Sampling and Sampling Distribution

Population: Total group under discussion or the group to which the results will be
generalized is called population. Number of units in a population is called size of the
population and is denoted by N.
Sample: Any subset of a population is called sample. Number of units in a sample is
denoted by n.
Target Population: A population about which we require information is called a target
population. e.g the students of all the colleges in the province.
Sampled Population: A population from which sample is chosen is known as sampled
Population.
Sampling: The process of drawing a sample from a population is called sampling.
Sampling Unit: A basic element or object which we select for a sample are called sampling
units, if we want to measure the average height of students, the students are sampling units.
Sampling frame: The complete list of all possible sampling units of the population is called
sampling frame.
Parameter: Numerical quantity describing a population called Parameter. Parameters are
usually denoted by Greek letters. Parameters are usually unknown. Parameters are fixed
(constant) values. Population mean (μ), Population Variance (σ2) and Population Proportion
(P).
Statistic: A statistic is a quantity that is calculated from a sample of data. It is used to give
information about unknown values in the corresponding population. For example, the average
of the data in a sample is used to give information about the overall average in the population
from which that sample was drawn.
It is possible to draw more than one sample from the same population and the value of a
statistic will in general vary from sample to sample. For example, the average value in a
sample is a statistic. The average values in more than one sample, drawn from the same
population, will not necessarily be equal.
Census: If all the respondents in a (universe/population) are asked to provide information,
this survey is called a Census.
Sample Survey: The collection of information from a part of the population is called sample
survey. In sample survey only a part of a population (Sample) is studied.
Errors in Sampling: There are two kinds of error may occur in the estimates, obtained
from samples.
(i) Sampling Error (ii) Non-Sampling Error
Sampling Error: The difference between statistic and actual value of population parameter
is called sampling error
Sampling Error = X − 
This error can be minimized by
➢ Increasing the size of the sample
➢ Improving the procedure of sample selection (sample design).
Non–sampling errors: Non-Sampling error are those errors arises due to biases in the
selection of the sample. Non-Sampling error may occur due to many reasons and at all stages
of an enquiry e.g. improper selection of respondents, wrong responses, non-response,
recording errors, selection of improper frames, biases, using poor questionnaire or poor tools
of data collection, poor training of enumerators, poor supervision etc.
Techniques of Sampling: There are two kinds of techniques in the selection of
samples.
(i) Probability Sampling (ii) Non-Probability Sampling
Probability sampling:
➢ Simple random sampling
➢ Systematic sampling
➢ Stratified sampling
➢ Cluster sampling
Non-probability sampling:
➢ Convenience sampling
➢ Judgmental sampling
➢ Quota sampling
➢ Snowball sampling
Accuracy: The degree to which the sample estimate is close to the true population value
maximum accuracy is obtained when the estimate equals the true population value
Precision: The variability of the sample estimate in repeated measurements (across different
samples) maximum precision is obtained when the estimate is the same across all samples.
Standard Error: The standard deviation of sampling distribution of a statistic is called
standard error and is denoted by S.E
Sampling Distribution: Sampling distribution is the probability distribution of the values of
^
statistic, such as sample mean ( X ) sample variance ( S 2 ) sample proportion ( P )

Computed from all possible sample of same size n, which can be drawn from population of
size N, when sampling is done with or without replacement.
Sampling With and Without Replacement
Sampling With Replacement: Sampling is said to be with replacement when we draw a
sampling unit from a population and return it to the population before next unit is drawn. In
this case sampling unit can be chosen more than once.
Sampling Without Replacement: if we do not return the sampling unit to the population and
drawn the next unit, the sampling is said to be without replacement. In this case the sampling
unit can be chosen only once.
Sampling distribution of sample Mean ( X )
The probability distribution of sample means computed from all possible samples of size ‘n’
which can be drawn from population size ‘N’ is called sampling distribution of sample mean.
Properties of Sampling distribution of sample Mean ( X ) (With Replacement)
(i) Mean of the sampling distribution of sample means is equal to the population mean i.e
X = 

