0% found this document useful (0 votes)
20 views20 pages

Sampling

The document discusses sampling methods and concepts including population, sample, parameter, statistic, sampling distribution, simple random sampling, sampling with and without replacement, stratified sampling, cluster sampling, advantages of sampling over complete enumeration, biases in surveys, and derivation of expectation and standard error of the sample mean.

Uploaded by

Baban Basak
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
20 views20 pages

Sampling

The document discusses sampling methods and concepts including population, sample, parameter, statistic, sampling distribution, simple random sampling, sampling with and without replacement, stratified sampling, cluster sampling, advantages of sampling over complete enumeration, biases in surveys, and derivation of expectation and standard error of the sample mean.

Uploaded by

Baban Basak
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 20

Sampling

• Population-aggregate objects/individuals
• Sample-A part of the population that is supposed
to represent the population
• Parameter-statistical measure based on
population values
• Statistic-statistical measure based on sample
observation
• Sampling fluctuation-change in value of statistic
from one sample to another
• Sampling distribution-probability distribution of a
statistic
• Statistic has expectation and standard error while
parameter does not have
• Standard error-standard deviation of a statistic
• Sampling-the process of drawing sample from
population
• Random sampling-each member of the population has
definite pre assigned probability of being included in
the sample
• Simple random sampling (SRS)- Each member of
population has same probability of being included in
the sample
• SRSWR-sample units are drawn with replacement
• SRSWOR-sample units are drawn with out
replacement
• Sampling experiment-
• Sample survey-no experiment; only data is collected
through mere observation or asking the respondents
Types of sampling
• Subjective and objective
• Objective sampling is divided into
probabilistic(random, stratified, multistage,
cluster), non probabilistic (convenience,
judgement, quota, snowball) and mixed
• SRSWR and SRSWOR
Advantage of sampling over Complete Enumeration

• Reduction of cost-money and man hours


• Greater scope-Some enquiries require highly trained
personnel or specialised equipment
• Gives data of better quality because of better
supervision, better trained personnel, reduction in
non sampling errors
• Non sampling errors can be gauged
• Only possible way when population is infinite or
hypothetical(tossing a coin infinite times)
• Complete enumeration may be destructive in nature
Different steps in large scale sample survey
• Defining the objectives
• Defining the population
• Determination of data to be collected
• Deciding on method of collection of data
• Getting a sampling frame
• Choice of sampling unit
• Deciding on the type of sampling to be used
• Drawing the sample
• Training of personnel
• Scrutiny of data, tabulation of data
• Statistical analysis
• Reporting and storing the information
Biases in survey
• Response
• Non-response
• Observational
• Interviewer
Techniques of random sampling
• Putting tickets in metallic cylinder and then
thrown into a rotating drum before drawing
each sample unit
• Take packet of cards as similar as can be. Then,
correspond each card with population unit.
Shuffle before each drawing
Random Sampling Number
• Sequence of digits that are perfectly random
order. That is prepared by special devices and
tested for randomness. It is prepared once for
all such that each digit has same probability of
occupying one place. If a digit is blindly selected
from 0, 1, 2,…,9, each digit has 1/10 probability
of being selected. If two consecutive digits are
selected from 00,01,..,99, each has 1/100
probability of being included in sample.
Method of Drawing Sample
• Assigning serial numbers
• Assigning identity numbers
• Start reading random number
• Amounts of rejection
• May have to assign more than one serial
numbers
• Random numbers are divided by 10 and
remainder is taken
Problem
• Consider a population X:1, 2, 7, 5, 9
• Draw all possible samples of size 2 using
• i) SRSWR
• Ii)SRSWOR
• Find population mean and sd.
• Illustrate sampling distribution of sample mean.
Find expectation and standard error of sample
mean
• Draw a SRSWOR sample of size 2 using random
number table.
Derivation of Expectation and standard error
of sample mean in SRSWR
• Suppose
  population consists of N observations
Also, suppose the population mean is and
population sd is . If we draw a sample of size
n, using SRSWR .
• E(
• Se(
• E()=
 
• E(
• V(-
=1/N
• V(V(

=
Prob Distn of xi
xi X1 X2 X3 XN
P(xi=X 1/100=P(xi= 1/100=P(xi= 1/100 1/100
alpha) X 1) X2)
Derivation of Expectation and standard
error of sample mean in SRSWOR
• Suppose
  population consists of N observations
Also, suppose the population mean is and
population sd is . If we draw a sample of size
n, using SRSWOR .
• E(
• Se(
•  P()=?
• P()=1/N
• P ()
=
=)
=
• E()=
 
• E(
• V(-
=1/N
• V(V(

= = [1-
•  )=E[- )]
=,=
=
= -]
= -]=-
• P(
 
• | ).P()
• =

You might also like