SAMPLING & SAMPLING TECHNIQUES
WHAT IS A POPULATION?
WHAT IS A SAMPLE?
Population
• The group of individuals under study is known as a
population or a universe.
• Types of population (based on countability):
1. Finite population: which can be counted
2. Infinite population: cannot be counted
Sample
• Finite subset of individuals in a population
• Sample size: total no. of individuals in a sample
Sampling process
- process of selection of sample from the population
Examples of sample/sampling:
• A housewife just picks up few grains of rice from
the cooking vessel and gets a fairly good idea
whether entire lot of rice is fully cooked or it
requires more cooking.
• Blood test, stool test, urine test
Objectives of sampling
• To minimize time, money and manpower, without
losing the accuracy of the conclusion
• Estimation of population parameters from the
sample statistics
• To test the hypothesis about the population from
which the sample/samples are drawn
• PARAMETER
- Value calculated from a
defined population
- Constant value
- Example: Mean (µ),
standard deviation (σ),
population size
• STATISTIC
- Value calculated from
sample
- May change according
to the sample
- Example: Mean (x̅),
Standard deviation (s),
sample size
Sampling frame:
• The list of population units from which the sample
units are to be selected is sampling frame.
• A good sampling frame is crucial to good sampling.
Qualities of sampling frame
• It should be exhaustive. i.e. all population units
must be included in the list.
• It should be up to date.
• The unit must not be repeated in the list.
• It must be maintained by an authority that can be
relied upon.
SAMPLING TECHNIQUES
PROBABILITY SAMPLING/RANDOM SAMPLING:
• For each element in the sample, the probability is
known and non-zero
• Every element of the population has the same
chance at being included in the sample
Simple random sampling:
• technique of drawing a sample in such a way that each
unit of the population has equal and independent
chance of being selected
• The sample can be selected either with replacement or
without replacement.
• It can be done in two ways:
1. Lottery method
2. Random number table
1. Lottery Method:
• Each of the ‘N’ population member is assigned a
unique number
• The numbers are placed in a bowl and thoroughly
mixed. Then a blind folded researcher selects ‘n’
numbers.
• Population members having the selected numbers
are included in the sample.
• For example; let us assume that we want to put 30
patients on a clinical trial out of 100 patients. To
ensure randomness of selection of 30 patients, we
may adopt lottery method by numbering the total
individuals from 1 to 100. Then we note down the
serial numbers of patients on 100 slips of papers, put
all the slips into a drum and shake well so that all the
slips are thoroughly mixed up and then pull 30 slips
out one by one.
2. Random number table:
Advantages:
• This method eliminates personal bias and more
representative than non probability sampling.
• The results are accurate as the sample size increases.
Disadvantages:
• It requires complete and up to date list of the
population.
• If unit selected in a study is random, samples are
dispersed widely.
• This method can be disruptive to isolate members from
a group.
• Numbering the population is too much tedious and
time consuming.
Systematic sampling:
• Individuals are chosen at regular intervals (for
example: every sixth unit) from the sampling frame
• Ideally, we randomly select a number to tell us
where to start selecting individuals from the list
For example, a systematic sample is to be selected
from 100 MBBS students of Batch 2020. The sample
size selected is 35. Then, the
Sampling interval(K) = Total population/sample size
= 100/35
= 2.8
Advantages:
• Simple and easy
• It gives more precise results than simple random
sampling when population is homogenous.
• The sample is spread more evenly over the
population.
• The time and labour involved in the collection of
sample is relatively small.
Disadvantages:
• The up to date and complete list of population is
mandatory.
• The system may interact with some hidden pattern
in the population. Ex: every 3rd MBBS student may
always be a boy.
Stratified random sampling:
• When the population characteristics are
heterogeneous then this method is applied to
obtain efficient results.
STEPS:
Stratify the target population into number of subgroups
known as strata such that:
• The units within each sub group or strata are as
homogenous as possible.
• The differences between various strata are marked as
far as possible.
• Various strata are non-overlapping. i.e. each and every
unit of the population belongs to one and only one
stratum.
• The units are drawn using simple random or
systematic sampling from each of the “K” strata.
Advantages:
• Every unit in a stratum has a chance of being
selected.
• Minimum knowledge of population needed.
• Easy to analyze data.
• Using the same sampling fraction for all strata
ensures proportionate representation in the sample
of the characteristic being stratified.
Disadvantages:
• It requires sample frame.
• The sampling frame has to be prepared separately
for each stratum.
• Not useful when there are no homogenous
subgroups.
