Small sample tests: t-test for single population mean

Null hypothesis H0 :There is no sig difference between the sample

mean and population mean.
√
(x − µ) × n − 1
Test statistic t =
s
with degree of freedom v = n − 1
where x is sample mean, µ is population mean, s is Standard
Deviation of sample and n is the size of sample.
s
P
2 P 2
X −X 2 x
s = (or ) s = − x2
n n

Conclusion:
Case (i): If calculated value of |t| < tα , H0 is accepted at a
particular level of significance
Case (ii): If calculated value of |t| > tα , H0 is rejected at a
particular level of significance.
Dr. P. Gurusamy
Small sample tests: t-test for single population mean

Note:
(i) 95% confident limit for population mean µ are given by
s
x ± √ × t0.05 .
n−1

(ii) 99% confident limit for population mean µ are given by

s
x ± √ × t0.01 .
n−1

Dr. P. Gurusamy
Problem No. 1

A random sample of 16 values from normal population showed a

mean of 41.5 inches and the sum of squares of deviation from this
mean equal to 135 sq inches. Show that the assumption of a mean
of 43.5 inches for population is not reasonable. Obtain 95% and
99% confidence limits for the same.
Solution:
Given : Size of sample n = 16 < 30 (small sample)
sample mean x = 41.5,
population mean µ = 43.5,
P
2
X − X = 135
Step (i): Null hypothesis H0 : µ = 43.5
Step (ii): Alternative hypothesis H1 : µ 6= 43.5 (two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 1

Step (iv): Calculations

s
P
2 r
X −X 135
s = = = 2.90
n 16
Test statistic
√ √
(x − µ) × n−1 (41.5 − 43.5) × 16 − 1 −7.7459
t = = = = −2.67
s 2.90 2.90
The calculated value of |t| = 2.671.
The table value of t at 5% level of significance for two tailed test
with d.f v = n - 1 = 16 - 1 = 15, t0.05 = 2.131.
Step (v): Conclusion:
Since |t| > t0.05 , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Problem No. 1

(i) Obtain 95% confidence limits are

 
s s s
x±√ × t0.05 ⇒ x − √ × t0.05 , x + √ × t0.05
n−1 n−1 n−1
 
2.90 2.90
⇒ 41.5 − √ × t0.05 , 41.5 + √ × t0.05
15 15
 
2.90 2.90
⇒ 41.5 − √ × 2.131, 41.5 + √ × 2.131
15 15
⇒ (41.5 − 1.595, 41.5 + 1.595)
⇒ (39.905, 43.095) .

Dr. P. Gurusamy
Problem No. 1

(ii) Obtain 99% confidence limits are

 
s s s
x±√ × t0.01 ⇒ x − √ × t0.01 , x + √ × t0.01
n−1 n−1 n−1
 
2.90 2.90
⇒ 41.5 − √ × t0.01 , 41.5 + √ × t0.01
15 15
 
2.90 2.90
⇒ 41.5 − √ × 2.947, 41.5 + √ × 2.947
15 15
⇒ (41.5 − 2.2066, 41.5 + 2.2066)
⇒ (39.2934, 43.7066) .

Dr. P. Gurusamy
Problem No. 2

The 9 items of a sample have the following values 50, 52, 48, 47,
49, 53, 51, 45 and 46. Does the mean of these values differ
significantly from assumed mean 47.5?.

Solution:
Given : Size of sample n = 9 < 30 (small sample)
population mean µ = 47.5,
sample mean P
x 441
X = = = 49
n 9
Step (i): Null hypothesis H0 : µ = 47.5
Step (ii): Alternative hypothesis H1 : µ 6= 47.5 (two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 2

Step (iv): Calculations

2
Sample values(x) X −X X −X
50 1 1
52 3 9
48 -1 1
47 -2 4
49 0 0
53 4 16
51 2 4
45 -4 16
46 -3 9
Total 60

Dr. P. Gurusamy
Problem No. 2

s
P
2 r
X −X 60
s = = = 2.582
n 9
Test statistic
√ √
(x − µ) × n−1 (49 − 47.5) × 9−1 4.242
t = = = = 1.643.
s 2.582 2.582
The calculated value of |t| = 1.643.
The table value of t at 5% level of significance for two tailed test
with d.f v = n - 1 = 9 - 1 = 8, t0.05 = 2.306.
Step (v): Conclusion:
Since |t| < t0.05 , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 3
An operator in a ice mill is told to fill bags with an average of 100
kg of rice per bag. A random sample of his work show the
following :
Bag number 1 2 3 4 5 6
Weight (kgs) 100.5 99.8 100.2 100.1 99.3 100.4
Test the hypothesis that he is doing the job correctly against the
alternative hypothesis that he is not at 5% level.
Solution:
Given : Size of sample n = 6 < 30 (small sample)
population mean µ = 100
sample mean P
x 600.3
X = = = 100.05
n 6
Step (i): Null hypothesis H0 : µ = 100
Step (ii): Alternative hypothesis H1 : µ 6= 100 (two tailed test)
Step (iii): Choose a level of significance α = 5%.
Dr. P. Gurusamy
Problem No. 3

Step (iv): Calculations

2
Sample values(x) X −X X −X
100.5 0.45 0.2025
99.8 -0.25 0.0625
100.2 0.15 0.0225
100.1 0.05 0.0025
99.3 -0.75 0.5625
100.4 0.35 0.1225
Total 0.975

Dr. P. Gurusamy
Problem No. 3

s
P
2 r
X −X 0.975
s = = = 0.403
n 6
Test statistic
√ √
(x − µ) × n−1 (100.05 − 100) × 5
t = = = 0.277
s 0.403
The calculated value of |t| = 0.277.
The table value of t at 5% level of significance for two tailed test
with d.f v = n - 1 = 6 - 1 = 5, t0.05 = 2.571.
Step (v): Conclusion:
Since |t| < t0.05 , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 4

Given a sample mean of 83, a sample S.D of 12.5 and a sample

size of 22, test the hypothesis that the value of the population
mean is 70 against alternative that it is more than 70. Use the
0.05 level of significance?