(ii) The variance of sampling distribution of sample mean is equal to population variance
divided by sample size.
 2 
X 2
=  
 n 
(iii) The standard deviation of sampling distribution of sample mean is equal to population
standard deviation divided by positive square root of sample size.
  
 x =  
 n

Example: A population consists of 5 values, 2, 4, 6, 8 and 10. Draw all possible random
samples of size 2 with replacement and then
i) Construct the sampling distribution of sample means.
ii) Verify the properties of sampling distribution of means.
iii) Find standard error of sampling distribution of sample means.
Solution: N=5 n=2
Total number of possible samples with replacement K= (N)n=52=25
Possible samples Sample means Possible samples Sample means
(2 , 2) 2 (6 , 8) 7
(2 , 4) 3 (6 , 10) 8
(2 , 6) 4 (8 , 2) 5
(2 , 8) 5 (8 , 4) 6
(2 , 10) 6 (8 , 6) 7
(4 , 2) 3 (8 , 8) 8
(4 , 4) 4 (8 , 10) 9
(4 , 6) 5 (10, 2 ) 6
(4 , 8) 6 (10 , 4) 7
(4 , 10) 7 (10, 6) 8
(6 , 2) 4 (10 , 8) 9
(6 , 4) 5 (10,10) 10
(6 , 6) 6

Sampling distribution of sample means:

X Frequency (F) Probability (P) PX PX 2
2 1 1/25 2/25 4/25
3 2 2/25 6/25 18/25
4 3 3/25 12/25 48/25
5 4 4/25 20/25 100/25
6 5 5/25 30/25 180/25
7 4 4/25 28/25 196/25
8 3 3/25 24/25 192/25
9 2 2/25 18/25 162/25
10 1 1/25 10/25 100/25
25 1 150/25 1000/25

From Sampling distribution of sample means:

60
Mean:  X = E ( X ) =  P X = =6
10
 1000
Variance  x =  P X − [ P X ] =
2 2
− [6] 2 = 40 − 36 = 4
2

25
Standard deviation:
x = 4 = 2
From Population:

Mean:  =
X =
30
=6
N 5

X X
2
2
 2

Variance:  2
= −  = 44 −  30  = 8
N  N  5  5 
 

X X
2
2
 2
44  30 
Standard deviation:  = −  = −   = 8 = 2.83
N  N  5  5 
 
Varification of properties
(i) X =  = 6
 2 
 X = 
2

(ii)  n 
8
4=
2
4=4

  
 x =  
(iii)  n
2=2
Properties of Sampling distribution of sample mean ( X ) (Without Replacement)
(i) Mean of the sampling distribution of sample means is equal to the population mean i.e
X = 

(ii) The variance of sampling distribution of sample mean is equal to population variance
divided by sample size.
  2  N − n 
X 2
=   
 n  N − 1 

 N −n
  is called Finite Population correction factor (F.P.C)
 N −1 
(iii) The standard deviation of sampling distribution of sample mean is equal to population
standard deviation divided by positive square root of sample size.
   N −n
 x =  
 n  N −1
Example: A population consists of 5 values, 2, 4, 6, 8 and 10. Draw all possible random
samples of size 2 without replacement and then
I. Construct the sampling distribution of sample means.
II. Verify the properties of sampling distribution of means
III. Find standard error of sampling distribution of sample means
Solution: N=5 n=2
 N  5
Total number of possible samples without replacement K=   =   = 10
 n   2
Possible samples Sample means
(2 , 4) 3
(2 , 6) 4
(2 , 8) 5
(2 , 10) 6
(4 , 6) 5
(4 , 8) 6
(4 , 10) 7
(6 , 8) 7
(6 , 10) 8
(8 , 10) 9

Sampling distribution of sample means:

X Frequency (f) Probability (P) PX PX 2
3 1 1/10 3/10 9/10
4 1 1/10 4/10 16/10
5 2 2/10 10/10 50/10
6 2 2/10 12/10 72/10
7 2 2/10 14/10 98/10
8 1 1/10 8/10 64/10
9 1 1/10 9/10 81/10
10 1 60/10 390/10
From Sampling distribution of Sample means:
60
Mean:  X = E ( X ) =  P X = =6
10
390
Variance  x =  P X − [ P X ] =
2 2
− [6] 2 = 39 − 36 = 3
2

