Sampling Distribution of the Sample Mean in Simple Random Sampling (SRS)

The sample mean y has a sampling distribution with the mean and SE given by
Mean of y = µ, for both SRSWR and SRSWOR
σ
SE of y = for SRSWR,
n
σ N −n
= for SRSWOR.
n N −1

An Example of SRS:
Consider a finite population of N = 5 values (Y1 = 1, Y2 = 3, Y3 = 4, Y4 = 5, Y5 = 7). The mean
and SD of the population are
1+ 3 + 4 + 5 + 7
µ= = 4.0
5
(1 − 4) 2 + (3 − 4) 2 + (4 − 4) 2 + (5 − 4) 2 + (7 − 4) 2
σ= = 2.0
5
Suppose we want to draw, from the above population, a random sample of size n = 2, say, (y1,
y2) and our objective is to find the sampling distribution of the sample mean (based on the
sample of size 2), namely,
y + y2
y= 1
2
a) SRSWR: Total number of samples of size 2 = 52 = 25;
Probability of each sample = 1/25.
The following table gives the different samples of size2, corresponding probabilities of
drawing them, and the values of the statistic whose sampling distribution we want, namely,
y + y2
the sample mean y = 1 .
2
Sample Sample Probabi- Sample Sample Sample Probabi- Sample
Number Values lity Mean Number Values lity Mean
(y1, y2) y (y 1 , y 2 ) y
1 (1, 1) 1/25 1.0 14 (4, 5) 1/25 4.5
2 (1, 3) 1/25 2.0 15 (4, 7) 1/25 5.5
3 (1, 4) 1/25 2.5 16 (5, 1) 1/25 3.0
4 (1, 5) 1/25 3.0 17 (5, 3) 1/25 4.0
5 (1, 7) 1/25 4.0 18 (5, 4) 1/25 4.5
6 (3, 1) 1/25 2.0 19 (5, 5) 1/25 5.0
7 (3, 3) 1/25 3.0 20 (5, 7) 1/25 6.0
8 (3, 4) 1/25 3.5 21 (7, 1) 1/25 4.0
9 (3, 5) 1/25 4.0 22 (7, 3) 1/25 5.0
10 (3, 7) 1/25 5.0 23 (7, 4) 1/25 5.5
11 (4, 1) 1/25 2.5 24 (7, 5) 1/25 6.0
12 (4, 3) 1/25 3.5 25 (7, 7) 1/25 7.0
13 (4, 4) 1/25 4.0 Total -- 1 --

Saibal Chattopadhyay
Example - SRS
 Note that the same value of the sample mean can occur for different samples. A table giving
the distinct values of the sample mean together with the corresponding probabilities will
provide the Sampling Distribution of the Sample Mean in this case.

Table: Sampling Distribution of the Sample Mean in SRSWR (N = 5, n = 2)

Sample 1.0 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 7.0 Total
mean( y )
Probabi- 1/25 2/25 2/25 3/25 2/25 5/25 2/25 3/25 2/25 2/25 1/25 1
lity

The mean and SD of the sampling distribution (more precisely, called the Expectation and the
Standard Error) of the sample mean can now be obtained from the above sampling distribution
just as one would obtain them from any probability distribution.

Mean (Expectation) = Mean of the Sample Mean = Σ y . Probability

1 + 4 + 5 + 9 + 7 + 20 + 9 + 15 + 11 + 12 + 7
= = 4.0
25
SD(Standard Error) =SD of the Sampling Distribution = ∑ ( y − mean) 2 . probability
9 + 8 + 4.5 + 3 + 0.5 + 0 + 0.5 + 3 + 4.5 + 8 + 9
= = 2.0 = 1.414
25
Check Using the Formulae for Expectation and SE:

As Calculated earlier, the mean and SD of the population are µ = 4.0 and σ = 2.0. Now, using
the formulae,

Expectation of sample mean = Population Mean = 4.0,

σ 2
Standard Error of Sample Mean (for SRSWR) = = = 2.0 =1.414
n 2

Saibal Chattopadhyay
Example - SRS
b) SRSWOR: Total number of samples of size 2 = 5C2 = 10;
Probability of each sample = 1/10.

The following table gives the different samples of size2, corresponding probabilities of drawing
them, and the values of the statistic whose sampling distribution we want, namely, the sample
y + y2
mean y = 1 .
2

Sample Sample Probabi- Sample

Number Values lity Mean
(y1, y2) y
1 (1, 3) 1/10 2.0
2 (1, 4) 1/10 2.5
3 (1, 5) 1/10 3.0
4 (1, 7) 1/10 4.0
5 (3, 4) 1/10 3.5
6 (3, 5) 1/10 4.0
7 (3, 7) 1/10 5.0
8 (4, 5) 1/10 4.5
9 (4, 7) 1/10 5.5
10 (5, 7) 1/10 6.0
Total -- 1 --

Note that the same value of the sample mean can occur for different samples. A table giving
the distinct values of the sample mean together with the corresponding probabilities will
provide the Sampling Distribution of the Sample Mean in this case.

Table: Sampling Distribution of the Sample Mean in SRSWOR (N = 5, n = 2)

Sample 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Total
mean( y )
Probabi- 1/10 1/10 1/10 1/10 2/10 1/10 1/10 1/10 1/10 1
lity

The mean and SD of the sampling distribution (more precisely, called the Expectation and the
Standard Error) of the sample mean can now be obtained as in the case of SRSWR.

Mean (Expectation) = Mean of the Sample Mean = Σ y . Probability

2 + 2.5 + 3 + 3.5 + 8 + 4.5 + 5 + 5.5 + 6
= = 4.0
10

Saibal Chattopadhyay
Example - SRS
SD (Standard Error) = SD of the sampling distribution = ∑ ( y − mean) 2 . probability =
4 + 2.25 + 1 + 0.25 + 0 + 0.25 + 1 + 2.25 + 4
= 1.5 = 1.225
10

Check Using the Formulae for Expectation and SE:

As Calculated earlier, the mean and SD of the population are µ = 4.0 and σ = 2.0. Now, using
the formulae,

Expectation of sample mean = Population Mean = 4.0,

σ N −n
Standard Error of Sample Mean (for SRSWR) =
n N −1
2 5−2
= = 1.5 =1.225.
2 5 −1

Thus it is not necessary to calculate the expectation and SE from the sampling distribution of the
sample mean, the formulas will suffice, knowing the values of N (the population size), n (the
sample size), and σ (the population SD).

Also note that the Standard Error of the sample mean y for SRSWOR is less than that of
SRWWR, suggesting that SRSWOR is to be preferred over SRSWR in this case.

Saibal Chattopadhyay
Example - SRS