Solution:
Given : Size of sample n = 22 < 30 (small sample)
population mean µ = 70
sample mean X = 83 and S.D of sample s = 12.5 Step (i): Null
hypothesis H0 : µ = 70
Step (ii): Alternative hypothesis H1 : µ > 70 (one tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 4

Test statistic
√ √
(x − µ) × n−1 (83 − 70) × 21
t = = = 4.7658
s 12.5
The calculated value of |t| = 4.7658.
The table value of t at 5% level of significance for one tailed test
with d.f v = n - 1 = 22 - 1 = 21, t0.05 = 1.721. since level
multiply by 2
Step (v): Conclusion:
Since |t| > t0.05 , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Problem No. 5
IQ test result of randomly selected five employees in an
organization is given below. Test whether minimum requirement of
average IQ level 87 is maintained in that company or not?
Employee code 234 232 121 343 111
IQ test 85 95 90 93 87

Solution:
Given : Size of sample n = 5 < 30 (small sample)
population mean µ = 87
sample mean P
x 450
X = = = 90
n 5
Step (i): Null hypothesis H0 : µ = 87
Step (ii): Alternative hypothesis H1 : µ > 87 (one tailed test)
Step (iii): Choose a level of significance α = 5%.
Dr. P. Gurusamy
Problem No. 5

Step (iv): Calculations

2
Sample values(x) X −X X −X
85 -5 25
95 5 25
90 0 0
93 3 9
87 -3 9
Total 68

Dr. P. Gurusamy
Problem No. 5

s
P
2 r
X −X 68
s = = = 3.6878
n 5
Test statistic
√ √
(x − µ) × n−1 (90 − 87) × 4
t = = = 1.626.
s 3.6878
The calculated value of |t| = 1.626.
The table value of t at 5% level of significance for one tailed test
with d.f v = n - 1 = 5 - 1 = 4, t0.05 = 2.132 since level multiply
by 2.
Step (v): Conclusion:
Since |t| < t0.05 , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 6

The weights of 10 people of a locality are formed to be 70, 67, 62,

68, 61, 68, 70, 64, 64, 66 kilograms. Is it reasonable to believe
that the average weight of the people of locality is greater than
64kg? Test at 5% level.

Solution:
Given : Size of sample n = 10 < 30 (small sample)
population mean µ = 64
sample mean P
x 660
X = = = 66
n 10
Step (i): Null hypothesis H0 : µ = 64
Step (ii): Alternative hypothesis H1 : µ > 64 (one tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 6

Step (iv): Calculations

2
Sample values(x) X −X X −X
70 4 16
67 1 1
62 -4 16
68 2 4
61 -5 25
68 2 4
70 4 16
64 -2 4
64 -2 4
66 0 0
Total 90

Dr. P. Gurusamy
Problem No. 6

s
P
2 r
X −X 90
s = = =3
n 10
Test statistic
√ √
(x − µ) × n−1 (66 − 64) × 9
t = = = 2.
s 3
The calculated value of |t| = 2.
The table value of t at 5% level of significance for one tailed test
with d.f v = n - 1 = 10 - 1 = 9, t0.05 = 1.833 since level multiply
by 2.
Step (v): Conclusion:
Since |t| > t0.05 , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Problem No. 7

The following is the information obtained from a random sample of

5 observations. Assume the population has a normal distribution
20, 18, 17, 22, 18. It is required to determine whether or not the
mean of the population from which this sample was taken is
significantly less than 21.
i) State null and alternate hypothesis
ii) Compute the standard error of the mean
iii) Determine the test statistic
iv) Determine p-value and at 95% confidence test whether or not
the mean of the population is significantly less than 21.

Solution:
Given : Size of sample n = 5 < 30 (small sample)
population mean µ = 21

Dr. P. Gurusamy
Problem No. 7

sample mean P
x 95
X = = = 19
n 5
Step (i): Null hypothesis H0 : µ = 21
Step (ii): Alternative hypothesis H1 : µ < 21 (one tailed test)
Step (iii): Choose a level of significance α = 5%. Step (iv):
Calculations

2
Sample values(x) X − X X −X
20 1 1
18 -1 1
17 -2 4
22 3 9
18 -1 1
Total 16

Dr. P. Gurusamy
Problem No. 7

s
P
2 r
X −X 16
s = = = 1.7889
n 5
s 1.7889
S.E = √ = = 0.8944
n−1 2
Test statistic
√ √
(x − µ) × n−1 (19 − 21) × 4 −4
t = = = = −2.236.
s 1.7889 1.7889
The calculated value of |t| = 2.236.
The table value of t at 5% level of significance for one tailed test
with d.f v = n - 1 = 5 - 1 = 4, t0.05 = 2.132 since level multiply
by 2.
Step (v): Conclusion:
Since |t| > t0.05 , H0 is rejected at 5% level of significance.
Dr. P. Gurusamy
Small sample tests: t-test for difference of two means

Case (i): Independent samples

Null hypothesis H0 :There is no sig difference between the

population means or H0 : µ1 = µ2
x −y
Test statistic t = q
S × n11 + 1
n2

with degrees of freedom v = n1 + n2 − 2.

where x is sample 1 mean, y is sample 2 mean, n1 is the size of
sample l, n2 is the size of sample 2, s1 is standard Deviation of
sample 1, s2 is standard Deviation of sample 2.
s
2 P
2
n1 s12 + n2 s22
P
X −X + Y −Y
S = (or ) S 2 =
n1 + n2 − 2 n1 + n2 − 2

Dr. P. Gurusamy
Small sample tests: t-test for difference of two means

Conclusion:
Case (i): If calculated value of |t| < tα , H0 is accepted at a
particular level of significance
Case (ii): If calculated value of |t| > tα , H0 is rejected at a
particular level of significance.

