Lecture #2 Sampling Distribution

Objectives of this lecture:

After reading this unit, you should be able to
• understand about the sampling distribution for mean
• understand about the sampling distribution for sample proportions

Sampling Distribution for Mean

Example 2.1: The ages of six executives of a company are
Name Mr. Ravi Mrs. Veena Mrs. Shanti Mr. Suresh Mr. Rajiv Mr. Anil
Age 54 50 52 48 50 52
i. How many samples of size are possible?
ii. Construct the sampling distribution of means by taking samples of size 2 and
organise the data.
iii. Calculate the mean of the sampling distribution and compare it with the population
mean.
The sample mean also have a variance. It measures the dispersion of the individual observations
(sample means) around their population means. One is the standard deviation of the entire
population which we usually denoted by σ . Then the standard deviation of a single sample, and
now we have the standard deviation σ X̄ of an entire set of sample means. Since σ X̄ measures
the dispersion of the sample means around μ , it gives a measure for the error in sampling.
Because of this, σ X̄ is called sampling error.
The standard error is the standard deviation of the sample means around the population mean and
is denoted by SE ( x̄) .
There is an alternative method to compute standard error of the means, SE ( x̄) , from a single
sample if we know the population standard deviation. By this method, we have the following
formula for obtaining the standard error of the mean for a finite population.
( N −n) σ 2
SE ( x̄)=
√ N −1 n

Sampling Distribution | Basic| Lecture #2

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

where N = population size = the total number of individuals, n = sample size = number of
individuals selected in the random sample. σ 2 = standard deviation of the individuals in the
population.
We use the above formula for computing SE ( x̄) , when N-n is not very large. When N is large,
relative to n, we use the formula,
2

√
SE ( x̄)= σ = σ
n √n
This formula can also be used for calculating SE ( x̄) for infinite population.

Example 2.2: The U.S. Bureau of census wishes to estimate the birth rates per 1,00,000 people in
the nation's largest cities. It is known that the standard deviation in the birth rates for these 100
urban centres is 12 births per 1,00,000 people. Then
(a) Calculate the variance and standard error of the sampling distribution of i) n = 8 cities ii) n =
15 cities.
(b) Compare the values obtained in both the cases.
Solution:
(a) (i) Here N = 100 and n = 8 and the population variance is 12. Therefore n / N less than 0.1.
Then we use the formula for calculating the variance as
2 2
12
σ 2x̄ = σ = =18
n 8
and the standard error is σ x̄ = √ 18=4.24
(ii) In this case N=100 and n=15 and therefore n slash N is greater than 0.1. Also, σ =12.
Therefore we use the formula for variance as
N −n σ 2
σ 2x̄ = ( )
N −1 n
=8.24

Hence, σ x̄ = √ 8.24=2.87
(b) On comparing both the values we observe that, the larger sample has a smaller standard error
and will tend to result in less sampling error in estimating the birth rates in the 100 cities.

Central Limit Theorem

The central limit theorem, first introduced by De Moivre during the early eighteenth century,
happens to be the most important theorem in statistics. According to this theorem, if we select a
large number of simple random samples, for example, from any population distribution and

Sampling Distribution | Basic| Lecture #2

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

determine the mean of each sample, the distribution of these sample means will tend to be described
by the normal probability distribution with a mean μ and variance σ 2 /n .
More formally, the CLT tells us that,
2
if 2
x ~ N ( μ , σ ) then x̄ ~ N ( μ , σ ) where n is sufficiently large
n

Or in other words, we can say that, the sampling distribution of sample means approaches to a
normal distribution. The sampling distribution of the mean of a random sample drawn from any
population is approximately normal for a sufficiently large sample size .
The central limit theorem tells us that no matter the population distribution, the sampling
distribution’s shape will approach normality as the sample size (N) increases.

This is useful, as the research never knows which mean in the sampling distribution is the same as
the population mean, but by selecting many random samples from a population, the sample means
will cluster together, allowing the research to make a very good estimate of the population mean.

Thus, the sampling error will decrease as the sample size (n) increases.

Sampling Distribution of the Sample Proportion

The population proportion ( π ) is a parameter that is as commonly estimated as the mean. It is
just as important to understand the distribution of the sample proportion, as the mean. With
proportions, the element either has the characteristic you are interested in or the element does not
have the characteristic. The sample proportion is p calculated by
x
p=
n
where x is the number of elements in your population with the characteristic and n is the sample
size.

Sampling Distribution | Basic| Lecture #2

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

The distribution of the sample proportion has a mean of μ p= π

and has a standard deviation of σ p=

√ π (1−π )
n
The sample proportion is normally distributed if n is very large and p isn’t close to 0 or 1.

Example 2.3: A statistician is analyzing the participation rate in a school’s math club. The school
has 6 students, and each student either participates in the math club (coded as 1) or does not
participate (coded as 0). The data is summarized as follows:
Student Participates in Math Club (1) / Does Not Participate (0)
A 1
B 1
C 0
D 1
E 0
F 1

The population proportion p of students participating in the math club is:

4
p= =0.67
6
The number of possible samples of size 2 from 6 students is given by the combination formula:
6 =15
()
2
List all possible samples of size 2 and calculate the sample proportion p^ for each.

Sample Student 1 Student 2 Sample Proportion p^

1 A (1) B (1) 1.00
2 A (1) C (0) 0.50
3 A (1) D (1) 1.00
4 A (1) E (0) 0.50
5 A (1) F (1) 1.00
6 B (1) C (0) 0.50
7 B (1) D (1) 1.00
8 B (1) E (0) 0.50
9 B (1) F (1) 1.00
10 C (0) D (1) 0.50
11 C (0) E (0) 0.00
12 C (0) F (1) 0.50
13 D (1) E (0) 0.50

Sampling Distribution | Basic| Lecture #2

 Lectured by
Subject: STT251: Sampling Technique
Md. Kaderi Kibria

Sample Student 1 Student 2 Sample Proportion p^

14 D (1) F (1) 1.00
15 E (0) F (1) 0.50

The sampling distribution is summarized below:

Sample Proportion p^ Frequency Probability

0.00 1 1 / 15 = 0.067
0.50 9 9 /15 = 0.600
1.00 5 5 /15 = 0.333

The mean of the sampling distribution μ ^p is:

μ ^p =0.633

The population proportion p of students participating in the math club is 0.67, which is close to the
mean of the sampling distribution μ ^p =0.633 . This demonstrates that the sampling distribution of
the sample proportion is an unbiased estimator of the population proportion.

References:
1. Hogg, R. V. and Craig, A. T. (2012), Introduction of Mathematical Statistics, 7 th Edition,
Pearson Education, New Delhi.
2. Hoel, P. G. (1984), Introduction to Mathematical Statistics, 5th Edition, John Wiley and
Sons, New York.
3. Mood, A. M., Graybill, F. A., Boes, D. C. (1974), Introduction to the Theory of Statistics,
3rd Edition, McGraw-Hill, USA.
4. https://people.ohio.edu/ruhil/statsbook/sampling.html#the-sampling-distribution-of-binary-
proportions
5. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://egyankosh.ac.in/bitstream/
123456789/14027/1/Unit-4.pdf

Sampling Distribution | Basic| Lecture #2