Topic

Probability and Statistics

F-Distribution
(Continuous Probability Distribution)

Prepared by:
Dr. Sunil
NIT Hamirpur (HP)
(Last updated on 21-11-2007)

F-Distribution: (Snedecor’s variance ratio distribution)

This distribution was introduces by the English statistician Prof. R.A. Fisher
(1890-1962), who had greatly influenced the development of modern statistics.

Introduction:
In testing the significance of the difference of two means of two samples, we
assume that the two samples came from the same population or population of same
variance. The object of F-test is to discover whether two independent estimates of
population variance differ significantly or whether the two samples may be regarded as
drawn from the normal populations having the same variance. Hence, before applying the
t-test for the significance of difference of two means, we have to test for the equality of
population variance by using F-test.

Let s12 be the sample variance of an independent sample of size n1 drawn from a

( )
normal population N µ1 , σ12 . Similarly, let s 22 be the sample variance of an independent

(
sample of size n 2 drawn from a normal population N µ 2 , σ 22 . )
1 n1 1 n2
Here s12 = ∑
n1 − 1 1
( x i − x ) 2
, s 2
2 = ∑
n2 −1 1
(y i − y )2 and x, y are the sample means.

Thus, s12 and s 22 are two variances of two random samples of sizes n1 and n 2 ,
respectively, drawn from two normal populations.
In order to determine whether the two samples come from two populations having
equal variances, consider the sampling distribution of the ratio of the variances of two
independent random samples defined by
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 2

s12 / σ12 σ 22 s12

F= = ,
s 22 / σ 22 σ12 s 22

which is an F-distribution with ν1 = n1 − 1 and ν 2 = n 2 − 1 degrees of freedom (d.f.).

To test whether these estimates s12 and s 22 are significantly different or if the samples
may be regarded as drawn from the same population or from two populations with

same variance σ 2 .

Null hypothesis (H0): First we set up the null hypothesis (H0): σ12 = σ 22 = σ 2 ,
i.e., the independent estimates of the common population do not differ significantly.
To carry out the test of significance of the difference of the variances, we calculate the
test statistic.
Under the hypothesis that two normal populations have the same variance,

( )
i.e., σ12 = σ 22 , we have

s12
F= .
s 22

F-distribution determines whether the ratio of two samples variances s12 and s 22 is too
small or too large. This distribution is also known as variance ratio distribution.
Conclusion:
If the calculated value of F exceeds F0.05 for ν1 = n1 − 1 and ν 2 = n 2 − 1 degrees
of freedom (d.f.) given in table, then we conclude that the ratio is significant at 5% level,
i.e., we conclude that the sample could have come from two normal populations with
same variance. This means that there is no significant difference between the variance
of the populations.
When F is close to 1, the two-sample variances s1 and s2 are nearly same. It is
expected, to take larger sample variance as the numerator, i.e., the larger of the variances
is placed in the numerator. F is always a positive number.
Geometrical representation:
Fα (ν1 , ν 2 ) is the value of F with ν1 and ν 2 dof such that the area under the F-

distribution curve the right of Fα is α . In table, Fα is tabulated (prepared by Snedecor)

for α = 0.05 and α = 0.01 for various combinations of the dof ν1 and ν 2 . The values of
F0.95 and F0.99 can be calculated from the tables using following result.
1
F1−α (ν1 , ν 2 ) = .
Fα (ν 2 , ν1 )
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 3

Properties of F-distribution:
(i) The F-distribution curve lies entirely in first quadrant and is unimodal.

(ii) The F-distribution is independent of the population variance σ 2 and depends on ν1

and ν 2 but also on the order in which they are stated.
(iii) The F-curve is J-shaped if ν 2 ≤ 2 and well shaped for ν1 > 2 .
(iv) It can be shown that the mode of F-distribution is less than unity.
Significance Test:
Snedecor’s F-tables give 5% and 1% points of significance for F. 5%
points of F mean that area under the F-curve to the right of the ordinate at a
value of F, is 0.05. Clearly, value of F at 5% significance is lower than that
at 1%. Thus, the table gives only single tail test. If however, we are testing
the hypothesis that the population variances are the same, we must use
both tail areas under the F-curve. In such cases, F-table will provide 10%
and 2% level significance.
Assumptions on which F-test is based:
• The populations of each sample must be normally distributed.
• The samples must be random and independent.

