Sampling Part-1

Dr. S. B. Chandak
Asst. Prof.,Mathematics
L. D. College of Engineering, Ahmedabad
[email protected]
9214663968
 Sampling
Population
In statistical investigation the assessment of the general
magnitude and study of variation with respect to one ofor
more characteristics relating to indivuals belonging to a
group. This group of individuals under study is called
population or universe. The population may be finite or
infinite. It is denoted by (N).
 Sampling
Sample
A finite subset of statistical individuals in a population
is called a sample and the number of individuals in a
sample is called sample size. It is denoted by (n).
The process of selecting the representative part of a
population for the purpose of determining
parameters or the characteristics of the whole
sample.
 Sampling
In sampling there are many observations and
measurements are derived on different properties
(such as length, weight, color, and location) of
observable bodies distinguised as independent
objects or individuals.
• With the help of sampling and its theory of
measured data which employed to guide an
individual or authority to make decision or policy
concern with the people.
 Sampling
The sampling process comprises of several stages:
1. Defining the population to be concern.
2. Specifying the sampling frame, a set of items or events
possible to measure.
3. Specifying a sampling method for selecting items or
events for the frame.
4. Determining the sample size and implementing the
sampling method.
5. Data collecting.
▪ There are two types of sampling: large sampling and
small sampling.
▪ The large sampling is used for samples when the no.
of samples are greater than 30 (n>30).
 Sampling
Type of the sampling
Some of the commanly know and frequently
used types of sampling are
(i) Purposive Sampling
(ii) Random Sampling
(iii)Stratified Sampling
 Sampling
Purposive Sampling
This type of the sampling is one in which the sample
units are selected with definite purpose in view. This
sampling suffers from the draw back of favouritism
and nepotism and dose not give a representative
sample of the population.
 Sampling
Random Sampling or Simple Sampling
Simple sample is random sampling in which each
units of population has an equal chance say p, of
being included in the sample and this probability is
independent of pervious drawings. Thus a simple
sample of size n from a population may be indetified
with series of n independent trails with constant
probability “p” of sussess for each trial.
 Sampling
Stratified Sampling
When the entire population is divided into a number of
homogenous group, usally termed as strata, which differ from
one another but each of these group is homogenous, within it
self. Then units are selected as random from each of these
stratum, the sample size in each stratum varies according to the
relative importance of the stratum in the population. The
sample which is aggregate of the sample units of each of the
stratum, is termed as stratified sample and the technique of
drawing this sample is know as stratified sampling.
 Sampling
Parameters:
The statisticsl constants of population such as mean
( ) , standard devation (  ), population proportion (P)
etc. and for sample mean( x ), standard deviation (s),
sample proportion (p) etc., are know as prameters

Standard Error: The standard deviation of the

sampling distribution of a statistic is known as its
standard error SE. Standard error plays a very
important role in,the large sample theory and forms
the basis of the testing of hypothesis.
Sampling
 Sampling
Utility of standard error:

Standard error (SE) plays a important role in the large

sample theory and it forms the basis of the testing of
hypothesis.

For Large sample test

 Sampling
Test of Siginificance:
The sampling theory is the study of the test of
siginificance, which enable us to decide on the basis
of the sample results if,
(i) The devation between the observed sample
statistic and the hypothetical parameter values or
(ii) the devation between two independent sample
statitic is siginificant or might be attributed to
change.
 Sampling
For Large sample n, all distributions, e.g. Binomial,
Poisson, Negative Binomial, Hypergeometic, Chi-
Square are nearly closed to normal probability curve,
we use the normal test of siginificance for large
sample.

For Small sample well known test of siginificance are

t- test, F-test and Fisher Z-transformation.
 Sampling
Hypothesis:
The technique of randomisation used for

selection of sample units makes the test of

significance valid for us is known as hypothesis. There
are two types of the hypohesis

