THIS IS VENKATESH .E WORKING AS ASSISTANT PROFESSOR AND THIS CONTENT OF SAMPLING WILL HELP TO THE M.SC I YER NURSING AND B.SC NURSING IV YEAR STUDENTS.THIS CONTENT WAS PREPARED AND REFERRED BY MY TEACHER G.ASHA KUMARI ,ASSOCIATE PROFESSOR. I HOPE THIS WILL ENHANCE KNOWLEDGE OF STUDENTS.
3. GENERAL OBJECTIVE:
On completion of the class group will gain in depth knowledge regarding sampling,
develop positive attitude regarding sampling and practice these skills in research process.
SPECIFIC OBJECTIVES:
After the completion of the class the group will be able to:
• define basic terminologies
• enlist the purposes of sampling
• enlist Characteristics of good sample
• enumerate the need of sampling
• describe the sampling process
• list the types of sampling
• explain about probability sampling techniques
• describe about non –probability sampling techniques
• discuss about differences between probability and non –probability sampling techniques
• explain about sampling error
4. TERMINOLOGIES
POPULATION:
Population is the set of people or entities to which the
results of a research are to be generalized.
--- Suresh.K.Sharma
It refers to the group of people, items or units under
investigation and includes every individual.
---Polit and Beck
5. • Target population:
A target population consists of the total number of
people or objects which meets the designated set of criteria
--Suresh.K.Sharma
It is the aggregation of all the cases with a certain
phenomenon ( or phenomena) about which the researcher
would like to make a generalization.
--polit and beck
6. ACCESIBLE POPULATION:
It is the aggregate of cases that conform to
designated criteria and are accessible as subjects
for a study.
--Suresh.K. Sharma
7. Sampling
Sampling is the process of selecting a
representative segment of the population under
study.
--Suresh.k.Sharma
It is a process of selecting a sample from
the population.
--polit and beck
8. Sample :
Sample may be defined as representative unit
of a target population, which is to be worked upon
by researchers during their study.
--Suresh.k.Sharma
Sample consists of a subset of units which
comprise the population selected by investigators
or researchers to participate in their project.
-- polit and beck
9. SAMPLINGFRAME:
It is a list of all the elements or subjects in the population from
which the sample is drawn. Sampling frame could be prepared by
the researcher or an existing frame may be used.
For example, A researcher may prepare a list of all the
household of a locality which have pregnant women or may use a
register of pregnant women for antenatal care available with the
local anganwadi worker.
--Suresh.K.Sharma
10. SAMPLINGERROR:
There may be fluctuations in the values of the
statistics of characteristics from one sample to
another, or even those drawn from the same
population.
-- Suresh.K.Sharma
11. SAMPLINGBIAS :
Distortion that arises when a sample is not representative
of the population from which it was drawn.
--Suresh.K.Sharma
SAMPLINGPLAN:
The formal plan specifying a sampling method, a sample
size and the procedure of selecting the subjects.
-- Suresh.K.Sharma
13. Characteristics of a good sample:
• REPRESENTATIVE: A representative sample is one that the
key characteristics of which are closely related to those of
the population. Representativeness of the sample makes it
possible to generalize the findings for the population.
• FREE FROM BIAS AND ERRORS: A good sample is one
which is free from deliberate selection of the subjects for
study. Sample should be free from sampling errors or
sampling bias..
14. • NO SUBSTITUTION AND INCOMPLETENESS: A sample
is said to be good if once a subject is selected for the
study, it is neither replaced nor it is incomplete in any
aspect of researchers interest.
• APPROPRIATE SAMPLE SIZE: Generally, it is believed
that in quantitative studies the large the sample size, the
better is the probability of the goodness of the sample.
However, in qualitative studies this notion is not considered
important.
15.
16.
17. SAMPLING PROCESS
DEFINE THE POPULATION
SPECIFY THE SAMPLING UNIT
SELECTION OF SAMPLING METHOD
DETERMINATION OF SAMPLING SIZE
SPECIFY SAMPLING PLAN
SELECTION OF SAMPLE
IDENTIFY THE SAMPLING FRAME
18. • IDENTIFY AND DEFINING TARGET POPULATION:
The first step of the sampling process is the identification and defining the
target population.