10
Standard deviation  x = 3 = 1.73
From Population:

Mean:  =
X =
30
=6
N 5

X X
2
2
 2

Variance:  2
= −  = 44 −  30  = 8
N  N  5  5 
 

X X
2
2
 44  30 
2

Standard deviation:  = −  = −   = 8 = 2.83

N  N  5  5 
 
Varification of properties

(i) X =  = 6
  2  N − n 
 X 2 =   
(ii)  n  N − 1 
85−2
3=  
2  5 −1 
3=3

   N −n
 x =  
 n  N −1
2.83 5 − 2
(iii) 1.73 =
2 5 −1
1.73 = 1.73
Example: The Arco transport company has a fleet of 5 buses, which have monthly running
costs of Rs. 2000, 1950, 1965,1985 and 1980. What are the mean variance and standard error
of mean for sampling distribution of size 3.
Solution:-
From population

Mean:  =
X = 1976
N

X X
2
2

Variance:  2
= −  = 294
N  N 
 
Mean Variance and standard error of sampling distribution of mean of size 3 without
replacement
(i)  X =  = 1976

N − n  2  5 − 3 2  2  294 
(ii)  x2 =  =  =   = 49
N − 1  n  5 − 1  3  4 3 
(iii) x = 7
Example: A large number of samples of size 50 were selected at random from a normal
population with mean  and variance  2 . The mean and standard error of the sampling
distribution of sample mean were obtained 2500 and 4 respectively. Find mean and variance
of the population.
Solution:
 X = 2500 X =4  X 2 = 16
,
 =  X = 2500
2
X2 =
n
 = n. X 2 = 50(16) = 800
2

Example: If the size of the sample from a population is 55, the variance of the sampling
distribution of sample means is 27, what must be the standard error of sample mean if n =
165.
Solution:

n = 55,  X = 27
2

2
X2 =
n
 2
27 =
55
 = 27(55) = 1485
2

 = 38.536

Standard error of sample mean if n = 165

 38.536
X = =3 =
n 165
Example: If the size of the sample from a population is 36 and standard error of mean is 2,
what must be the size of the sample become if the standard error is to be reduced to 1.2.
Solution:
n = 36,  X = 2

X =
n

2=
36
 = 2(6) = 12

X =
n
12
1.2 =
n
12
n=
1.2
n = 10
n = 100

Example: A random sample of 36 is taken from a population with a mean of 2 and a standard
deviation of 3. Find mean and standard error of sampling distribution of sample mean ( X ).
Solution:
n = 36,  = 2and = 3
 X=  = 2
 3 3
X = = = =2
n 36 6
Example: A random sample of 100 is taken from a population with a mean of 30 and a standard
deviation of 5. Find mean and standard error of sampling distribution of sample mean ( X ).
Solution:
n = 100,  = 30and = 5
 X =  = 30
 5 5
X = = = = 0.5
n 100 10

Sampling distribution of Difference between two sample means ( X 1 − X 2 )

Sampling distribution of difference between two sample means is the probability distribution

of all possible differences between X 1 and X 2 obtained from all possible samples of size
‘n1 and n2’ which can be drawn from two population size ‘N1 and N2’ is called sampling
distribution of difference between two sample means.
Properties of Sampling distribution of Difference between two sample
means ( X 1 − X 2 ) (With Replacement)
(i) Mean of the sampling distribution of X 1 − X 2 is equal to difference of two population
means i.e  X 1 −X 2 = 1 −  2

(ii) The variance of sampling distribution of X 1 − X 2 is equal to sum of two population

variances divided by respective sample size.
 2  2 
 X2 1−X 2
=  1 + 2 
 n1 n2 

(iii) The standard deviation of sampling distribution of X 1 − X 2 is equal to sum of two

population standard deviation divided by positive square root of respective sample size.
 2 2 
X = 1 + 2 
1−X 2  n1 n2 
 