Cluster sampling:
• It is a method in which population is classified into
different sub groups called cluster in such a way that
the characteristics within the cluster are heterogeneous
and between the clusters are homogeneous.
• Used where a complete list of subjects is impossible or
impractical to construct.
• A random sampling technique is then used on any
relevant clusters to choose which clusters to use in
study.
• Often used to evaluate vaccination coverage in EPI
Advantages:
• It cuts down cost of preparing sampling frame and
has a greater speed.
• The sampling frame is required only for the selected
clusters and individuals in selected clusters.
• One can take a larger sample with cluster sampling
than with the other methods for given fixed budget.
Disadvantages:
• Higher sampling error.
• Further analysis is difficult.
• Standard errors of the estimates are high compared
to other sampling designs with same sample size.
• Cluster may not be the representative of the whole
population.
Multistage sampling:
• A complex form of cluster sampling and involves
several stages in which the sampling process is carried
out
• In the first stage, large groups or clusters are selected.
• These clusters are designed to contain more
population units than are required for the final
sample.
• In the second stage, population units are chosen from
selected clusters to derive a final sample.
• This type of sampling is frequently used in national
level health research.
Advantages:
• It is flexible and economic.
• It does not require a complete list of members.
• It greatly reduces the field cost.
Disadvantages
• Sampling error is increased compared with simple
random sampling of the same size.
NON-PROBABILITY SAMPLING
• Non- probability sampling is the process where the
probability of selection of the elements is not
considered.
• The choice of the selection of the items depends
largely upon the judgment of the researcher.
When to use non-probability sampling techniques:
• This type of sampling can be used when demonstrating
that a particular trait exists in the population.
• It can also be used when the researcher aims to do
a qualitative, pilot or exploratory study.
• It can be used when randomization is impossible like
when the population is almost limitless.
• It can be used when the research does not aim to
generate results that will be used to generalize to the
entire population.
• It is also useful when the researcher has limited
budget, time and workforce.
Techniques of non-probability sampling:
• Purposive or Judgmental Sampling
• Convenient Sampling
• Quota Sampling
• Snowball Sampling
Convenience sampling:
• Haphazard/Accidental sampling
• The researcher using such a sample cannot
scientifically make generalizations about the total
population.
• It involves the sample being drawn from that part
of the population which is close to hand.
• A sample that can be taken easily without random
selection.eg. People walking by on the street or
samples which are easily available without much
effort on time, energy and cost.
Examples:
• A reporter can take the views of people in certain
issues at street.
• In clinical practice, we might use clients who are
available to us as our sample
Advantages:
• Useful for making pilot studies, particularly for
testing the research instruments.
• Cheap and quick.
Disadvantages:
• Difficult to generalize the population from sampling
results.
• Result obtained by this method are generally
biased and unsatisfactory, since the result obtained
can hardly be representative of the population
parameters.
• Sample unit is selected deliberately or purposively
depending upon the objective of study.
• The researcher includes only those items in the
sample which he thinks are most typical with
regards to the characteristics under investigation.
Purposive/Judgmental sampling:
Advantages:
• It is very cheap and if selection is done carefully,
gives relevant results.
Disadvantages:
• It is highly subjective in nature since the selection
of the sample entirely depends upon the personal
convenience, beliefs and prejudices of the
investigator.
Quota sampling:
• In quota sampling, a population is first segmented
into mutually exclusive sub-groups i.e. quota just
as in stratified sampling.
• Then judgment or convenience sampling is used to
select the subjects or units from each segment
based on a specified proportion.
Advantages:
• Easy to conduct and easy to get the information in
short period of time.
• The cost and time involved in getting information from
the sample will be relatively less for quota sampling.
• It is very useful in public opinion studies, election
forecast polls, as there is not sufficient time to adopt a
probability sampling scheme.
Disadvantages
• Does not represent the population.
• Selection of the sample is non-random so the
samples may be biased.
Snowball sampling:
• It is also called as a chain or a network sampling.
• This technique can be used to assess “hard to
reach” or hidden population like: drug addicts,
homeless people, individuals with HIV/AIDS, STIs,
prostitutes and so on.
Advantages:
• This method is cheap, simple and cost effective.
• The chain of referral allows researcher to reach
population that are difficult to sample when other
sampling techniques are used.
Disadvantage:
• The main drawback of this sampling technique is
that the sample is not representative of the
population.

sampling & sampling techniques.pdf techniques.pdf

  • 1.
  • 2.
    WHAT IS APOPULATION? WHAT IS A SAMPLE?