• DESCRIBING ACCESSIBLE POPULATION AND ENSURING SAMPLING
FRAME:
It is not always possible to have access to each subject included in the
target population. Therefore, researcher must establish a description about the
accessible population which is readily available for research. Researcher must
have a sampling frame available to select a sample from accessible population.
19. SPECIFYING THE SAMPLING UNIT:
Researcher must establish the specific inclusion and
exclusion criteria to select a particular sampling unit.
SPECIFYING SAMPLE SELECTION METHODS:
It is one of the important stages of the sampling process
where the researcher decides whether sample will be drawn from
the population by using probability or non- probability sampling
techniques.
20. • DETERMINING THE SAMPLE SIZE:
It is very essential to determine the size of the sample so that the
researcher can plan the sampling process accordingly.
• SPECIFYING SAMPLING PLAN:
Before the selection of a particular sample, the researcher must have a
final sampling plan.
• SELECTING A DESIRED SAMPLE:
Finally, a researcher draws a representative sample from the
accessible population which is used by the researcher for data collection
in research study
21. FACTORS INFLUENCING SAMPLING PROCESS
Nature of the researcher
Inexperienced investigator
Lack of interest
Lack of honesty
Intensive work load
Inadequate supervision
22. Nature of the sample
Inappropriate sampling techniques
Sample size
defective sampling frame
Circumstances
Lack of time
Large geographic area
Lack of cooperation
Natural calamities
23. TYPES OF SAMPLINGTECHNIQUES
PROBABILITY SAMPLING NON-PROBABILITY SAMPLING
SIMPLE RANDOM SAMPLING PUPOSIVE SAMPLING
STRATIFIED RANDOM SAMPLING CONVENIENCE SAMPLING
SYSTEMATIC RANDOM SAMPLING QUOTA SAMPLING
CLUSTER SAMPLING SNOWBALL SAMPLING
SEQUENTIASAMPLING CONSECUTIVE SAMPLING
VOLUNTEER SAMPLING
GENEALOGY SAMPLING
24.
25. PROBABILITY SAMPLING
It is based on the theory of probability. It involves
random selection of the elements/ members of the
population. In this, every subject in a population has
equal chance to be selected as study sample.
Probability sampling technique is used to enhance
the representativeness of the selected sample for a
study. In probability sampling techniques, the chances
of systematic bias are relatively less because subjects
are randomly selected.
--Suresh.k.Sharma
26. FEATURES OF PROBABILITY SAMPLING:
• The technique probability sampling involves the samples to be
gathered in a process that provides equal chance to all the
individuals in the population of getting selected
• In this sampling technique, the researcher surely has to
guarantee that each individual has an equal opportunity for
getting selected. This is feasible only if the researcher uses
randomization.
27. • The absence of both sampling and systematic bias is
the advantage of utilizing a random sample. If
random selection is made in a proper manner, the
sample is representative of the whole population.
• The effect of this is an absent or minimal systematic
bias and is a difference between the results from the
population. Since the subjects are randomly
selected, therefore, sampling bias is eliminated.
28.
29.
30. Methods Of Simple Random Sampling:-
1.Lottery method
2.Table f random numbers
3.Computer table
31. 1.Lottery Method :-
In case of small population size, names can be
written on a piece of paper, which are then placed in a containership
and bowl, mixed well to draw adequate size of sample.
NO. NAME OF THE STUDENT
1 Ramandepep
2 kulwinder
3 navdeep
4 deepinder
5 kusum
6 kamini
7 lalita
8 jasbir
9 jayamala
10 savita
NO. NAME OF THE STUDENT
11 harpreet
12 hardeep
13 mandeep
14 manjeet
15 kuldeep
16 shivani
17 shalini
18 sharmilla
19 shalony
20 lakhwinder
NO. NAME OF THE STUDENT
21 shruthi
22 navpreet
23 gurdeep
24 jaswinder
25 sharika
26 sidhika
27 tanveer
28 anitha
29 bindu
30 rajbeer
32. 2.Table of random numbers:-
Once the sampling frame has been
developed and numbers are assigned consecutively. A table of
random number should be developed to draw desired number of
sample size draw the sample from random table, researcher blindly
put the finger/pencil on anyone number then move the finger/pencil
up/down or right or left to select desired number of sample.