Dr. P. Gurusamy
Problem No. 1
The I.Qs of 16 students from one class of an institution showed a
mean of 107 with a S.D of 10, while the I.Qs of 14 students from
another class showed a mean of 112 with a S.D of 8. Check
whether there is an appreciable difference between the I.Qs of the
two groups at (i) 1% level and (ii) 5% level.

Solution:
Given : n1 is the size of sample l = 16 < 30,
n2 is the size of sample 2 = 14 < 30 (small sample test)
X is sample 1 mean = 107,
Y is sample 2 mean = 112
s1 is standard Deviation of sample 1 =10
s2 is standard Deviation of sample 2 =8
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 6= µ2 (two tailed test)
Step (iii): Choose a level of significance α = 5%.
Dr. P. Gurusamy
Problem No. 1

Step (iv): Calculations

s s
16 × 102 + (14 × 82 )


2 2
n1 s 1 + n2 s 2
S = =
n1 + n2 − 2 16 + 14 − 2

1600 + 896 √
r
= = 89.142 = 9.4415 ⇒ S = 9.4415
28

x −y 107 − 112
Test statistic t = q = q
S × n11 + 1
n2 9.4415 × 161
+ 1
14

−5 −5
= = = −1.447 ⇒ t = −1.447.
9.4415 × 0.366 3.4556
The calculated value of |t| = 1.447.

Dr. P. Gurusamy
Problem No. 1

(i)The table value of t at 5% level of significance for two tailed

test with degrees of freedom v = n1 + n2 − 2 = 16 + 14 − 2 = 28
is t0.05 = 2.048.
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.
(ii)The table value of t at 1% level of significance for two tailed
test with degrees of freedom v = n1 + n2 − 2 = 16 + 14 − 2 = 28
is t0.01 = 2.763.
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 1% level of significance.

Dr. P. Gurusamy
Problem No. 2
Two types of batteries are tested for their length of life and the
following data are obtained.
Type No of samples Mean life Variance
Type A 9 600 hours 121
Type B 8 640 hours 144
Is there significance difference in the two means at 5% level?
Solution:
Given : n1 is the size of sample l = 9 < 30,
n2 is the size of sample 2 = 8 < 30 (small sample test)
X is sample 1 mean = 600,
Y is sample 2 mean = 640
s1 is standard Deviation of sample 1 =11
s2 is standard Deviation of sample 2 =12
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 6= µ2 (two tailed test)
Step (iii): Choose a level of significance α = 5%.
Dr. P. Gurusamy
Problem No. 2

Step (iv): Calculations

s r
n1 s12 + n2 s22 (9 × 121) + (8 × 144)
S = =
n1 + n2 − 2 9+ 8− 2

1089 + 1152 √
r
= = 149.4 = 12.223 ⇒ S = 12.223
15
x −y 600 − 640
Test statistic t = q = q
S × n11 + 1
n2 12.223 × 19 + 1
8

−40 −40
= = = −6.7362 ⇒ t = −6.7362.
12.223 × 0.4858 5.938
The calculated value of |t| = 6.7362.

Dr. P. Gurusamy
Problem No. 2

The table value of t at 5% level of significance for two tailed test

with degrees of freedom v = n1 + n2 − 2 = 9 + 8 − 2 = 15 is
t0.05 = 2.131.
Step (v): Conclusion:
Since |t| > tα , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Problem No. 3
In the test given to two groups of students the marks obtained
were as follows:
Group 1 18 20 36 50 49 36 34 49 41
Group 2 29 28 26 35 30 44 46

Examine the significance difference between the means of marks

secured by students of the above two groups.

Solution:
Given : n1 is the size of sample l = 9 < 30,
n2 is the size of sample 2 = 7 < 30 (small sample test)
X is sample 1 mean = 333/9 = 37,
Y is sample 2 mean = 238/7 = 34
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 6= µ2 (two tailed test)
Step (iii): Choose a level of significance α = 5%.
Dr. P. Gurusamy
Problem No. 3

Step (iv): Calculations

2

2
G1(X) X −X X −X G2(Y) Y −Y Y −Y
18 -19 361 29 -5 25
20 -17 289 28 -6 36
36 -1 1 26 -8 64
50 13 169 35 1 1
49 12 144 30 -4 16
36 -1 1 44 10 100
34 -3 9 46 12 144
49 12 144
41 4 16
Total 1134 380

Dr. P. Gurusamy
Problem No. 3
s
2 P
2
P r
X −X + Y −Y 1134 + 380
S = =
n1 + n2 − 2 9+ 7 − 2
r
1520
S = = 10.42
14
x −y 37 − 34
Test statistic t = q = q
S × n11 + n12 10.42 × 19 + 17
3
= = 0.5713 ⇒ t = 0.5713.
10.42 × 0.504
The calculated value of |t| = 0.5713.
The table value of t at 5% level of significance for two tailed test
with degrees of freedom v = n1 + n2 − 2 = 9 + 7 − 2 = 14 is
t0.05 = 1.76.
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.
Dr. P. Gurusamy
Problem No. 4
Samples of two types of electric light bulbs were tested for length
of life and following data were obtained.
Type I Type II
Size of sample 8 7
Sample mean 1234 hrs 1036 hrs
Standard deviation of sample 36 hrs 40 hrs

Is the difference in the mean sufficient to warrant that type I is

superior to type II regarding length of life ?.