  • 3.
    Population • The groupof individuals under study is known as a population or a universe. • Types of population (based on countability): 1. Finite population: which can be counted 2. Infinite population: cannot be counted
  • 4.
    Sample • Finite subsetof individuals in a population • Sample size: total no. of individuals in a sample
  • 5.
    Sampling process - processof selection of sample from the population
  • 6.
    Examples of sample/sampling: •A housewife just picks up few grains of rice from the cooking vessel and gets a fairly good idea whether entire lot of rice is fully cooked or it requires more cooking. • Blood test, stool test, urine test
  • 7.
    Objectives of sampling •To minimize time, money and manpower, without losing the accuracy of the conclusion • Estimation of population parameters from the sample statistics • To test the hypothesis about the population from which the sample/samples are drawn
  • 9.
    • PARAMETER - Valuecalculated from a defined population - Constant value - Example: Mean (µ), standard deviation (σ), population size • STATISTIC - Value calculated from sample - May change according to the sample - Example: Mean (x̅), Standard deviation (s), sample size
  • 10.
    Sampling frame: • Thelist of population units from which the sample units are to be selected is sampling frame. • A good sampling frame is crucial to good sampling.
  • 11.
    Qualities of samplingframe • It should be exhaustive. i.e. all population units must be included in the list. • It should be up to date. • The unit must not be repeated in the list. • It must be maintained by an authority that can be relied upon.
  • 12.
  • 14.
    PROBABILITY SAMPLING/RANDOM SAMPLING: •For each element in the sample, the probability is known and non-zero • Every element of the population has the same chance at being included in the sample
  • 15.
    Simple random sampling: •technique of drawing a sample in such a way that each unit of the population has equal and independent chance of being selected • The sample can be selected either with replacement or without replacement. • It can be done in two ways: 1. Lottery method 2. Random number table
  • 16.
    1. Lottery Method: •Each of the ‘N’ population member is assigned a unique number • The numbers are placed in a bowl and thoroughly mixed. Then a blind folded researcher selects ‘n’ numbers. • Population members having the selected numbers are included in the sample.
  • 17.
    • For example;let us assume that we want to put 30 patients on a clinical trial out of 100 patients. To ensure randomness of selection of 30 patients, we may adopt lottery method by numbering the total individuals from 1 to 100. Then we note down the serial numbers of patients on 100 slips of papers, put all the slips into a drum and shake well so that all the slips are thoroughly mixed up and then pull 30 slips out one by one.
  • 18.
  • 19.
    Advantages: • This methodeliminates personal bias and more representative than non probability sampling. • The results are accurate as the sample size increases.
  • 20.
    Disadvantages: • It requirescomplete and up to date list of the population. • If unit selected in a study is random, samples are dispersed widely. • This method can be disruptive to isolate members from a group. • Numbering the population is too much tedious and time consuming.
  • 21.
    Systematic sampling: • Individualsare chosen at regular intervals (for example: every sixth unit) from the sampling frame • Ideally, we randomly select a number to tell us where to start selecting individuals from the list
  • 22.
    For example, asystematic sample is to be selected from 100 MBBS students of Batch 2020. The sample size selected is 35. Then, the Sampling interval(K) = Total population/sample size = 100/35 = 2.8
  • 23.
    Advantages: • Simple andeasy • It gives more precise results than simple random sampling when population is homogenous. • The sample is spread more evenly over the population. • The time and labour involved in the collection of sample is relatively small.
  • 24.
    Disadvantages: • The upto date and complete list of population is mandatory. • The system may interact with some hidden pattern in the population. Ex: every 3rd MBBS student may always be a boy.
  • 25.
    Stratified random sampling: •When the population characteristics are heterogeneous then this method is applied to obtain efficient results.
  • 26.
    STEPS: Stratify the targetpopulation into number of subgroups known as strata such that: • The units within each sub group or strata are as homogenous as possible. • The differences between various strata are marked as far as possible. • Various strata are non-overlapping. i.e. each and every unit of the population belongs to one and only one stratum. • The units are drawn using simple random or systematic sampling from each of the “K” strata.
  • 28.
    Advantages: • Every unitin a stratum has a chance of being selected. • Minimum knowledge of population needed. • Easy to analyze data. • Using the same sampling fraction for all strata ensures proportionate representation in the sample of the characteristic being stratified.
  • 29.
    Disadvantages: • It requiressample frame. • The sampling frame has to be prepared separately for each stratum. • Not useful when there are no homogenous subgroups.
  • 30.