• The ratio of σ12 to σ 22 should be equal to 1 or greater than 1. That is why

we take the larger variance in the numerator of the ratio.
Applications:
• F-distribution is extremely useful in testing the equality of several population
means, comparing sample variances, and forms the backbone of analysis of
variance.
• F-test is used to test, whether two independent samples have been drawn from

normal populations with the same variance σ 2 .

• F-test is used to test, whether two independent estimates of the population
variance are homogeneous or not.

Fisher’s z-Distribution:
1
Changing the variable F to z by the substitution z = log e F ⇒ F = e 2z in the F-
2
distribution, we get the Fisher’s z-distribution, which is of the form

( )
y = y 0 e ν1z ν1e 2z + ν 2 , where y0 is so chosen that the area under the curve is unity.
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 4

It is more nearly symmetrical than F-distribution. Table showing the values of z

that will be exceeded in simple sampling with probabilities 0.05 and 0.01 have been
prepared for various values of ν1 and ν 2 .

Significant Test:
As z-tables give only critical values corresponding right hand tail areas, therefore
5% (or 1%) points of z imply that the area to the right of the ordinate at z is 0.05% (or
0.01%). In other words, 5% and 1% points of z correspond to 10% and 2% levels of
significance, respectively.

Now let us solve some problems based on F-distribution:

Q.No.1.: For an F-distribution, find

(a). F0.05 with v1 = 7 and v2 = 15,
(b). F0.01 with v1 = 24 and v2 = 19,
(c). F0.95 with v1 = 19 and v2 = 24,
(d). F0.99 with v1 = 28 and v2 = 12.
Sol.: (a). From table, F0.05 with v1 = 7 and v2 = 15 is 2.71. Ans.
(b). From table, F0.01 with v1 = 24 and v2 = 19 is 2.92. Ans.
1 1
(c). From table, F0.95 (19, 24) = = = 0.473933 . Ans.
F0.05 (24,19 ) 2.11

1 1
(d). From table, F0.99 (28, 12) = = = 0.34482 . Ans.
F0.01 (12,28) 2.90
Q.No.2.: Determine the probability that the variance of the first sample size n1= 9 will be
at least 4 times as large as variance of the second sample of size n2 = 16 if the
two samples are independent random samples from a normal population.
Sol.: From table, F0.01 = 4.0 for v1 = n 1 − 1 = 9 − 1 = 8, v 2 = n 2 − 1 = 16 − 1 = 15 .
Then, the desired probability is 0.01.
Q.No.3.: The household net expenditure on health care in south and north India, in two
samples of households, expressed as percentage of total income is shown the
following table:
South 15.0 8.0 3.8 6.4 27.4 19.0 35.3 13.6
North 18.8 23.1 10.3 8.0 18.0 10.2 15.2 19.0 20.2
Test the equality of variances of households net expenditure on health care in
south and north India.
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 5

Sol.: Let the net expenditure of the south and north be considered as N µ1 , σ12 and ( )
( )
N µ 2 , σ 22 , respectively.

Step 1.: Null hypothesis (H0): σ12 = σ 22 , i.e., two normal population have the same
variance.

Step 2.: Alternative Hypothesis (H1): σ12 ≠ σ 22 , i.e., difference in variances.

