Research
Methodology in
Rehabilitation
D R .WA J E E H A
A SS I S TA N T P R O F E SS O R
Sampling
Techniques
Content
Sampling technique in research
NON-PROBABILITY SAMPLING
PROBABILITY SAMPLING
Researchers rarely have the
opportunity to study all the
individuals who possess the
characteristics of interest
within a study
Sampling
Larger group to which research results are generalized is called the population.
Sampling is the process by which a subgroup of participants is selected for study from a
larger group of potential participants.
Assignment is the process by which participants in the sample are assigned to groups
within the study.
SAMPLING Simple Random Sampling
PROBABILITY Systematic Sampling
SAMPLING Stratified Sampling
Disproportional Sampling
Cluster Sampling
Samples of Convenience
NON- Snowball Sampling
PROBABILITY Purposive Sampling
SAMPLING Quota Sampling
Nonprobability sampling
All the elements of the population do not have an equal chance of being
selected under these circumstances, we cannot readily assume that the sample
represents the target population.
The probability exists that some segment of the population will be
disproportionately represented.
Nonprobability sampling
If studies are self-funded, subject selection is confined to a single setting with a limited number
of available patients, so the researcher often chooses to study the
entire accessible population.
Nonprobability sampling is widely used in rehabilitation research and is distinguished from
probability sampling by the absence of randomization.
One reason for the predominance of nonprobability sampling in rehabilitation research is
limited funding.
1. Convenience Sampling
Samples of convenience involve the use of readily available
participants
With this method, subjects are chosen on the basis of availability.
The major limitation of this method, however, is the potential bias
of self-selection. It is not possible to know what attributes are
present in those who offer themselves as subjects, as compared with
those who do not, and it is unclear how these attributes may affect
the ability to generalize experimental outcomes.
The term “sample of convenience” seems to give the negative
implication that the researcher has not worked hard enough at the
task of sampling.
In addition, we already know that probability samples tend to have
less sampling error than nonprobability samples.
But before one totally discounts the validity of samples of
convenience, it should be pointed out that the accessible
populations can also be viewed as large samples of convenience.
Convenience Sampling
Most used and practical approach to convenience sampling is consecutive
sampling, which involves recruiting all patients who meet the inclusion and
exclusion criteria as they become available.
Essentially, a consecutive sample includes the entire accessible population
within the defined time period of the study.
Consecutive samples are used in a prospective study in which the population
does not exist at the beginning of the study; in other words, a sampling frame
does not exist
In a consecutive sample, all patients who meet the criteria are placed into the
study as they are identified. This continues until a specified number of patients
are collected, a specified time frame has passed, or certain statistical outcomes
are seen.
This can be problematic if the study period is too short,
and a sufficient number of qualified subjects cannot be
obtained, or if the time period does not allow a
representative group.
Example
For example, if we were
studying causes of low back
pain, and we only included
patients who came to the clinic
in the summer months, we
might miss patients whose pain
is caused by shoveling snow or
slipping on ice.
2. Purposive Sampling
Researcher handpicks subjects on the basis of specific criteria.
Locate subjects by chart review or interview patients to determine if they fit
the study.
When there is specific reason for selecting particular participants
for study.
Researcher must exercise fair judgment to make this process
meaningful.
Used for qualitative research.
Example
If there are 40 elementary schools in a
district and the researchers have to select
two schools.
For a purposive sample, the researchers
might pick one school because it is large and
students are from families with high median
incomes and pick a second school because it
is small and draws students from families with
modest median incomes
For a sample of convenience, the researchers
might select two schools in close proximity.
Convenience /Purposive
Sampling
Convenience sampling uses
whatever units are readily available, purposive sampling
uses handpicked units that meet the researcher’s needs.
Purposive sampling is similar to convenience
sampling, but differs in that specific choices
are made, rather than simple availability.
Both are biased sample
3. Snowball Sampling
To study sensitive topics, rare traits, personal networks, and social
relationships.
When the potential members of the sample are difficult to identify.
Researchers identify a few participants who are then asked to
identify other potential members of the sample.
Strong verification for the eligibility of each respondent to ensure a
representative group.
Carried out in stages
First stage, a few subjects who meet selection criteria are identified
and tested or interviewed.
Second stage, these subjects are asked to identify others who have
the requisite characteristics.
“Chain referral" or "snowballing" is continued until an adequate
sample is obtained.
Example
Snowball sampling is used
extensively, for example, in
studies involving homeless
persons and drug abusersY-15
4. Quota Sampling
Sampling technique happens on basis of a pre-set standard.
