Sampling Techniques

SAMPLING TECHNIQUES
By
Dr. H.S. ABZAL BASHA, M.B.A., Ph.D.
Assistant Professor,
Department of Management Studies,
G. Pullaiah College of Engineering & Technology, Kurnool.

CONTENTS
 Introduction
 Need for sampling
 Sampling Process
 Essentials of Sampling
 Methods of Sampling
 Probability Sampling
 Non Probability Sampling
 Errors in Sampling
 References

INTRODUCTION
 Population/Universe: in statistics denotes the
aggregate from which sample (items) is to be taken.
 A population can be defined as including all people
or items with the characteristic one wishes to
understand.
 Because there is very rarely enough time or money
to gather information from everyone or everything
in a population, the goal becomes finding a
representative sample (or subset) of that population.

INTRODUCTION
 Sampling frame is the list from which the
potential respondents are drawn.
 A sample is “a smaller (but hopefully
representative) collection of units from a
population used to determine truths about that
population” (Field, 2005)

SAMPLING
 Sampling: the process of learning about
population on the basis of sample drawn from it.
 Three elements in process of sampling:
Selecting the sample
Collecting the information
Making inference about population
 Statistics: values obtained from study of a
sample.
 Parameters: such values from study of
population.

NEED FORSAMPLING
DATA
(Acc.to source)
Primary Secondary
1.ORIGINALIN
CHARACTER
2. GENERATED IN LARGE
NO. OFSURVEYS
OBTAINED FROM
1. PUBLISHED SOURCES
2.UNPUBLISHED
SOURCES

NEED FORSAMPLING
 When secondary data are not available for the
problem under study , primary data is
collected.
 Two methods –
 Census method or complete enumeration method
 Sample method

CENSUS (CompleteEnumeration Survey)
 Merits
 Data obtained from each and every unit of
population.
 Results: more representative, accurate, reliable.
 Basis of various surveys.
 Demerits
 More effort ,money , time.
 Big problem in underdeveloped countries.

ADVANTAGES OF SAMPLING
 Less resources (time, money)
 Less workload.
 Gives results with known accuracy that can be
calculated mathematically.

THEORETICAL BASISOF SAMPLING
 On the basis of sample study we can predict
and generalize the behavior of mass
phenomena.
 There is no statistical population whose
elements would vary from each other without
limit.

THEORETICAL BASISOF SAMPLING
 Law of Statistical Regularity-
 Sample is taken at random from a population, it is
likely to possess same characteristics as that of
population.
 Law of inertia of large numbers-
 Larger the size of sample, more accurate the results
are likely to be.

SAMPLING PROCESS
 Defining the population of concern.
 Specifying a sampling frame, a set of items
or events possible to measure.
 Specifying a sampling method for selecting
items or events from the frame.
 Determining the sample size.
 Implementing the sampling plan.
 Sampling and data collection

ESSENTIALS OF SAMPLING
 Representativeness- ensure by random
selection
 Adequacy - sample size
 Independence - same chance of selection
 Homogeneity - no basic difference in nature of
units.

SAMPLINGMETHODS
NON
PROBABILITY
PROBABILITY MIXED
JUDGMENT
QUOTA
CONVENIENCE
SNOWBALL
SIMPLERANDOM
STRATIFIED
RANDOM
SYSTEMATIC
CLUSTER
MULTISTAGE
MULTIPHASE
LOTQUALITY
ASSURANCE

SIMPLE RANDOM SAMPLING
 Each unit has an equal opportunity of being
selected.
 Chance determines which items shall be
included.
 The sample is a simple random sample if any of
the following is true (Chou) –
 All items selected independently.
 At each selection , all remaining items have same
chance of being selected.
 All the possible samples of a given size are equally
likely to be selected.

Simple or unrestricted randomsampling
Lottery method Random numbertables

SIMPLE RANDOM SAMPLING
 Merits
 No personal bias.
 Sample more representative of population.
 Accuracy can be assessed as sampling errors follow
principals of chance.
 Demerits
 Requires completely catalogued universe.
 Cases too widely dispersed- more time and cost.

STRATIFIED RANDOM SAMPLING
 Universe is sub divided into mutually
exclusive groups.
 A simple random sample is then chosen
independently from each group.