Solution:
Given : n1 is the size of sample l = 8 < 30,
n2 is the size of sample 2 = 7 < 30 (small sample test),
X is sample 1 mean = 1234, Y is sample 2 mean = 1036
s1 is standard Deviation of sample 1 =36
s2 is standard Deviation of sample 2 = 40
Dr. P. Gurusamy
Problem No. 4

Step (i): Null hypothesis H0 : µ1 = µ2

Step (ii): Alternative hypothesis H1 : µ1 > µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%. Step (iv):
Calculations
s r
n1 s12 + n2 s22 (8 × 1296) + (7 × 1600)
S = =
n1 + n2 − 2 8+ 7− 2
√
= 1659.08 = 40.7318 ⇒ S = 40.7318

Dr. P. Gurusamy
Problem No. 4

x −y 1234 − 1036
Test statistic t = q = q = 9.3925.
S × n11 + 1
n2 40.7318 × 18 + 1
7

The calculated value of |t| = 9.3925.

The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n1 + n2 − 2 = 8 + 7 − 2 = 13 is
t0.05 = 1.77. (since level multiply by 2)
Step (v): Conclusion:
Since |t| > tα , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Problem No. 5

The dean of students at UTC has said that the average grade of
UTC students is higher than that of the students at GSU. Random
samples of grades from the two schools are selected and the results
are shown below.
UTC GSU
Size of sample 14 12
Sample mean 2.85 2.61
Standard deviation of sample 0.40 0.35
Mode of sample 2.5 3.0

i) Give the hypothesis

ii) Compute test statistic
iii) At 1% level of significance, test the dean of students statements

Dr. P. Gurusamy
Problem No. 5

Solution:
Given : n1 is the size of sample l = 14 < 30,
n2 is the size of sample 2 = 12 < 30 (small sample test),
X is sample 1 mean = 2.85, Y is sample 2 mean = 2.61
s1 is standard Deviation of sample 1 = 0.40
s2 is standard Deviation of sample 2 = 0.35
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 > µ2 (One tailed test)
Step (iii): Choose a level of significance α = 5%. Step (iv):
Calculations
s s
14 × 0.42 + 12 × 0.362



2 2
n1 s1 + n2 s2
S = = = 0.3932
n1 + n2 − 2 14 + 12 − 2

Dr. P. Gurusamy
Problem No. 5

x −y 2.85 − 2.61
Test statistic t = q = q = 1.552.
S × n11 + 1
n2 0.3932 × 14 1
+ 1
12

The calculated value of |t| = 1.552.

The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n1 + n2 − 2 = 14 + 12 − 2 = 24 is
t0.05 = 1.711. (since level multiply by 2)
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 6

A group of five patients treated with Medicine A weights 42, 39,

48, 60 and 41kg. A second group of 7 patients from the same
hospital treated with Medicine B weights 38, 42, 56, 64, 68, 69
and 62kg. Do you agree with the claim that Medicine B increases
the weight significantly?

Solution:
Given : n1 is the size of sample l = 5 < 30,
n2 is the size of sample 2 = 7 < 30 (small sample test)
X is sample 1 mean = 230/5 = 46,
Y is sample 2 mean = 399/7 = 57
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 < µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 6

P P
X 230 Y 399
X = = = 46, Y = = = 57
n 5 n 7
Step (iv): Calculations


2

2
A(X) X −X X −X B(Y) Y −Y Y −Y
42 -4 16 38 -19 361
39 -7 49 42 -15 225
48 2 4 56 -1 1
60 14 196 64 7 49
41 -5 25 68 11 121
69 12 144
- 62 5 25
230 290 399 926

Dr. P. Gurusamy
Problem No. 6

s
2 P
2
P r
X −X + Y −Y 290 + 926
S = = = 11.027
n1 + n2 − 2 5+ 7 − 2

x −y 46 − 57
Test statistic t = q = q
S × n11 + 1
n2 11.027 × 15 + 1
7

−11 −11
= = = −1.699 ⇒ t = −1.699.
11.027 × 0.587 6.473
The calculated value of |t| = 1.699.
The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n1 + n2 − 2 = 5 + 7 − 2 = 10 is
t0.05 = 1.812. (since level multiply by 2)
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.
Dr. P. Gurusamy
Problem No. 7

The height of six randomly chosen sailors in inches 63, 65, 68, 69,
71 and 72. Those 10 randomly chosen soldiers are 61, 62, 65, 66,
69, 69, 70, 71, 72 and 73. Discuss the hypothesis that sailor are on
the average taller than soldiers.

Solution:
Given : n1 is the size of sample l = 6 < 30,
n2 is the size of sample 2 = 10 < 30 (small sample test)
X is sample 1 mean = 68,
Y is sample 2 mean = 67.8
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 > µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 7
P P
X 408 Y 678
X = = = 68, Y = = = 67.8
n 6 n 10
Step (iv): Calculations


2

2
A(X) X −X X −X B(Y) Y −Y Y −Y
63 -5 25 61 -6.8 46.24
65 -3 9 62 -5.8 33.64
68 0 0 65 -2.8 7.84
69 1 1 66 -1.8 3.24
71 3 9 69 1.2 1.44
72 4 16 69 1.2 1.44
70 2.2 4.84
71 3.2 10.24
72 4.2 17.64
73 5.2 27.64
408 60 678 153.6
Dr. P. Gurusamy
Problem No. 7

Step (iv): Calculations

s
2 P
2
P r
X −X + Y −Y 60 + 153.6
S = = = 3.906
n1 + n2 − 2 6 + 10 − 2

x −y 68 − 67.8
Test statistic t = q = q = 0.099.
S × n11 + 1
n2 3.906 × 16 + 1
10

The calculated value of |t| = 0.099.

