Sampling: Iiird Year Resident

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 26

SAMPLING

DR ANSHULI TRIVEDI
IIIRD YEAR RESIDENT
DEFINITION-
 Sampling is that part of statistical practice concerned
with the selection of an unbiased or random subset of
individual observations within a population of
individuals intended to yield some knowledge about the
population of concern, especially for the purposes of
making predictions based on statistical inference.
 Sampling frame-The sampling frame operationally
defines the target population from which the sample is
drawn and to which the sample data will be generalized.
NEED FOR SAMPLING-
 Advantages of sampling-
 cost is lower
 data collection is faster
 Since the data set is smaller it is possible to
ensure homogeneity and to improve the accuracy
and quality of the data.
CHARACTERISTIC OF SAMPLING METHOD-
 Precision-Depends on sample size.
 Precision=√n
 s n=sample size, s=SD
 By increasing sample size, increase accuracy and
decrease margin of error
 • The smaller the population, the larger the needed ratio
of the sample size to the population size
 • Aim for is a 95% confidence level.
 Unbiased character.
PROCESS OF 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


TYPES OF SAMPLING-
 Probability
 Non probability
PROBABILITY SAMPLING-
 A probability sampling scheme is one in which every unit
in the population has a chance (greater than zero) of being
selected in the sample, and this probability can be
accurately determined.
 Types-
Simple Random Sampling
Systematic Sampling
Probability Proportional to Size Sampling
Stratified Sampling
Cluster or Multistage Sampling.
SIMPLE RANDOM SAMPLING:
 Each member of the population has an equal and
independent chance of being selected in the sample.
 Used when the population is small, homogeneous and
readily available-
 Randomness by
 Lottery method(Population is small)
 Table of Random numbers (Population is large).
 Ex-Suppose we have to select a sample (n) of 100 subjects from a
population of 500 (N).
 We first make a serial list of each and every subject of the 500
 subjects in the population. Then from the random number table,
random numbers are selected row wise or column wise.
Random Number Table
32 50 92 46 24 69 48 93 77 87 47 17 29 36 55
81 34 70 46 99 27 95 04 69 59 71 30 74 42 36
45 11 49 20 50 86 16 75 80 55 33 98 93 66 76
13 56 08 38 43 12 11 01 21 41 13 87 08 47 98
64 61 65 94 30 17 51 54 45 85 41 22 96 26 64
38 09 93 01 49 43 06 09 24 42 23 23 21 65 14
95 76 09 00 24 54 15 04 34 41 58 61 05 09 82
97 30 78 89 23 44 66 18 71 83 08 21 74 18 91
STRATIFIED RANDOM SAMPLING:
 Population under study is first divided into
homogeneous groups called STRATA & a sample is
drawn from each Stratum by simple Random
Sampling method
 Useful when population is not homogeneous.

This method is useful where one is interested in analyzing the


data by a certain characteristics of population like,
religion, caste, occupation or age groups, Gender etc.
For example-if we draw a simple random sample from a
population, a sample of 100 may contain only 10 to 15
high income groups, 20 to 30 middle income groups and
80 to 90 low income groups.
Stratify each SE-Class

Take simple random no. from each size as per the sample size.

The advantage is that minority group is not left out but we


require a sampling frame for each stratum separately.
SYSTEMATIC RANDOM SAMPLING: (Quasi)
 A systematic random sample is obtained by
selecting one unit on a random basis and then
choosing additional units at evenly spaced
intervals until the desired number of sample size
is obtained.
 Used when- Systematic sampling is a commonly
employed technique, when
 complete and up to date list of sampling units is

available.
The advantage of systematic sampling is that it is easier
to draw, but the demerit is that the sample may exhibit
a pattern or periodicity.
For example, if there are 100 students (N) in a class and we wish to
select a sample of 20 students (n) from them by systematic random.
first step is to write the names of 100 students in alphabetical order
or their roll numbers one below the other.

In the systematic sampling procedure we divide N by n to get the


sampling fraction (k).Thus in the example k=100/20 = 5.

Next we randomly select any number between 1 to k i.e. between 1


to 5. Suppose the number we select is 4. Then the student number 4
is selected in the sample. Thereafter every kth student is selected in
the sample until we reach the last one.