12 85 73 97 99 31 01 04 39 07
13 17 93 94 89 66 42 57 89 58
10 63 20 33 47 12 54 79 48 21
37 75 26 64 06 09 90 92 25 76
9 90 92 25 76 10 85 79 97 33
12 54 79 48 21 14 17 93 94 89
66 42 57 89 52 37 63 20 20 33
31 01 04 39 89 66 42 57 89 58
54 90 92 25 54 79 48 21 02 8
55 54 79 42 09 08 03 05 08 2
33. 3.Computer Table:-
In this method ,a random
number of table is created with the help
of computer software to draw a desired
number of sample for study.
34. Types of Simple Random Sampling:-
1.Simple random sampling with replacement
2.Simple random sampling without replacement
35. Simple randomsampling with replacement:-
It is a method of selection of sample from
accessible population in such a way that at one
stage of selection, removing samples have same
chance of being selected .
36. Simple random sampling without replacement :
It is a method of drawing sample in
which researcher draw the sample from the
accessible population one by one and at each
stage chances for selection of sample will be
reduced.
37. MERITS:
Most reliable and unbiased method
Requires minimum knowledge of study population
Free from sampling errors/ bias
DEMERITS:
Needs up-to-date complete list of all the members of
the population
Expensive and time consuming
Lots of procedures need to be done before sampling
is accomplished.
38.
39. MERITS:
Convenient and simple to carry out
Distribution of sample over entire population
Statistically more efficient and provides a better
representative sample when population elements are
randomly distributed
DEMERITS:
Less representative sample if subjects are non randomly
distributed.
Sometimes may result in biased sample
40.
41. TYPES OF STRATIFIED RANDOM SAMPLING:
1. Proportionate stratified random sampling
2. Disproportionate stratified random sampling
42. •Proportionate stratified random sampling:
In this, sample chosen from each stratum is in
proportionate to the size of total population. When viewed
against the entire population, the sample size of each
stratum in this technique seems proportionate to the
population size of the stratum. It is understood that each
stratum has a similar sampling fraction.
STRATUM
Population size
Sampling fraction
Final sample size
A
100
1/2
50
B
200
1/2
100
C
300
1/2
150
43. DISPROPORTIONATE STRATIFIED RANDOM SAMPLING:
In this sub type the sample chosen from each
stratum, is not in proportionate to the size of total
population in that stratum. Sampling fraction is the
only difference between proportionate and
disproportionate stratified random sampling.
STRATUM
Population size
Sampling fraction
Final sample size
A
100
1/2
50
B
200
1/4
50
C
300
1/6
50
44. MERITS:
•Ensures representative sample in heterogenous
population
•Comparison is possible in two groups
DEMRITS:
•Requires complete information of population
•Large population is required
•Chances of faulty classification of strata
47. ONE STAGE CLUSTER SAMPLE:
It occurs when the researcher includes all the high school students
from all the randomly selected clusters as sample
TWO STAGE CLUSTER SAMPLE:
In this, initially the researcher lists all the clusters appearing in the
population. After this, the clusters are selected normally by simple random
sampling. Then usually by simple random sampling or often by systematic
sampling, the elements or units in the selected clusters of the first stage
are then sampled in the second stage.
48. • MULTISTAGE CLUSTER SAMPLE:
In multistage cluster sample, the sampling is done at
more than the two levels by initially identified clusters as population
at different levels and selecting them using simple random sampling
technique and finally selecting units (elements) using simple random
sampling or systematic sampling.
PROBABILITY PROPORTION TO SIZE CLUSTER SAMPLE:
It is a variant of cluster sampling, when size of clusters is
not same and there is risk of over-sampling from smaller size
clusters and under sampling from the large size clusters. Then it is
most useful when the sampling units vary considerably in size
because it assures that those in large sites have the same
probability of getting into the sample as those in the smaller sites,
and vice versa.