Properties of Sampling distribution of Difference between two sample
means ( X 1 − X 2 ) (Without Replacement)
(i) Mean of the sampling distribution of X 1 − X 2 is equal to difference of two population
means i.e  X 1 −X 2 = 1 −  2

(ii) The variance of sampling distribution of X 1 − X 2 is equal to sum of two population

variances divided by respective sample size.
 12  N1 − n1   22  N 2 − n2 
 X2 =  +  
1−X 2
n1  N1 − 1  n2  N 2 − 1 

(iii) The standard deviation of sampling distribution of X 1 − X 2 is equal to sum of two

population standard deviation divided by positive square root of respective sample size.
 2 N −n  2  N 2 − n2  
X =  1  1 1  + 2  
1−X 2  n1  N1 − 1  n2  N 2 − 1  

Example: Find Mean and Standard error of sampling distribution of X 1 − X 2 .
Given that
n1 = 30, 1 = 78and 12 = 150
n2 = 50,  2 = 75and 22 = 200
Solution: X 1−X 2
= 1 −  2 = 78 − 75 = 3

 2 2 
X = 1 + 2 
1−X 2  n1 n2 
 
150 200
X 1−X 2
= + = 5+4 = 9
30 50
X 1−X 2
=3

Example: The television picture tubes of manufacturer A have a mean lifetime of 6.5 years
and a standard deviation of 0.9 years, while those of manufacturer B have a mean lifetime of
0.6 years and a standard deviation of 0.8 years. A random sample of size 36 tubes is selected
from manufacturer A, other random sample of size 49 tubes is selected from manufacturer B.
Find mean and standard error of the sampling distribution of ( X 1 − X 2 )
Solution: Given that
Manufacturer A n1 = 36, 1 = 6.5and 1 = 0.9
Manufacturer B n2 = 49,  2 = 6and 2 = 0.8
X 1−X 2
= 1 −  2 = 6.5 − 6 = 0.5

 2 2 
X = 1 + 2 
1−X 2  n1 n2 
 
0.81 0.64
X 1−X 2
= + = 0.0225 + 0.0130 = 0.0356
36 49
X 1−X 2
= 0.1885

Example: Random samples of each size 100 are drawn from two independent probability
distribution and their means ( X 1 ) and ( X 2 ) computed. If the means and standard deviations
of the two populations are 1 = 10 and  1 = 2 ,  2 = 8 and  2 = 1 .Find mean and standard
error of the sampling distribution of ( X 1 − X 2 )
Solution: Given that
n1 = 100, 1 = 10and 1 = 2
n2 = 100,  2 = 8and 2 = 1
X 1−X 2
= 1 −  2 = 10 − 8 = 2

 2 2 
X = 1 + 2 
1−X 2  n1 n2 
 
4 1
X 1−X 2
= + = 0.04 + 0.01 = 0.05
100 100
X 1−X 2
= 0.224
Sampling distribution of sample proportion (P )
Sampling distribution of sample proportion ‘P’ is the probability distribution of the
proportion of success obtained from all possible samples of size ‘n’ which can be drawn from
population size ‘N’ with proportion ‘π’ is called sampling distribution of sample proportion.
Properties of Sampling distribution of sample proportion (P ) (With
Replacement)
(i) Mean of the sampling distribution of sample proportion is equal to the population
proportion i.e  P = 
(ii) The variance of sampling distribution of sample proportion is equal
  (1 −  ) 
 P2 =  
 n 
(iii) The standard deviation of sampling distribution of sample proportion is equal

 (1 −  )
P =
n
Example: A population consists of 4 values 1, 3, 6 and 8. Draw all possible samples of size 2
with replacement and find construct the sampling distribution of even numbers in the
samples. Also find mean and variance of the sampling distribution.
Solution: N=4 n=2 method=with replacement
Total number of possible samples with replacement K= (N)n=42=16
Possible samples Sample Proportion (p) Possible samples Sample Proportion (p)