Thus the student’s corresponding to numbers 4, 9, 14, 19, …..99


are to be selected in the sample.
CLUSTER RANDOM SAMPLING:
 Any method of sampling where in a group is taken as a Sampling Unit is called
Cluster Sampling
 Cluster is a group consisting of units such as villages, wards, factories, slums,
town, children of a school etc.

 A special form of cluster sampling called the


 “30 cluster sampling”, has been recommended by the WHO
 for field studies in assessing vaccination coverage.
 In this a list of all villages (clusters) for a given geographical area is
made.
 30 clusters are selected using Probability Proportional
 to Size (PPS).
 From each of the selected clusters, 7 subjects are
 randomly chosen. Thus a total sample of 30 x 7 = 210 subjects
 is chosen.
 Multistage Sampling-
 In this method, the whole population is divided in first stage
 sampling units from which a random sample is selected.
 The selected first stage is then subdivided into second stage
 units from which another sample is selected.
 Third and fourth stage sampling is done in the same manner if
necessary.
 For example, in an urban survey in a state, a sample of towns may
 be taken first and then in each of the selected towns, a second
 stage sample of households may be taken.
 If needed, further from each of the selected household, a third
stage sample of individuals may be selected. Since the samples are
selected at each stage the method is called ‘Multi stage sampling’.
MULTI PHASE SAMPLING-
 Information is collected from whole sample &
part of information from subsample.
Montux test is done on whole population

Mtx +ve are screened by X-ray

Xray +ve are tested by Elisa


NON PROBABILITY SAMPLING-
 It is a sampling method where some elements of the population
have no chance of selection or where the probability of
selection can't be accurately determined.

 Hence, because the selection of elements is nonrandom, non-


probability sampling does not allow the estimation of sampling
errors. These conditions give rise to exclusion bias placing
limits on how much information a sample can provide about
the population. Information about the relationship between
sample and population is limited, making it difficult to
extrapolate from the sample to the population.
Non-probability Samples
• Convenience
• Purposive
• Quota
Convenience Sample
• Subjects selected because it is easy to access them.
• No reason tied to purposes of research.
• Students in your class, people on State,Street, friends
PURPOSIVE SAMPLING:
(Judgmental Sampling or deliberate sampling)
 Cases that serve specific purpose are chosen for study.
 Ex-Clinical trials are done on volunteers who agree to take
part in study.
QUOTA SAMPLING:

Startified random sampling minus randomization.

Ex-Imagine that in a town there are 50 % farmes ,


25% small busimessman,25% workers.
When we want to take the sample in the population
the sample should have the same % of these
groups.
COMPARISON OF ALL SAMPLING METHODS-

Method Desired size of Merits De merits


target population
Simple Small  Scientific  Needs complete list of
 More study population.
representative  Less number of subjects
sample is not true
representative
 Time & cost due to travel
can be high.
Stratified Small  More  Difficult to divide pop.
representative. into homogenous group.
 Greater accuracy.  Sometimes there may be
 More economical overlap.
Method Desired size of Merits De merits
target population

Systematic Medium Simple & Used only when


convenient to pop. is
adopt. homogenous.

Cluster Large Allow small no.of Sample may not be


target pop. To be representative
sampled.

Multistage Very large Allow large scale Requires sampling


surveys fame only for each
nested unit.

Multiphase Large Less cost,less labor Cumbersome


More purposeful
SAMPLE SIZE DETERMINATION-
 Quantitative data-N=4(SD)² SD= Standard Deviation, E=Allowable
error
 E²
 Ex-Mean pulse rate of a population is believed to be 70/min with
standard deviation of 8 beats. Calculate the minimum size of sample
to verify this. E=+ 1beat at 5% risk.
 N=4×8×8=64
 1×1
Qualitative Data-N=4PQ P=+VE Character,Q=1-P, E=Allowable
error

Ex-Incidence rate in last, Influenza epidemic was found to be 5% of
the population exposed. What should be sample size to find rate in
current epidemic in E=0.005.
P=0.05,Q=0.95,E=0.005 N=4×0.05×0.95=7600
0.005×0.005
Thanks
Happy Diwali

You might also like