The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n1 + n2 − 2 = 6 + 10 − 2 = 14 is
t0.05 = 1.76. (since level multiply by 2)
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 8

Two independent samples of 8 and 7 items respectively has the

following values:
Samle 1 9 11 13 11 15 9 12 14
Sample 2 10 12 10 14 9 8 10

Is the difference between the means of the samples significance?

Solution:
Given : n1 is the size of sample l = 8 < 30,
n2 is the size of sample 2 = 7 < 30 (small sample test)
X is sample 1 mean = 11.75,
Y is sample 2 mean = 10.43
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 6= µ2 (two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 8

P P
X 94 Y 73
X = = = 11.75, Y = = = 10.43
n 8 n 7
Step (iv): Calculations


2

2
A(X) X −X X −X B(Y) Y −Y Y −Y
9 -2.75 7.562 10 -0.4 0.176
11 -0.75 0.562 12 1.6 5.856
13 1.25 1.562 10 -0.4 0.176
11 -0.75 0.562 14 3.6 12.816
15 3.25 10.562 9 -1.4 2.016
9 -2.75 7.562 8 -2.4 5.858
12 0.25 0.0625 10 -0.4 0.176
14 2.25 5.062
94 33.5 73 23.72

Dr. P. Gurusamy
Problem No. 8

Step (iv): Calculations

s
2 P
2
P r
X −X + Y −Y 33.5 + 23.72
S = = = 2.098
n1 + n2 − 2 8+ 7 − 2

x −y 11.75 − 10.43
Test statistic t = q = q = 1.2151.
S × n11 + 1
n2
1
2.099 × 8 + 7 1

The calculated value of |t| = 1.2151.

The table value of t at 5% level of significance for two tailed test
with degrees of freedom v = n1 + n2 − 2 = 8 + 7 − 2 = 13 is
t0.05 = 2.09. (since level multiply by 2)
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Small sample tests: t-test for difference of two means

Case (ii): Dependent samples (or) Paired samples

Null hypothesis H0 :There is no sig difference between the

population means or H0 : µ1 = µ2

d √
Test statistic t = × n
s
with degrees of freedom v = n - 1
where P P 2
d 2 d 2
d = , d = x − y, s = −d
n n
s
P
2
di − d
s=
n−1

Dr. P. Gurusamy
Small sample tests: t-test for difference of two means

Conclusion:
Case (i): If calculated value of |t| < tα , H0 is accepted at a
particular level of significance
Case (ii): If calculated value of |t| > tα , H0 is rejected at a
particular level of significance.

Note:
Always use the one tailed test for dependent sample.

Dr. P. Gurusamy
Problem No. 1

A drug was administered to 12 patients and increments in this

blood pressure were recorded to be 5, 2, 8, -1, 3, 0, -2, 1, 5, 0, 4
and 6. Is it reasonable to believe that the drug has no effective on
change of blood pressure?

Solution:
Given : n = the size of sample = 12 < 30
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 < µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 1
Step (iv): Calculations
X
di = 5 2 8 − 1 3 0 − 2 1 5 0 4 6 ⇒ di = 31
P
di 31
d = = = 2.58
n 12
s
P
2 r
di − d 104.92
s= = = 3.088
n−1 11
√
d √ 2.58 × 12
Test statistic t = × n= = 2.89.
s 3.088
The calculated value of |t| = 2.89.
The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n - 1 = 11 is t0.05 = 1.80.
Step (v): Conclusion:
Since |t| > tα , H0 is rejected at 5% level of significance.
Dr. P. Gurusamy
Problem No. 2
The daily production rates for a sample of factory workers before
and after a training program are shown below.
Worker 1 2 3 4 5
Before 6 10 9 8 7
After 9 12 10 11 9
We want to determine if the training program was effective
i) Give the hypothesis for this problem
ii) Compute test statistic
iii) At 95% confidence, test the hypotheses, that is did the training
program actually increase the production rates?

Solution:
Given : n = the size of sample = 5 < 30
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 < µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%.
Dr. P. Gurusamy
Problem No. 2

Step (iv): Calculations

2
Worker Before(x) After(y) d=y-x di − d
1 6 9 3 0.64
2 10 12 2 0.04
3 9 10 1 1.44
4 8 11 3 0.64
5 7 9 2 0.04
Total 11 2.8

Dr. P. Gurusamy
Problem No. 2

P
di 11
d = = = 2.2
n 5
s
P
2 r
di − d 2.8
s= = = 0.84
n−1 4
√
d √ 2.2 × 5
Test statistic t = × n= = 5.86.
s 0.84
The calculated value of |t| = 5.86.
The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n - 1 = 4 is t0.05 = 2.776.
Step (v): Conclusion:
Since |t| > tα , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Problem No. 3
Ten persons were appointed in the officer cadre in an office. Their
performance was noticed by giving a test and the marks were
recorded out of 100.

Employee A B C D E F G H I J
Before training 80 76 92 60 70 56 74 56 70 56
After training 84 70 96 80 70 52 84 72 72 50

By applying the t- test, can it be concluded that the employees

have been benefited by the training?