    Cluster sampling: • Itis a method in which population is classified into different sub groups called cluster in such a way that the characteristics within the cluster are heterogeneous and between the clusters are homogeneous. • Used where a complete list of subjects is impossible or impractical to construct. • A random sampling technique is then used on any relevant clusters to choose which clusters to use in study. • Often used to evaluate vaccination coverage in EPI
  • 32.
    Advantages: • It cutsdown cost of preparing sampling frame and has a greater speed. • The sampling frame is required only for the selected clusters and individuals in selected clusters. • One can take a larger sample with cluster sampling than with the other methods for given fixed budget.
  • 33.
    Disadvantages: • Higher samplingerror. • Further analysis is difficult. • Standard errors of the estimates are high compared to other sampling designs with same sample size. • Cluster may not be the representative of the whole population.
  • 34.
    Multistage sampling: • Acomplex form of cluster sampling and involves several stages in which the sampling process is carried out • In the first stage, large groups or clusters are selected. • These clusters are designed to contain more population units than are required for the final sample. • In the second stage, population units are chosen from selected clusters to derive a final sample. • This type of sampling is frequently used in national level health research.
  • 36.
    Advantages: • It isflexible and economic. • It does not require a complete list of members. • It greatly reduces the field cost.
  • 37.
    Disadvantages • Sampling erroris increased compared with simple random sampling of the same size.
  • 38.
    NON-PROBABILITY SAMPLING • Non-probability sampling is the process where the probability of selection of the elements is not considered. • The choice of the selection of the items depends largely upon the judgment of the researcher.
  • 39.
    When to usenon-probability sampling techniques: • This type of sampling can be used when demonstrating that a particular trait exists in the population. • It can also be used when the researcher aims to do a qualitative, pilot or exploratory study. • It can be used when randomization is impossible like when the population is almost limitless.
  • 40.
    • It canbe used when the research does not aim to generate results that will be used to generalize to the entire population. • It is also useful when the researcher has limited budget, time and workforce.
  • 41.
    Techniques of non-probabilitysampling: • Purposive or Judgmental Sampling • Convenient Sampling • Quota Sampling • Snowball Sampling
  • 42.
    Convenience sampling: • Haphazard/Accidentalsampling • The researcher using such a sample cannot scientifically make generalizations about the total population. • It involves the sample being drawn from that part of the population which is close to hand. • A sample that can be taken easily without random selection.eg. People walking by on the street or samples which are easily available without much effort on time, energy and cost.
  • 43.
    Examples: • A reportercan take the views of people in certain issues at street. • In clinical practice, we might use clients who are available to us as our sample
  • 44.
    Advantages: • Useful formaking pilot studies, particularly for testing the research instruments. • Cheap and quick.
  • 45.
    Disadvantages: • Difficult togeneralize the population from sampling results. • Result obtained by this method are generally biased and unsatisfactory, since the result obtained can hardly be representative of the population parameters.
  • 46.
    • Sample unitis selected deliberately or purposively depending upon the objective of study. • The researcher includes only those items in the sample which he thinks are most typical with regards to the characteristics under investigation. Purposive/Judgmental sampling:
  • 47.
    Advantages: • It isvery cheap and if selection is done carefully, gives relevant results.
  • 48.
    Disadvantages: • It ishighly subjective in nature since the selection of the sample entirely depends upon the personal convenience, beliefs and prejudices of the investigator.
  • 49.
    Quota sampling: • Inquota sampling, a population is first segmented into mutually exclusive sub-groups i.e. quota just as in stratified sampling. • Then judgment or convenience sampling is used to select the subjects or units from each segment based on a specified proportion.
  • 50.
    Advantages: • Easy toconduct and easy to get the information in short period of time. • The cost and time involved in getting information from the sample will be relatively less for quota sampling. • It is very useful in public opinion studies, election forecast polls, as there is not sufficient time to adopt a probability sampling scheme.
  • 51.
    Disadvantages • Does notrepresent the population. • Selection of the sample is non-random so the samples may be biased.
  • 52.
    Snowball sampling: • Itis also called as a chain or a network sampling. • This technique can be used to assess “hard to reach” or hidden population like: drug addicts, homeless people, individuals with HIV/AIDS, STIs, prostitutes and so on.
  • 53.
    Advantages: • This methodis cheap, simple and cost effective. • The chain of referral allows researcher to reach population that are difficult to sample when other sampling techniques are used.
  • 54.
    Disadvantage: • The maindrawback of this sampling technique is that the sample is not representative of the population.