
Probability sampling…
 Steps to Achieve a Systematic Random
Sampling
 Number the units in the population from
1to N (total number of study population)
 Decide the sample size (n) that you need
Probability sampling…
 Calculate the interval size (k) = N/n
 Randomly select number between 1 to k
 Then take every Kth unit
Probability sampling…
 Example

 You are about to select 100 samples from

1200 students using systematic random
sampling
 Then: N =1200, n = 100
 Sampling interval (K) =N/n = 1200/100 = 12
Probability sampling…
 After this, the number of the first
student to be included in the sample
is chosen randomly from1 to 12
 For instance, let’s say that number 6 is
picked by chance
Probability sampling…
 Then, starting with student number 6, every
twelfth student will be included in the
sample until 100 students are selected
 Accordingly, the numbers selected would be
6, 18, 30, 42, etc.
 What about if you select number 1 by
chance?
Probability sampling…
 Systematic sampling is usually less time
consuming and easier to perform than simple
random sampling
 However, there is a risk of bias if we are
dealing with populations having “hidden
periodicities”
 The selected kth unit may be belonged to the
population with the same characteristics
Probability sampling…
3. Stratified Random Sampling
 In stratified random sampling the population
is first divided into different homogeneous
group or strata based on certain criteria (e.g.,
age, sex, religion, residence, etc.,) to ensure
the inclusion of all groups
Probability sampling…
 Then random samples of a pre-determined
size will have to be obtained from each
group (stratum)
 Stratified sampling is only possible if
proportion of each group of the study
population is known
Probability sampling…
 This sampling is employed when we
assumed that the population in different
stratum has heterogeneous characteristics
 Stratified random sampling can be either
proportionate or disproportionate
Probability sampling…
 Proportionate Stratified Random Sampling
 In case of this method, the researcher
stratifies the population according to known
characteristics
 Then, randomly draws the sample from
each stratum according to its proportion to
total sample size
Probability sampling…
 Example
 Assume that you are going to select 150
students from 1000 total students after
stratified them according to their study year
which consists 60% of 4th year, 30% of 3rd
year and 10% of 2nd year students.
Probability sampling…
 Then calculate the number for each batch:-

 4th year = 150x60% = 90 students

 3rd year = 150x30% = 45 students

 2nd year = 150x10% = 15 students

Probability sampling…
 Disproportionate Stratified Random
Sampling
 In this method, the sampling unit in
each stratum is not necessarily be as
per their population proportion
Probability sampling…
 Rather, the equal number of the
sample selected from each stratum
especially when we interested to
compare the strata depending on
different characteristics
Probability sampling…
 Example
 if you interest to compare the level of test
anxiety between male and female students, it
is better to take equal number of male and
female students without consideration of
their total number
Probability sampling…
 This is a biased type of sampling and in
this case some stratum is over-
represented and some are less
represented
 It is not truly representative sampling,
still this to be used in some special cases
Probability sampling…

4. Cluster (area) Random sampling

 This is the probability sampling in
which we select a groups of study
units (clusters) instead of study units
(individuals)
Probability sampling…
 In this sampling, the total population is
divided into a number of relatively small
sub divisions or groups (clusters) and
then some of these clusters are randomly
selected for inclusion in the sample
Probability sampling…
 We use this sampling when we assume that
the population in each cluster is
homogeneous
 Clusters are often geographic units (e.g.,
districts, villages, kebales, etc.,) or
organizational units (e.g., clinics, schools,
etc.,)
Probability sampling…
 In cluster sampling, we follow these steps:
 Divide population into clusters (usually along
geographic boundaries)
 Randomly select some clusters (usually 30%
assumed to be representative)
 Include all units within sampled clusters or
select another sample by using other sampling
methods
Probability sampling…
5. Multi-stage sampling
 For very large and diverse populations,
sampling may be done in two or more
stages
 This is often the case in community-based
studies, in which people are to be
interviewed from at stages
Probability sampling…
 Example:

 In a study of utilization of mental

health services in Mattu town, a total
of 135 households are to be visited
for interviews with family members
Probability sampling…
 Let’s say the town has 6 kebales and
each kebale has 5 Gots
 Assume that you decided to include 3
kebales; and 3 Gots from each kebale
 Then follow the following procedure
Probability sampling…
 Select 3 kebales randomly
 From each kebale, select 3 gots again
randomly
 Then, divide 135 households for 9 gots
(which is 15 HH from each got)
 Then divide the total number of HHs in that
got by 15 to obtain K value (interval)
Probability sampling…
 But, how we start the household to start
from?
 Go to the center of selected got
 Choose a direction in a random way (spin a
bottle on the ground and choose the
direction of the bottle neck)
Probability sampling…

 Walk in the chosen direction and select the

HH every Kth interval calculated until you

get 15 HHs
Probability sampling…
 If you reach the boundary of the Got
and you still do not have 15
households, return to the centre of the
Got; walk in the opposite direction and
continue to select your sample in the
same way until you have 15 households
Probability sampling…
 If there is nobody in a chosen household,
take the next nearest one
 Decide beforehand whom to interview
(for example the head of the household,
if present, or the oldest adult who lives
there and who is available)
Probability sampling…

6. Multi-phase sampling

 In this method, part of the information

is collected from the whole sample

and part from the sub-sample

Probability sampling…
 Example: - In Tuberculosis survey
 Phase I - Mantoux test will be performed on all cases
of the sample
 Phase II - All persons positive for Mantoux test will
be subjected to X-ray test
 Phase III - All those who were having clinical
symptoms and X-ray positive will be
subjected to sputum examination for confirmation of
the diagnosis
Non-probability samplings
 These are the sampling methods in
which the probability/chance of
selection for the individual study unit
from the target population is unknown
Non-probability samplings…

 Some groups of the population may included

in large and the others may remains
unselected
 Because of this, sometimes it is difficult to
generalize the findings when we utilized these
samplings
 The following are common types non-
probability samplings
Non-probability samplings…
1. Consecutive sampling
 This is the sampling technique in which
we select participants consecutively until
the intended number of sample achieved
 This is common sampling technique in
the study that involves clinical patients
Non-probability samplings…

2. Convenience sampling
 is a method in which the researcher
includes the participants for convenience
sake as they are easily accessible
Example
Interviewing passengers on the road about
some topic/issue
Non-probability samplings…
3. Purposive sampling
 This method is typically used in qualitative
research when focusing on a limited number
of informants, whom we select strategically
so that their in-depth information will give
optimal insight into an issue about which
little is known
Non-probability samplings…

4. Quota sampling

 In this sampling, we select people non-

randomly according to some fixed quota

Non-probability samplings…

5. Snowball or chain sampling

 This approach is particularly suitable for locating
key informants or critical cases
 We start with one or two information-rich key
informants and ask them if they know persons who
know a lot about the topic of interest
Example
 Research concerning homosexuality
Thanks