Step 3.: Computations: n1 = 8, n2 = 9.
8
∑ x 1i = 15 + 8 + 3.8 + 6.4 + 27.4 + 19 + 35.3 + 13.6 = 128.5 .
i =1

8
∑ x 12i = (15)2 + 8 2 + (3.8)2 + (6.4 )2 + ......... + (13.6 )2 = 2887.21 .
i =1

9 9
∑ x 2i = 142.8, ∑ x 22i = 2485.26 .
i =1 i =1

1  8 (∑ x1i )2  1  (128.5)2  = 117.596996 .

s12 ∑ 1i
2
So = x −  =  2887 .21 − 
n 1 − 1  i =1

n1  7  8 

1  9 2 (∑ x 2i ) 
2
1  (142.8)2  = 27.4375 .
Also s 22 = ∑ x 2i − = 2485.26 − 
n 2 − 1  i =1 n2  8  9 

Step 4.: Test statistic:

s12 117.59696
F= = = 4.28599 .
s 22 27.4375

Now for ν1 = 8, ν 2 = 9 , we have (from Table) F0.05 = 3.23.

Conclusion: Since the calculated value of F is greater than the tabulated value.
Hence, hypothesis H0 is rejected.
i.e., there is significance difference between the population variances,
Thus, the population variances are significantly different.
Q.No.4.: Two samples of sizes 9 and 8 give the sum of squares of deviations from their
respective means equal to 160 inches2 and 91 inches2, respectively. Can these
be regarded as drawn from the same normal population?

Sol.: Null hypothesis (H0): σ12 = σ 22 , i.e., two normal population have the same variance.

1 n1 n2
Since we know s12 = ∑ (x i − x )2 and s 22 = 1 ∑ (y i − y )2 ,
n1 − 1 1 n2 −1 1
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 6

where x, y are the sample means.

Here given ∑ (x − x )2 = 160 and ∑ (y − y)2 = 91 .

160 91
∴ s12 = = 20 and s 22 = = 13 .
8 7

s12 20
Hence, F = = = 1.54 nearly.
s 22 13

In F-distribution, ν1 = n 1 − 1 and ν 2 = n 2 − 1 are degrees of freedom.

Now for ν1 = 8, ν 2 = 7 , we have (from Table) F0.05 = 3.73.
Conclusion: Since the calculated value of F is less than F0.05 (tabulated value).
Hence, hypothesis H0 is accepted.
∴ There is no significant difference between the variance of the populations.
Thus, the two samples can be regarded as drawn from two normal populations with the
same variance. If the two populations are to be same, their means should also be the
same, which can be verified by applying t-test provided the sample means are known.
Q.No.5.: Measurements on the length of a copper wire were taken in 2 experiments A
and B as under
A’s measurements (mm) : 12.29, 12.25, 11.86, 12.13, 12.44, 12.78, 12.77, 11.90, 12.47.
B’s measurements (mm) : 12.39, 12.46, 12.34, 12.22, 11.98, 12.46, 12.23, 12.06 .
Test whether B’s measurements are more accurate than A’s (the readings taken
in both cases being unbiased).
Sol.: Readings in both cases being unbiased, B’s measurements will be taken more
accurate if its population variance in less than that of A’s measurements.

Null hypothesis (H0): σ12 = σ 22 , i.e., two normal population have the same variance.
Under this hypothesis, we have

s12
F= with ν1 = n 1 − 1 = 8 and ν 2 = n 2 − 1 = 7 .
s 22
We calculate the s.d.’s of two series as follows:
A’s measurements B’s Measurements
x u = 100(x − 12 ) u2 y v = 100(y − 12 ) v2
12.29 29 841 12.39 39 1521
12.25 25 625 12.46 46 2116
11.86 − 14 196 12.34 34 1156

12.13 13 169 12.22 22 484

F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 7

12.44 44 1936 11.98 −2 4

12.78 78 6084 12.46 46 2116

12.77 77 5929 12.23 23 529
11.90 − 10 100 12.06 6 36

12.47 47 2209
289 18089 214 7962

1  (289)2  = 1 (18089 − 9280) = 1101.1 .

∴ s12 = 18089 − 
n 1 − 1  n 1  8

1  (214)2  = 1 (7962 − 5724) = 319.7 .

s 22 = 7962 − 
n 2 − 1  n 2  7

s12 1101.1
F= = = 3.44 .
s 22 319.7

For ν1 = 8 and ν 2 = 7 , from table, F0.05 = 3.73 and F0.01 = 6.84.