Solution:
Given : n = the size of sample = 10 < 30
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 < µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%.
Dr. P. Gurusamy
Problem No. 3

Step (iv): Calculations

2
Employee Before(x) After(y) d=x-y di − d
A 80 84 -4 0
B 76 70 6 100
C 92 96 -4 0
D 60 80 -20 256
E 70 70 0 16
F 56 52 4 64
G 74 84 -10 36
H 56 72 -16 144
I 70 72 -2 4
J 56 50 6 100
Total -40 720

Dr. P. Gurusamy
Problem No. 3

P
di −40
d = = = −4
n 10
s
P
2 r
di − d 720
s= = = 8.94
n−1 9
√
d √ −4 × 10
Test statistic t = × n= = −1.414.
s 8.94
The calculated value of |t| = 1.414.
The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n - 1 = 9 is t0.05 = 1.83.
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 4

The memory capacity of the students was tested before and after
the training. State whether the training was effective from the
following data
Student 1 2 3 4 5 6 7 8 9
Before 10 15 9 3 7 12 16 17 4
After 12 17 8 5 6 11 18 20 3

Solution:
Given : n = the size of sample = 9 < 30
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 < µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 4

Step (iv): Calculations

Student Before(x) After(y) d=x-y d2
1 10 12 -2 4
2 15 17 -2 4
3 9 8 1 1
4 3 5 -2 4
5 7 6 1 1
6 12 11 1 1
7 16 18 -2 4
8 17 20 -3 9
9 4 3 1 1
Total -7 29

Dr. P. Gurusamy
Problem No. 4

P
di −7
d = = = −0.777
n 9
rP r
d2 29
s = − d¯ =
2 − (−0.777)2 = 1.618
n 9
√
d √ −0.777 × 9
Test statistic t = × n= = −1.358.
s 1.618
The calculated value of |t| = 1.358.
The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n - 1 = 8 is t0.05 = 1.860.
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 5

The following are average weekly losses of working hours deu to

accidents in 10 industrial plants before and after an introduction of
a safety program was put in to operation
Before 45 73 46 124 33 57 83 34 26 17
After 36 60 14 119 35 51 77 29 24 11

Use at 0.05 level of significance to test whether the safety program

is effective.
Solution:
Given : n = the size of sample = 10 < 30
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 < µ2 (one tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 5

Step (iv): Calculations

Before(x) After(y) d=x-y d2
45 36 9 81
73 60 13 169
46 14 32 1024
124 119 5 25
33 35 -2 4
57 51 6 36
83 77 6 36
34 29 5 25
26 24 2 4
17 11 6 36
82 1440

Dr. P. Gurusamy
Problem No. 5

P
di 82
d = = = 8.2
n 10
rP r
d2 1440 √
s = − d¯2 = − (8.2)2 = 144 − 67.24 = 8.761
n 10
√
d √ 8.2 × 10
Test statistic t = × n= = 2.959.
s 8.761
The calculated value of |t| = 2.959.
The table value of t at 5% level of significance for one tailed test
with degrees of freedom v = n - 1 = 9 is t0.05 = 1.833.
Step (v): Conclusion:
Since |t| > tα , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Small sample tests: F -test for equality of two variances
Null hypothesis H0 :There is no significance difference between the
sample variances or H0 : σ12 = σ22 .

S12
Test statistic F = with S12 > S22
S22

with degrees of freedom (v1 , v2 ).

where v1 = n1 − 1 = degrees of freedom for sample having larger
variance and v2 = n2 − 1 = degrees of freedom for sample having
smaller variance.
x is sample 1 mean, y is sample 2 mean, n1 is the size of sample l,
n2 is the size of sample 2, s1 is standard Deviation of sample 1, s2
is standard Deviation of sample 2.
P
2 P
2
X −X Y −Y
S12 = &S22 =
n1 − 1 n2 − 1

Dr. P. Gurusamy
Small sample tests: F -test for equality of two variances

If s1 is standard Deviation of sample 1, s2 is standard Deviation of

sample 2 then

n1 s12 n2 s22
S12 = & S22 =
n1 − 1 n2 − 1

Conclusion:
Case (i): If calculated value of F < Fα , H0 is accepted at a
particular level of significance
Case (ii): If calculated value of F > Fα , H0 is rejected at a
particular level of significance.

Dr. P. Gurusamy
Small sample tests: F -test for equality of two variances

Note:
1. Test statistic F = larger variance/ smaller variance.
2. To test whether the samples have equal means
then we apply t – test.
3. To test whether the samples have equal variance
then we apply F – test.
4. To test whether the samples have come from the same normal
population, first we apply F – test then we apply t – test.
5. Suppose F- test accepted then we go to t – test.
6. If F – test fails, then we can not apply t – test and we say that
the samples have not come from same normal population.
7. If F – test accepted and t – test is also accepted then we say
that the samples have come from same normal population

Dr. P. Gurusamy
Problem No. 1

In one sample of 8 observations the sum of the squares of

deviations of the sample values from the sample mean was 84.4
and in the other sample of 10 observations, it was 102.6. Test
whether this difference is significant at 5% level, given that the 5%
point of F for v1 = 7 & v2 = 9 degrees of freedom is 3.29.

Solution:
Given : n1 = the size of sample 1 = 8 < 30
n2 = the size of sample 2 = 10 < 30 (small sample test)
Step (i): Null hypothesis H0 : σ12 = σ22 .
Step (ii): Alternative hypothesis H1 : σ12 6= σ22 .(two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 1

Step (iv): Calculations

Given
X
2 X
2
X − X = 84.4 & Y − Y = 102.6
P
2
X −X 84.4
S12= = = 12.057
n1 − 1 7
P
2
2 Y −Y 102.6
S2 = = = 11.4
n2 − 1 9
S12
Test statistic F = with S12 > S22
S22
12.057
Since S12 > S22 , Test statistic F = = 1.0576.
11.4
The calculated value of F = 1.0576.

Dr. P. Gurusamy
Problem No. 1

v1 = n1 − 1 = degrees of freedom for sample having larger

variance = 8 - 1 = 7
v2 = n2 − 1 = degrees of freedom for sample having smaller
variance = 10 - 1 = 9.
The table value of F at 5% level of significance for two tailed test
with degrees of freedom (v1 = 7, v2 = 9) is F0.05 = 3.29.
Step (v): Conclusion:
Since F < Fα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 2

Two random samples drawn from normal populations

Sample 1 20 16 26 27 23 22 18 24 25 19
Sample 2 27 33 42 35 32 34 38 28 41 43 30 37

Obtain the estimates of variances of the population and test

whether the population have the same variances.