(1,1) 0 (6,6) 1
(1,3) 0 (6,8) 1
(1,6) 1/2 (8,1) 1/2
(1,8) 1/2 (8,3) 1/2
(3,1) 0 (8,6) 1
(3,3) 0 (8,8) 1
(3,6) 1/2
(3,8) 1/2
(6,1) 1/2
(6,3) 1/2
From population;
X 2
Population proportion of even numbers π = = = 0.5
N 4
Sampling distribution of sample proportions:
p Frequency (f) Probability (P) P(p) P(p2)
0 4 4/16 0 0
1/2 8 8/16 8/32 8/64
1 4 4/16 8/32 16/64
16 1 16/32 24/64

From Sampling distribution:

16
 p =  P( p) = = 0.5
32
 p2 =  p 2 ( P) −  P( p) = − (0.5) = 0.375 − 0.25 = 0.125
2 24 2

64
P = p 2
( P) −  P( p) 2
= 0.125 = 0.353
Verification of properties:
(i)  P =  = 0.5
 (1 −  )
 pˆ 2 =
n
0.5(0.5)
(ii)  pˆ =
2

2
0.125 = 0.125
 (1 −  )
 pˆ =
n
0.5(0.5)
(iii)  pˆ =
2
0.353 = 0.353
Properties of Sampling distribution of sample proportion (P ) (Without
Replacement)
(i) Mean of the sampling distribution of sample proportion is equal to the population
proportion i.e  P = 
(ii) The variance of sampling distribution of sample proportion is equal
  (1 −  )  N − n 
 P2 =   
 n  N − 1 
(iii) The standard deviation of sampling distribution of sample proportion is equal

 (1 −  )  N − n 
P =  
n  N −1 
Example: A population consists of 4 values 1, 3, 6 and 8. Draw all possible samples of size 2
without replacement and find construct the sampling distribution of even numbers in the
samples. Also find mean and variance of the sampling distribution.
Solution: N=4 n=2 method=without replacement
 N   4
Total number of possible samples without replacement K=   =   = 6
 n   2
Possible samples Sample Proportion (p) Possible samples Sample Proportion (p)

(1,3) 0 (3,6) 1/2

(1,6) 1/2 (3,8) 1/2
(1,8) 1/2 (6,8) 1

Sampling distribution of sample proportions:

p Frequency (f) Probability (P) P(p) P(p2)
0 1 1/6 0 0
1/2 4 4/6 4/12 4/24
1 1 1/6 2/12 4/24
6 1 6/12 8/24

From population;
X 2
Population proportion of even numbers π = = = 0.5
N 4
From Sampling distribution:
6
 p =  P( p) = = 0.5
12
 p2 =  p 2 ( P) −  P( p) = − (0.5) = 0.333 − 0.25 = 0.0833
2 8 2

24
P = p 2
( P) −  P( p) 2
= 0.0833 = 0.2886
Verification of properties:
(i)  P =  = 0.5
  (1 −  )  N − n 
 P2 =   
 n  N − 1 
 0.5(0.5)  4 − 2 
(ii) 0.0833 =   
 2  4 − 1 
0.0833 = (0.125)(0.6666)
0.0833 = 0.0833
   (1 −  )  N − n 
P =   
 n  N − 1 
 0.5(0.5)  4 − 2 
(iii) 0.2886 =   
 2  4 − 1 
0.2886 = 0.2886
Example: The Zarco marketing research institute has a total of 8 members whose ages are
32, 33, 26, 43, 28, 52, 25 and 27 years. The institute has a rule that its president must be at
least 33 years old. Let’s assume that a sample of size 4 is selected to provide an estimate of
the population proportion (P) eligible for the president. What would be the mean and standard
error of the sampling distribution of sample proportion?
Solution:-
From population
π = Proportion of the members eligible for president= π =3/8 = 0.375
Mean, variance and standard error of sampling distribution of size 4 without
replacement
(i) Mean:  P =  = 0.375
  (1 −  )  N − n   0.375(0.625)  8 − 4 
 P2 =   =  
(ii) Variance:  n  N − 1   4  8 − 1 
 P 2 = 0.0585(0.5714) = 0.0334