Conclusion: Since the calculated value of F is less than both F0.05 and F0.01 (tabulated
values).
Hence, hypothesis H0 is accepted.
Thus, the result is not significant at both 5% and 1% level.
Hence, there is no reason to say that B’s measurements are more accurate than
those of A’s .
Q.No.6.:In two independent samples of sizes 8 and 10 the sum of squares of derivations
of the sample values from the respective sample means were 84.4 and 102.6.
Test whether the difference of variances of the populations is significant or not.

Sol.: Null hypothesis (H0): σ12 = σ 22 , i.e., two normal population have the same variance.

s12
Under this hypothesis H0: F = ~ F(ν1 , ν 2 d.f .) ,
s 22

where ν1 = n 1 − 1 , n1 = sample I size = 8;

ν 2 = n 2 − 1 , n1 = sample II size = 10.

∑ (X1 − X1 ) = 84.4 ; ∑ (X 2 − X 2 ) = 102.6 .

2 2

s12 =
∑ (X1 − X1 )
2
=
84.4
= 12.057 ; s 22 =
∑ (X2 − X2 )
2
=
102.6
= 11.4 .
n1 − 1 7 n 2 −1 9

s12 12.057
F=
s 22
=
11.4
= 1.0576 . (Q s 2
1 > s 22 )
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 8

The tabulated value of F at 5% level of significance for (7, 9) d.f. is 3.29.

∴ F0.05 = 3.29
Conclusion: Since the calculated value of F is less than F0.05 (tabulated value).
Hence, hypothesis H0 is accepted.
∴ There is no significant difference between the variance of the populations.
Q.No.7.: Two independent samples of sizes 7 and 6 had the following values:
Sample A 28 30 32 33 31 29 34
Samples B 29 30 30 24 27 28

Examine whether the samples have been drawn from normal populations
having the same variance.

Sol.: Null hypothesis (H0): σ12 = σ 22 , i.e., two normal population have the same variance.

Alternative Hypothesis (H1): σ12 ≠ σ 22 .

Under null hypothesis, the statistic
s12
F= , (if s 2
1 > s 22 )
s 22
Computations for s12 and s 22 :

X1 X 1 − X1 (X1 − X1 )2 X2 X2 − X2 (X 2 − X 2 )2
28 −3 9 29 1 1
30 −1 1 30 2 4
32 1 1 30 2 4
33 2 4 24 −4 16
31 0 0 27 −1 1
29 −2 4 28 0 0
34 3 9
28 26

∑ (X1 − X1 ) = 28 .
2
X1 = 31 ; n1 = 7;

∑ (X 2 − X 2 ) = 26 .
2
X 2 = 28 ; n2 = 6;

s12 =
∑ (X1 − X1 )
2
=
28
= 4.666 ; s 22 =
∑ (X 2 − X 2 )
2
=
26
= 5.2
n1 − 1 6 n2 −1 5

s 22 5.2
F=
s12
=
4.666
= 1.1158 . (Q s 2
2 > s12 )
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 9

The tabulated values of F at ν1 = 6 − 1 and ν 2 = 7 − 1 d.f. for 5% level of significance is

4.39.
Conclusion: The calculated value of F is less than the tabulated value of F.
Hence, hypothesis H0 is accepted.
i.e., there is no significance difference between the population variances,
i.e., the samples have not been drawn from the normal population with same variance.
Q.No.8.: Two random samples drawn from two normal populations have the variable
values as below:
Sample I 19 17 16 28 22 23 19 24 26
Sample II 28 32 40 37 30 35 40 28 41 45 30 36
Obtain the estimate of variance of the population and test whether the two
population have the same variance.