Solution:
Given : n1 = the size of sample 1 = 10 < 30
n2 = the size of sample 2 = 12 < 30 (small sample test)
Step (i): Null hypothesis H0 : σ12 = σ22 .
Step (ii): Alternative hypothesis H1 : σ12 6= σ22 .(two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 2
Step (iv): Calculations : X = 22, Y = 35


2

2
I(X) X −X X −X II(Y) Y −Y Y −Y
20 -2 4 27 -8 64
16 -6 36 33 -2 4
26 4 16 42 7 49
27 5 25 35 0 0
23 1 1 32 -3 9
22 0 0 34 -1 1
18 -4 16 38 3 9
24 2 4 28 -7 49
25 3 9 41 6 36
19 -3 9 43 8 64
30 -5 25
- 37 2 4
220 120 420 314

Dr. P. Gurusamy
Problem No. 2

P P
X 220 Y 420
X = = = 22, Y = = = 35
n1 10 n2 12
X
2 X
2
X − X = 120 & Y − Y = 314
P
2
X −X 120
S12 = = = 13.33
n1 − 1 9
P
2
Y −Y 314
S22 = = = 28.55
n2 − 1 11
S22 28.55
Since S22 > S12 , Test statistic F = 2
= = 2.1417.
S1 13.33
The calculated value of F = 2.1417.

Dr. P. Gurusamy
Problem No. 2

v1 = n2 − 1 = degrees of freedom for sample having larger

variance = 12 - 1 = 11
v2 = n1 − 1 = degrees of freedom for sample having smaller
variance = 10 - 1 = 9.
The table value of F at 5% level of significance for two tailed test
with degrees of freedom (v1 = 11, v2 = 9) is F0.05 = 3.11.
Step (v): Conclusion:
Since F < Fα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 3

Two independent samples of 8 and 7 items respectively has the

following values
Sample I 9 11 13 11 15 9 12 14
Sample 2 10 12 10 14 9 8 7

Is the difference between the variances of the samples significance?

Solution:
Given : n1 = the size of sample 1 = 8 < 30
n2 = the size of sample 2 = 7 < 30 (small sample test)
Step (i): Null hypothesis H0 : σ12 = σ22 .
Step (ii): Alternative hypothesis H1 : σ12 6= σ22 .(two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 3
Step (iv): Calculations
P P
X 94 Y 70
X = = = 11.75, Y = = = 10.
n1 8 n2 7

2

2
I(X) X −X X −X II(Y) Y −Y Y −Y
9 -2.75 7.5625 10 0 0
11 -0.75 0.5625 12 2 4
13 1.25 1.5625 10 0 0
11 -0.75 0.5625 14 4 16
15 3.25 10.5625 9 -1 1
9 -2.75 7.5625 8 -2 4
12 0.25 0.0625 7 -3 9
14 2.25 5.0625
94 33.5 70 34

Dr. P. Gurusamy
Problem No. 3

P P
X 94 Y 70
X = = = 11.75, Y = = = 10.
n1 8 n2 7
X
2 X
2
X − X = 33.5 & Y − Y = 34.
P
2
X −X 33.5
S12 = = = 4.7857
n1 − 1 7
P
2
2 Y −Y 34
S2 = = = 5.6667
n2 − 1 6
S22 5.6667
Since S22 > S12 , Test statistic F = 2
= = 1.1840.
S1 4.7857
The calculated value of F = 1.1840.

Dr. P. Gurusamy
Problem No. 3

v1 = n2 − 1 = degrees of freedom for sample having larger

variance = 7 - 1 = 6
v2 = n1 − 1 = degrees of freedom for sample having smaller
variance 8 - 1 = 7.
The table value of F at 5% level of significance for two tailed test
with degrees of freedom (v1 = 6, v2 = 7) is F0.05 = 3.87
Step (v): Conclusion:
Since F < Fα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 4

Two random samples give the following results

Sample size sample mean sum of squares of
deviations from the mean
1 10 15 90
2 12 14 108

Test whether the samples come from the same population.

Solution:
(i) First we apply F-test Given : n1 = the size of sample 1 =
10 < 30
n2 = the size of sample 2 = 12 < 30 (small sample test)
Step (i): Null hypothesis H0 : σ12 = σ22 .
Step (ii): Alternative hypothesis H1 : σ12 6= σ22 .(two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 4

Step (iv): Calculations

Given X
2 X
2
X − X = 90 & Y − Y = 108.
P
2
X −X 90
S12 = = = 10
n1 − 1 9
P
2
2 Y −Y 108
S2 = = = 9.818
n2 − 1 11
10
Since S12 > S22 , Test statistic F = = 1.02.
9.818
The calculated value of F = 1.02.

Dr. P. Gurusamy
Problem No. 4

v1 = n1 − 1 = degrees of freedom for sample having larger

variance = 10 - 1 = 9
v2 = n2 − 1 = degrees of freedom for sample having smaller
variance 12 - 1 = 11.
The table value of F at 5% level of significance for two tailed test
with degrees of freedom (v1 = 9, v2 = 11) is F0.05 = 2.90.
Step (v): Conclusion:
Since F < Fα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 4

If F- test accepted the we apply t-test

(ii) t-test for two independent means

Solution:
Given : n1 is the size of sample l = 10 < 30,
n2 is the size of sample 2 = 12 < 30 (small sample test)
x1 is sample 1 mean = 15,
x2 is sample 2 mean = 14
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 6= µ2 (two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 4

Step (iv): Calculations

s
2 P
2 r
P
X − X̄ + Y − Ȳ 90 + 108
S = = = 3.146
n1 + n2 − 2 10 + 12 − 2

x̄ − ȳ 15 − 14
Test statistic t = q = q = 0.742.
S × n11 + 1
n2 3.146 × 101
+ 1
12

The calculated value of |t| = 0.742.