  (1 −  )  N − n 
(iii) Standard Error:  P =    = 0.0334 = 0.1828
 n  N − 1 
Example: Suppose that 60% of a city population favours public finding for a proposed
recreational facility. If 150 persons are to be randomly selected and interviewed, what is the
mean and standard error of the sample proportion favouring this issue?
60
Solution:  = 60% = = 0.6 n = 150
100
Mean, variance and standard error of sampling distribution of Sample proportion (P)
(i) Mean:  P =  = 0.6
  (1 −  )   0.6(0.4) 
(ii) Variance:  P =  =  = 0.0016
2

 n   150 

  (1 −  ) 
(iii) Standard Error:  P =   = 0.0016 = 0.04
 n 
Example: A small professional society has N= 4500 members. The president has mailed n =
400 questionnaire to a random sample of members asking whether they wish to affiliate with
a large group. Assuming that the proportion of entire membership favouring consolidation is
π = 0.7. Find mean and standard error of the sample proportion P.
Solution:  = 0.7 n = 400 N = 4500
Mean, variance and standard error of sampling distribution of Sample proportion (P)
(i) Mean:  P =  = 0.7
  (1 −  )  N − n   0.7(0.3)  4500 − 400 
 P2 =   =  
(ii) Variance:  n  N − 1   400  4500 − 1 
 P 2 = 0.000525(0.9113) = 0.000478

  (1 −  )  N − n 
(iii) Standard Error:  P =    = 0.000478 = 0.022
 n  N − 1 
Sampling distribution of Difference between two sample Proportion ( P1 − P2 )
Sampling distribution of difference between two sample (P1-P2) is the probability distribution
of all possible differences between P1 and P2 obtained from all possible samples of size ‘n1
and n2’ which can be drawn from two population size ‘N1 and N2’ with proportion π1 and π2
is called sampling distribution of difference between two sample proportion (P1-P2).
Properties of Sampling distribution of Difference between two sample
Means ( P1 − P2 ) (With Replacement)
(i) Mean of the sampling distribution of P1 − P2 is equal to difference of two population

proportion i.e  P1 −P2 =  1 −  2

(ii) The variance of sampling distribution of P1 − P2 is equal

  1 (1 −  1 )  2 (1 −  2 ) 
 P2 − P =  + 
 n1 n2 
1 2

(iii) The standard deviation of sampling distribution of P1 − P2 is equal

  1 (1 −  1 )  2 (1 −  2 ) 
 P − P =  + 
 n1 n2 
1 2

Properties of Sampling distribution of Difference between two sample

Means ( P1 − P2 ) (Without Replacement)
(i) Mean of the sampling distribution of P1 − P2 is equal to difference of two population

proportion i.e  P1 −P2 =  1 −  2

(ii) The variance of sampling distribution of P1 − P2 is equal

 1 (1 −  1 )  N1 − n1   2 (1 −  2 )  N 2 − n2 
 P2 − P =   +  
n1  N1 − 1  n2  N2 −1 
1 2

(iii) The standard deviation of sampling distribution of P1 − P2 is equal

 1 (1 −  1 )  N1 − n1   2 (1 −  2 )  N 2 − n2 
 P −P =   +  
 N1 − 1 
n1 n2  N2 −1 
1 2

Example: The percentage of families with a monthly income of Rs. 1000 or more in city A
and city B is 25% and 20% respectively. If a random sample of 100 families is selected from
each of these two cities and the proportion of families earning Rs. 1000 or more in two
samples are compared, what is the mean and standard error of (P1 - P2).
Solution:  1 = 25% = 0.25 n1 = 100
 2 = 20% = 0.20 n1 = 100
Mean, variance and standard error of sampling distribution of (P1 – P2)
(i) Mean:  P1 − P2 =  1 −  2 = 0.25 − 0.20 = 0.05
  (1 −  1 )  2 (1 −  2 )   0.25(0.75) 0.20(0.80) 
(ii) Variance:  P21 − P2 =  1 +  =  +  = 0.0035
 n1 n2   100 100 

  (1 −  1 )  2 (1 −  2 ) 
(iii) Standard Error:  P1 − P2 =  1 +  = 0.0035 = 0.059
 n1 n2 