 Issues involved in stratification
 Base of stratification
 Number of strata
 Sample size within strata.
Sample size withinstrata
Proportional
(proportion in eachstratum)
Disproportional
(equal no. in eachstratum)

Strata
Formation
General(30%)
SC(15%)
ST
(25%)
OBC
(30%
)
KURNOOL
CITY
150
Random
sampling
Forsample sizeof1000
STRATIFIED RANDOM SAMPLING -EXAMPLE

 Merits
 More representative.
 Greater accuracy.
 Greater geographical concentration.
 Demerits
 Utmost care in dividing strata.
 Skilled sampling supervisors.
 Cost per observation may be high.

SYSTEMATIC SAMPLING
 Selecting first unit at random.
 Selecting additional units at evenly spaced
intervals.
 Complete list of population available.
k=N/n
k=samplinginterval
N=universe size
n=Samplesize
Classof 95students: roll no. 1 to 95
Sample of 10students
k=9.5 or10
1ststudent random then every10th

SYSTEMATIC SAMPLING
 Merits
 Simple and convenient.
 Less time consuming.
 Demerits
 Population with hidden periodicities.

CLUSTER SAMPLING
 A sampling technique in which the entire
population of interest is divided into groups, or
clusters, and a random sample of these clusters is
selected.
 Each cluster must be mutually exclusive and
together the clusters must include the entire
population .
 After clusters are selected, then all units within the
clusters are selected. No units from non-selected
clusters are included in the sample.

CLUSTER SAMPLING
 In cluster sampling, the clusters are the
primary sampling unit (PSU’s) and the units
within the clusters are the secondary sampling
units (SSU’s)

STRATIFICATIONV/S CLUSTERING
Stratification Clustering
All strata are represented in
the sample.
Only a subset of clusters are
in thesample.
Less errorcomparedto
simple random.
More errorcompared to
simple random.
More expensive to obtain
stratification information
before sampling.
Reduces costs to sample
only some areas or
Organizations.

CLUSTER SAMPLING- STEPS
 Identification of clusters
 List all cities, towns, villages & wards of cities with
their population falling in target area under study.
 Calculate cumulative population & divide by 30, this
gives sampling interval.
 Select a random no. less than or equal to sampling
interval having same no. of digits. This forms 1st
cluster.
 Random no.+ sampling interval = population of 2nd
cluster.
 Second cluster + sampling interval = 3rd cluster.
 Last or 30th cluster = 29th cluster + sampling interval

• III 1500 6500
• IV 4000 10500 3
• V 5000 15500 4,5
• VI 2500 18000 6
• VII 2000 20000 7
• VIII 3000 23000 8
XIII. 3500
XIV. 2000
XV. 3000
• IX 3500 26500 9
• X 4500 31000 10
• XI 4000 35000 11,12
• XII 4000 39000 13
• 44000 14,15
• 46000
• 49000 16
• Freq cf cluster • XVI 3500 52500 17
• I 2000 2000 1 • XVII 4000 56500 18,19
• II 3000 5000 2 • XVIII 4500 61000 20
30
XXVIII
XXIX
XXX
1000
1000
1000
• XIX 4000 65000 21,22
• XX 4000 69000 23
• XXI 2000 71000 24
• XXII 2000 73000
• XXIII 3000 76000 25
• XXIV 3000 79000 26
• XXV 5000 84000 27,28
• XXVI 2000 86000 29
• XXVII 1000 87000
• 88000
• 89000
• 90000
• 90000/30 = 3000 samplinginterval
CLUSTERSAMPLING

CLUSTER SAMPLING
 Merits
 Most economical form of sampling.
 Larger sample for a similar fixed cost.
 Less time for listing and implementation.
 Reduce travel and other administrative costs.
 Demerits
 May not reflect the diversity of the community.
 Standard errors of the estimates are high, compared to
other sampling designs with same sample size .

MULTISTAGE SAMPLING
 Sampling process carried out in various stages.
 An effective strategy because it banks on multiple
randomizations.
 Used frequently when a complete list of all
members of the population does not exist and is
inappropriate.

MULTISTAGE SAMPLING
 Merits
 Introduces flexibility in the sampling method.
 Enables existing divisions and sub divisions of
population to be used as units.
 Large area can be covered.
 Valuable in under developed areas.
 Demerits
 Less accurate than a sample chosen by a single
stage process.