The table value of t at 5% level of significance for two tailed test
with degrees of freedom v = n1 + n2 − 2 = 10 + 12 − 2 = 20 is
t0.05 = 2.086.
Step (v): Conclusion:
Since |t| < tα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 4

Final Conclusion:
Both F-test and t-test are accepted at 5% level of significance.
Hence we may conclude that the given samples have been drawn
from the same normal population.

Dr. P. Gurusamy
Problem No. 5

Two random samples give the following results

Sample size sample mean sample variance
1 8 9.6 1.2
2 11 16.5 2.5

Test whether the samples come from the same population.

Solution:
(i) First we apply F-test Given : n1 = the size of sample 1 =
8 < 30
n2 = the size of sample 2 = 11 < 30 (small sample test)
Step (i): Null hypothesis H0 : σ12 = σ22 .
Step (ii): Alternative hypothesis H1 : σ12 6= σ22 .(two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 5

Step (iv): Calculations

Given

X = 9.6, Y = 16.5, s12 = 1.2, s22 = 2.5

n1 s12 8 × 1.2
S12 = = = 1.37
n1 − 1 7
n2 s22 11 × 2.5
S22 = = = 2.75
n2 − 1 10
S22 2.75
Since S22 > S12 , Test statistic F = 2
= = 2.01
S1 1.37
The calculated value of F = 2.01.

Dr. P. Gurusamy
Problem No. 5

v1 = n2 − 1 = degrees of freedom for sample having larger

variance = 11 - 1 = 10
v2 = n1 − 1 = degrees of freedom for sample having smaller
variance 8 - 1 = 7.
The table value of F at 5% level of significance for two tailed test
with degrees of freedom (v1 = 10, v2 = 7) is F0.05 = 3.64.
Step (v): Conclusion:
Since F < Fα , H0 is accepted at 5% level of significance.

Dr. P. Gurusamy
Problem No. 5

If F- test accepted the we apply t-test

(ii) t-test for two independent means

Solution:
Given : n1 is the size of sample l = 8 < 30,
n2 is the size of sample 2 = 11 < 30 (small sample test)
x1 is sample 1 mean = 9.6,
x2 is sample 2 mean = 16.5
Step (i): Null hypothesis H0 : µ1 = µ2
Step (ii): Alternative hypothesis H1 : µ1 6= µ2 (two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 5

Step (iv): Calculations

s
(8 × 1.2) + (11 × 2.5) √
r
n1 s12 + n2 s22
S= = = 2.182 = 1.477
n1 + n2 − 2 8 + 11 − 2

x̄ − ȳ 9.6 − 16.5
Test statistic t = q = q = −10.05.
S × n11 + 1
n2 1.477 × 18 + 1
11

The calculated value of |t| = 10.05.

The table value of t at 5% level of significance for two tailed test
with degrees of freedom v = n1 + n2 − 2 = 8 + 11 − 2 = 17 is
t0.05 = 2.11.
Step (v): Conclusion:
Since |t| > tα , H0 is rejected at 5% level of significance.

Dr. P. Gurusamy
Problem No. 5

Final Conclusion:
F-test is accepted and t-test is rejected at 5% level of significance.
Hence we may conclude that the given samples have not been
drawn from the same normal population.

Dr. P. Gurusamy
Problem No. 6

Can you say that the following two samples come from the same
normal population?
Sample I 24 27 26 23 25
Sample 2 27 30 28 31 22 36

Solution:
Given : n1 = the size of sample 1 = 5 < 30
n2 = the size of sample 2 = 6 < 30 (small sample test)
Step (i): Null hypothesis H0 : σ12 = σ22 .
Step (ii): Alternative hypothesis H1 : σ12 6= σ22 .(two tailed test)
Step (iii): Choose a level of significance α = 5%.

Dr. P. Gurusamy
Problem No. 6

Step (iv): Calculations

P P
X 125 Y 174
X = = = 25, Y = = = 29.
n1 5 n2 6

2

2
X X −X X −X Y Y −Y Y −Y
24 -1 1 27 -2 4
27 2 4 30 1 1
26 1 1 28 -1 1
23 -2 4 31 2 4
25 0 0 22 -7 49
36 7 49
125 10 174 108

Dr. P. Gurusamy
Problem No. 6

P P
X
125 Y 174
X = = = 25, Y = = = 29.
n1 5 n2 6
X
2 X
2
X − X = 10 & Y − Y = 108.
P
2
X −X 10
S12= = = 2.5
n1 − 1 4
P
2
2 Y −Y 108
S2 = = = 21.6
n2 − 1 5
S22 21.6
Since S22 > S12 , Test statistic F = 2
= = 10.285.
S1 2.5
The calculated value of F = 10.285.

Dr. P. Gurusamy
Problem No. 6

v1 = n2 − 1 = degrees of freedom for sample having larger

variance = 6 - 1 = 5
v2 = n1 − 1 = degrees of freedom for sample having smaller
variance 5 - 1 = 4.
The table value of F at 5% level of significance for two tailed test
with degrees of freedom (v1 = 5, v2 = 4) is F0.05 = 6.26
Step (v): Conclusion:
Since F > Fα , H0 is rejected at 5% level of significance.
If F- test is rejected then we can not apply t-test at 5% level of
significance.
Final Conclusion:
F-test is rejected at 5% level of significance. Hence we may
conclude that the given samples have not been drawn from the
same normal population.

Dr. P. Gurusamy