(I) Null Hyphothesis

(ii) Alternative Hypothesis

 Sampling
Null Hypothesis (Ho) : A definite statement about the
population parmeter , such a hypothesis which is usually
a hypothesis of no difference is called Null Hypohesis.
According to R. A. Fisher null hypothesis is the hypothesis
which is tested for possible rejection under the
assumption that it is true. It is denoted by Ho. It asserts
that there is no significant difference between the
statistic and the population parameter and whatever
observed difference exists, is merely due to the
fluctuations in sampling from the same population.
 Sampling
Alternative Hypothesis(H1) : Any hypothesis which is
complementary to the null hypothesis is called an
alternative hypothesis. It is denoted by H1. It is set in
such a way that the rejection of null hypothesis
implies the acceptance of alternative hypothesis.
For example if we want to test the null hypothesis
that population mean is  o i.e.
H0 :  =  o
The alternative hypothesis could be
H1 :    o (Two tailed test )
H1 :    o ( Right tailed test )
H1 :    o ( Left tailed test )
 Sampling
• Errors in Hypothesis Testing:
The main objective in sampling theory is to draw valid
inferences about the population parameters on the
basis of the sample results. There is every chance that
a decision regarding a null hypothesis may be correct
or may not be correct. There are two types of errors.
 Sampling
• Type I error ( ) : It is the error of rejecting the null
hypothesis H0, when it is true. It occurs when a null
hypothesis is true, but the difference of means is
significant and the hypothesis is rejected. If the
probability of making a type I error is denoted
by , the level of significance, then the probability
of making a correct decision is (1 -  ).
Type I error Reject H0 when it is true
 Sampling
Type II error ( ) : It is the error of accepting the null
hypothesis H0, when it is false. It occurs when a null
hypothesis is false, but the difference of means is
insignificant and the hypothesis is accepted. The
probability of making a type II error is denoted by ( ) .
Type two error : Accept Null Hypothesis when it is
wrong.
Sampling
Sampling
Sampling
Sampling
Sampling
Sampling
Sampling
Sampling
Sampling
 Sampling
• The various steps in testing of a statistical
hypothesis are as follows
(i) Null Hypothesis: Set up the null hypothesis H0
(ii) Alternative Hypothesis : Set up the Alternative
hypothesis H1
(iii) Level of significance: Set level of siginificance 
(iv) Test statics : Calculate test statistics


 Sampling
(i) Critical value: Find the critical value (tabulated) z
of z for given level of significance.
(vi) Compare the calulated and tabulated value of Z.
If z  z then we accepet our null hyphothesis.
if z  z then we reject our null hyphothesis.
 Sampling
Sampling Distribution of Proportions
Test for Single Proportion
The sampling proportions for an attribute of a
characteristics is found by dividing the number of
sample observation having that attribute by sample
size.
Let X be the number of success in n trails with
constant probability ‘P’ of success in each trial. Then
E(X) = nP and V(X) = nPQ where Q = 1-P
 Sampling
Let Observed proportion of Success is p= X/n
E(X/n) = (1/n) E(X) & V(X/n) = (1/n2 ) V(X)
=(1/n) nP = (1/n2 )nPQ
= P =PQ/n
Standard Error of proportion of success = PQ
n
Let X & (X/n) both tends to normal distribution hence
the test static under considertation will be
 Sampling
If we have a sample from finite population of size N
then  N − n  PQ 
S .E.( p ) =   
 n −1  n 

For Normal variance z lies between  3

 P(− 3  z  3) = 0 .9973
 P( z  3) = 0 .9973
 X − E( X ) 
 P  3  = 0 .9973
 V (X ) 

( )
 P E ( X ) − 3 V ( X )  X  E ( X ) + 3 V ( X ) = 0 .9973
 Sampling
There probable limits for the observed proportion of
success are

Since p is the unbiased estimeter of proportion P.

Hence if P is not known the the probable lomits for
the proportion of succes are
Sampling
Sampling
Sampling
Sampling
 Sampling
Example 1
Sampling
 Sampling
Example 2
Sampling
 Sampling
Example 3
Sampling
 Sampling
Test for significance for single mean
Sampling
Sampling
 Sampling
Example 2
 Sampling
Test for significance for difference of mean
Sampling
 Sampling
Example: 1
 Sampling
Test for significance for difference of standard deviations
Sampling
Sampling
Sampling
Sampling
 Sampling
Example 1

Solution: Here n= 50,

Probability for getting head p= ½, q=1/2

Mean = = np = 960(1/2)= 480,

variance = npq=960(1/2)(1/2) = 240, hence

=

Number of success x = 183

Sampling