49. MERITS:
•Cheap, quick, and easy for a large population
•Population of parameters can be estimated for
sample size
DEMERITS:
•Possibility of high sampling error
•Chances of least representative sample due to
over-represented or under-represented cluster
50. SEQUENTIAL SAMPLING
This method of sample selection is slightly different
from other methods. Here the sample size is not fixed. The
investigator initially selects small sample and tries out to
make inferences, if not to draw results, he or she then adds
more subjects until clear- cut inferences can be drawn.
FOR EXAMPLE, A researcher is studying association
between smoking and lung cancer. Initially researcher takes
a small sample and tries to draw inferences. If unable to
draw any inferences he or she continues to draw the sample
until meaningful inferences are drawn.
51. MERITS
• Facilitates to conduct a study on best-possible smallest
representative sample
• Helping in ultimately finding the inferences of the study
DEMERITS:
• With the sampling technique, it is not possible to study a
phenomenon which needs to be studied at one point of
time.
• Requires repeated entries into the field to collect the
sample
52. NON- PROBABILITY SAMPLING
Non- probability sampling is a
technique wherein the samples are
gathered in a process that does
not give all the individuals in the
population equal chances of being
selected in the sample.
53.
54.
55. MERITS:
Simple to draw sample and useful in explorative
studies
Saves resources, requires less field work
DEMERITS:
•Requires considerable knowledge about the
population under study
•It is not always reliable sample, as conscious bias
may exist
56.
57. MERITS:
•Easiest, cheapest and least time consuming
•Helps in saving time, money and resources
•DEMERITS:
•Chances of non representative sample,
systematic error/ bias because only interested
people participate in study
•Study results may lack the generalisability
58.
59. MERITS:
•Economically cheap
•Suitable where the field work has to be done
like studies related to market and public
opinion polls
•DEMERITS:
•Always does not guarantee representative
sample
•Chances of sampling bias
60.
61. Types of snowball sampling :
Linear or single chain :
In this type, the early sample refer or register only one next sample for
study and at the end of completion a single chain will be formed.
A B C D
Exponential non discriminative snowball sampling:
In this subtype, initial/ early sample is requested to refer or register at
least two next samples for the study.
B D
E
A
F
C G
62. Exponential discriminative snowball sampling:
It is similar to non discriminative sampling where one subject or
sample will register at least two samples.
G
C D F
A B
E
63.
64.
65. MERITS:
•Facilitates sampling for people difficult to locate
•Cheap, simple and cost efficient
•Needs little planning and lesser work force
DEMERITS:
•Little control of researcher of over the sampling
method
•Representativeness of the sample is not
guaranteed
•Chances of poor coverage of entire population
66. •Target subjects are informed
through mass media to participate
in study and interested participants
may voluntarily contact researcher
to participate in the study.
VOLUNTEER SAMPLING
67. MERITS:
Cost effective sampling technique
Needs very limited efforts and time to
locate the study participants.
This technique helps to collect large size
data in limited time period.
68. DEMERITS:
Only interested people contact to
participate, so there are very less
chances that sample may not be a
representative sample.
Study results may lack the
generalizability.
69. CONSECUTIVE SAMPLING
• Picks up all the available
subjects who are meeting the
present inclusion and exclusion
criteria.
70. MERITS:
There is very little effort on the part of the
researcher when performing this sampling
technique.
It is not expensive, not time consuming,
and not work force intensive.
Ensures more representativeness of the
selected sample.
71. DEMERITS:
The researcher has no set plans about the
sample size and sampling schedule.
It always does not guarantee the selection of
representative sample.
Results from this sampling technique
cannot be used to create conclusions and
interpretations pertaining to the entire
population.
72. GENEALOGY SAMPLING
• A participant is identified and then
he/she is asked to refer his/her
relative families to participate in
study irrespective of their location
of stay.
73. MERITS:
This sampling technique is useful in drawing
a representative sample from traditional rural
communities, which are socio-culturally and
economically homogenous.
Saves the time and efforts in locating the
study subjects because participants are
identified through reference from previous
participants.
74. DEMERITS:
This sampling technique encounters
problem of systematic errors or bias.
It lacks the diversity of sample
participant characteristics because
subjects are selected from a family or
related families. Thus, it has limited
usefulness.