MULTIPHASE SAMPLING
 Used for studies to be carried out in multiple
phases.
 For e.g. A cross - sectional study on nutrition
may be carried out in phases
Phase-1: K.A.P. study in allfamilies
Phase-2: Dietary assessment insubsample
Phase-3:anthropometric examination insub-
sample of family members covered in2nd phase

LOT QUALITYASSURANCE SAMPLING
 Originated in the manufacturing industry for
quality control purposes.
 Manufacturers were interested in determining
whether a batch, or lot, of goods met the
desired specifications.
 The only outcome in this type of sampling is
“acceptable” or “not acceptable”

 The sample size is the number of units that are
selected from each lot.
 The decision value is the number of “defective”
items that need to found before the lot is deemed
unacceptable.
 There are two types of risks
 the risk of accepting a “bad” lot, referred to as
Type I error
 the risk of not accepting a “good” lot, referred to as
Type II error.

 Information from lots can be combined to
obtain the overall proportion of defects.
 The population is first divided into a complete
set of non-overlapping lots.
 Samples are then taken from every lot, and the
proportion of defective items in each lot is
calculated.
 The LQAS method is an example of stratified
sampling, where the lots play the role of the
strata.

LOT QUALITYASSURANCE
SAMPLING
 The advantage of the LQAS method over a
traditional stratified sampling design is that
the response for each lot is binary (acceptable
or not), and therefore smaller sample sizes can
be used.

LOT QUALITYASSURANCE
SAMPLING
 Can be used for evaluating a number of health
programmes e.g. immnunisation coverage ,
knowledge of ORS, etc
 Despite successful trials of LQAS in health
surveys , its routine use has not been
established yet.

ERRORS
SAMPLING
ERRORS
NON SAMPLING
ERRORS
SAMPLE
SAMPLEAND
CENSUS

JUDGMENT SAMPLING
 Judgment/Purposive/Deliberate sampling.
 Depends exclusively on the judgment of
investigator.
 Sample selected which investigator thinks to
be most typical of the universe.

JUDGMENT SAMPLING
 Merits
 Small no. of sampling units
 Study unknown traits/case sampling
 Urgent public policy & business decisions
 Demerits
 Personal prejudice & bias
 No objective way of evaluating reliability of results

JUDGMENT SAMPLING-
EXAMPLE
CLASSOF20STUDENTS
Sample size for astudy=8
JUDGMENT
SAMPLE OF8
STUDENTS

CONVENIENCE SAMPLING
 Convenient sample units selected.
 Selected neither by probability nor by
judgment.
 Merit – useful in pilot studies.
 Demerit – results usually biased and
unsatisfactory.

CONVENIENCE SAMPLING -EXAMPLE
Classof 100students
20 Students selected asper
convenience

QUOTASAMPLING
 Most commonly used in non probability
sampling.
 Quotas set up according to some specified
characteristic.
 Within the quota , selection depends on
personal judgment.
 Merit- Used in public opinion studies
 Demerit – personal prejudice and bias

Quota
Formation
Interview 500people
judgement
Radiolistening
survey
60%
housewives
25%
farmers
15%children
under age15
300
125
Personal
75
500people
QUOTA SAMPLING -EXAMPLE

SNOWBALLSAMPLING
 A special non probability method used when
the desired sample characteristic is rare.
 It may be extremely difficult or cost
prohibitive to locate respondents in these
situations.
 Snowball sampling relies on referrals from
initial subjects to generate additional subjects.

SNOWBALLSAMPLING - STEPS
 Make contact with one or two
cases in the population.
 Ask these cases to identify further cases.
 Ask these new cases to identify further new
cases.
 Stop when either no new cases are given or the
sample is as large as is manageable.

SNOWBALLSAMPLING
 Merit
 access to difficult to reach populations (other
methods may not yield any results).
 Demerit
 not representative of the population and will result
in a biased sample as it is self-selecting.

NON SAMPLING ERRORS
• Data specification inadequate & inconsistent
with respect to objective of census.
• Inaccurate or inappropriate methods of
interview, observation, definitions.
• Lack of trained & experienced investigators.
• Errors due to non response.
• Errors in data processing operations
• Errors committed during presentation.
MORE IN COMPLETEENUMERATIONSURVEY

REFERENCES
 Methods in Biostatistics by BK Mahajan
 Statistical Methods by SP Gupta
 Basic & Clinical Biostatistics by Dawson and
Beth.

Sampling Techniques

More Related Content

What's hot (19)

Similar to Sampling Techniques (20)

More from Dr. Abzal Basha H S (20)

Recently uploaded (20)

Sampling Techniques