Sample is formed on basis of specific attributes, the created
sample will have the same attributes that are found in the total
population.
It is an extremely quick method of collecting samples.
Quota Sampling
Researcher can control for the potential confounding effect of
known characteristics of a population by guiding the sampling
process so that an adequate number of subjects are obtained for
each QUOTA
A type of stratified sample in which selection of cases within strata
is entirely random.
Potential for nonprobability biases, incorporate elements of
stratification.
Improved by proportionally representing each segment of the
population in the sample.
Example
For example, using the previous
example examining students'
attitudes toward the elderly, we
could call for volunteers to take the
questionnaire and stop the call
once we have achieved the proper
number of subjects for each class.
PROBABILITY SAMPLING
Simple Random Sampling
Systematic Sampling
Stratified Sampling
Cluster Sampling
1. Simple Random Sampling
A random sample is unbiased in that each selection is
independent, and no one member of the population has any more
chance of being chosen than any other member.
A random sample is drawn from the accessible population
Accessible population is actually defined according to the available
listings.
As easy as assigning numbers to the individuals (sample) and then
randomly choosing from those numbers through an automated
process.
Random selections can be made by computer, using
statistical packages, given a numbered list of subjects in
a data set.
Total data set usually will represent an
accessible population.
Computer can then generate a random
list of any specified size to select the
sample
Example
Listing of persons, such as membership directories, or institutions,
such as lists of accredited hospitals.
For example, if we use a professional membership directory to
create a sample of therapists, the accessible population would be
defined as therapists who were members. Not all therapists belong
to the association, however, and it may not be valid to generalize
responses of such a sample to all practicing therapists.
Computer Generated, Lottery Method, Fish Bowl, Sealed Opaque
Enveloped
2. Systematic Sampling
Selects every nth person on a list.
An efficient alternative to simple random sampling, and it often
generates samples that are as representative of their populations as
simple random sampling.
Equivalent to random sampling, as long as no recurring pattern or
particular order exists in the listing
Systematic Sampling
Exception to this is if the ordering system used somehow
introduces a systematic error into the sample.
Random sampling can be a laborious technique, unless the
accessible population is organized as a short, pre-numbered list.
When lists are arranged alphabetically or in some other ordered
fashion, an alternative approach can be used that simplifies this
procedure, called systematic sampling.
Example
To use this sampling technique, the researcher divides the total number of
elements in the accessible population by the number of elements to be
selected. Therefore, to select a sample of 100 from a list of 1,000 students,
every tenth person on the list is selected.
Interval between selected elements is called the sampling interval.
The starting point on the list is determined at random, often using a table of
random numbers.
Least time consuming and most convenient way to obtain a sample from an
available listing of potential subjects.
3. Stratified Random Sampling
Identifying relevant population characteristics, and partitioning members of a
population into homogeneous, non-overlapping subsets, or strata, based on
these characteristics.
A larger population can be divided into smaller groups, that usually don’t
overlap but represent the entire population together.
National samples for surveys and polls are often stratified by
geographic area so that the distribution of regional variables mirrors
the population.
Stratification can be done on more than one variable when this is
appropriate. Sometimes, however, variables are correlated in such a
way that only one variable needs to be the basis for stratification.
For instance, age and college class are generally related, and
therefore, stratification on class should also control for variations in
age.
Example
Proportional stratified sample
To control for this, we can create a proportional stratified sample, by first
separating the population into the four classes and then drawing random or
systematic samples from each class in the proportion that exists in the population.
Let us assume that our accessible population consists of 300 freshmen, 300
sophomores, 200 juniors, and 200 seniors. In a simple random sample drawn from
the list of 1,000 students, it is possible, just by chance, that the distribution of
subjects will not reflect the differential proportions of these classes in the
population.
Therefore, to obtain a sample of 100, we would choose 30 freshmen, 30
sophomores, 20 juniors, and 20 seniors. The resulting sample would intuitively
provide a better estimate of the population than simple random sampling.
Cluster sampling
It is a way to randomly select participants when they are geographically spread
out.
Cluster sampling usually analyzes a particular population in which the sample
consists of more than a few elements, for example, city, family, university etc.
Clusters are then selected by dividing the greater population into various
smaller sections
[Link]
kwdXEXC7yE https://
[Link]/watch?v=K8lcSHlB64w
https://
[Link]/watch?v=pTuj57uXWlk
https://
[Link]/watch?v=9PaR1TsvnJs