Sampling distribution of sample variance (S 2 )

Sampling distribution of sample variance (S2) is the probability distribution of all possible
variances obtained from all possible samples of size ‘n’ which can be drawn from two
population size ‘N’ with variance σ2 is called sampling distribution of sample variance (S2).
Property of Sampling distribution of sample Variance (S 2 ) (With Replacement)
n −1 2
Mean of the sampling distribution of sample variance is equal. i.e  S2 = 
n
Property of Sampling distribution of sample Variance (S 2 ) (Without
Replacement)
 N  n −1 2
Mean of the sampling distribution of sample variance is equal. i.e  S2 =   
 N −1 n
 Example: A population consists of 5 values, 2, 4, 6, 8 and 10. Draw all possible random
samples of size 2 with replacement and then
i) Construct the sampling distribution of sample variance.
n −1 2
ii) Verify the property of mean of sampling distribution of sample variance. i.e  S2 = 
n
Solution: N=5, n=2, Total number of possible samples with replacement K= (N)n=52=25
Possible Sample means Sample variance Possible Sample means Sample means
Samples X
S 2
=
(X − X ) 2 samples X
S 2
=
(X − X ) 2

n n
(2 , 2) 2 0 (6 , 8) 7 1
(2 , 4) 3 1 (6 , 10) 8 4
(2 , 6) 4 4 (8 , 2) 5 9
(2 , 8) 5 9 (8 , 4) 6 4
(2 , 10) 6 16 (8 , 6) 7 1
(4 , 2) 3 1 (8 , 8) 8 0
(4 , 4) 4 0 (8 , 10) 9 1
(4 , 6) 5 1 (10, 2 ) 6 16
(4 , 8) 6 4 (10 , 4) 7 9
(4 , 10) 7 9 (10, 6) 8 4
(6 , 2) 4 4 (10 , 8) 9 1
(6 , 4) 5 1 (10,10) 10 0
(6 , 6) 6 0
Sampling distribution of sample variances:
S2 Frequency (F) Probability (P) PS 2
0 5 5/25 0
1 8 8/25 8/25
4 6 6/25 24/25
9 4 4/25 36/25
16 2 2/25 32/25
25 1 100/25
100
From Sampling distribution of sample variances:  S2 =  PS 2 = =4
25
From Population: (X − X ) ( X − X )2
X
2 -4 16 =
X =
30
=6
N 5
4 -2 4
6 0 0 S 2
=
(X − X ) 2

=
40
=8
N 5
8 2 4
10 4 16
30 0 40

Varify the property

n −1 2
 S2 = 
n
2 −1
4= (8)
2
4=4
Example: A population consists of 5 values, 2, 4, 6, 8 and 10. Draw all possible random
samples of size 2 without replacement and then
I. Construct the sampling distribution of sample variance.
II. Verify the property of sampling distribution of sample variance
Solution: N=5 n=2
 N  5
Total number of possible samples without replacement K=   =   = 10
 n   2
Possible Sample means Sample variance
samples X
S2 =
(X − X ) 2

n
(2 , 4) 3 1
(2 , 6) 4 4
(2 , 8) 5 9
(2 , 10) 6 16
(4 , 6) 5 1
(4 , 8) 6 4
(4 , 10) 7 9
(6 , 8) 7 1
(6 , 10) 8 4
(8 , 10) 9 1
From Population:

X (X − X ) ( X − X )2
=
X =
30
=6
N 5
2 -4 16
4 -2 4 S 2
=
(X − X ) 2

=
40
=8
6 0 0 N 5

8 2 4
10 4 16
30 0 40

Sampling distribution of sample variances:

S2 Frequency (F) Probability (P) PS 2
1 4 4/10 4/10
4 3 3/10 12/10
9 2 2/10 18/10
16 1 1/10 16/10
10 1 50/10

From Sampling distribution of sample variances:

50
 S2 =  PS 2 = =5
10
Varify the property
 N  n −1 2
 S2 =   
 N −1 n
 5  2 −1
5=  (8)
 5 −1 2
5=5