Sol.: X1 =
∑ X1 = 21.55 , n = 9 , X2 =
∑ X2 = 35.166 , n 2 = 12 .
1
n1 n2

X1 d1 = X1 − 17 d12 X2 d 2 = X 2 − 28 d 22

19 2 4 28 0 0
17 0 0 32 4 16
16 −1 1 40 12 144
28 11 121 37 9 81
22 5 25 30 2 4
23 6 36 35 7 49
19 2 4 40 12 144
24 7 49 28 0 0
26 9 81 41 13 169
45 17 289
30 2 4
36 8 64

∑ d12 = 321 ∑ d 22 = 964

s12 =
∑ (X 1 − X1 )
2
=
∑ (
d12 − n 1 X1 − A )2 = 321 − 9(21.55 − 17 )2 = 16.834 ;
n1 − 1 n1 − 1 9 −1

s 22 =
∑ (X2 − X2 )
2
=
∑ d 22 − n 2 (X 2 − A )2
=
964 − 12(35.166 − 28)2
= 31.616
n 2 −1 n 2 −1 12 − 1
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 10

s 22 31.616
F= = = 1.878 . (Q s 2
2 > s12 )
s12 16.834

The calculated value of F is 1.878.

The tabulated value of ν 2 = 12 − 1 = 11 , ν1 = 9 − 1 = 8 d.f. at 5% level of significance is
3.315.
Conclusion: Since the calculated value of F is less than the tabulated value.
Hence, hypothesis H0 is accepted.
i.e., there is no significance difference between the population variance,
i.e., the two populations has the same variance.

Remarks: If the two populations are to be same, their means should also be the same,
which can be verified by applying t-test provided the sample means are known.

Q.No.9.: Two random samples are drawn from two normal populations are as follows:
A 17 27 18 25 27 29 13 17
B 16 16 20 27 26 25 21

Test whether the tests are drawn from same normal population.
Sol.: To test whether the two independent samples have been drawn from the same
population, we have to test
(i) equality of the means by applying the t-test and
(ii) equality of the population variances by applying F-test.
Since the t-test assumes that the sample variances are equal, we shall first apply the F-
test.
(i). F-test:

Null hypothesis (H0): σ12 = σ 22 , i.e., two normal population have the same variance.

Alternative Hypothesis (H1): σ12 ≠ σ 22 .

s12
Test statistic: F =
s 22
, (if s 2
1 )
> s 22 .

Computations for s12 and s 22 :

X1 X2
X 1 − X1 (X1 − X1 )2 X2 − X2 (X 2 − X 2 )2
17 − 4.625 21.39 16 − 2.714 7.365
27 5.735 28.89 16 − 2.714 7.365
18 − 3.625 13.14 20 1.286 1.653
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 11

25 3.375 11.39 27 8.286 68.657

27 5.735 28.89 26 7.286 53.085
29 7.735 54.39 25 6.286 39.513
13 − 8. 625 74.39 21 2.286 5.226
17 − 4.625 21.39

∑ (X1 − X1 ) = 253.87 .
2
X1 = 21.625 ; n1 = 8;

∑ (X 2 − X 2 ) = 182.859 .
2
X 2 = 18.714 ; n2 = 7;

s12 =
∑ (X 1 − X1 )
2
=
253.87
= 36.267 ; s 22 =
∑ (X 2 − X 2 )
2
=
182.859
= 30.47 .
n1 − 1 7 n 2 −1 6

s12 36.267
F= = = 1.190 .
s 22 30.47

The table value of F by ν1 = 7 and ν 2 = 6 degrees of freedom at 5% level is 4.21.

Conclusion: The calculated value of F is less than the tabulated value of F.
Hence, hypothesis H0 is accepted.
Hence, we conclude that the variability in two populations is same.
(ii). t-test:
Null hypothesis (H0): µ1 = µ 2 , i.e., the population means are equal.
Alternative Hypothesis (H1): µ1 ≠ µ 2 .
Test of statistic

s 2
=
∑ (X1 − X1 ) + ∑ (X 2 − X 2 )
2 2
=
253.87 + 182.859
= 33.594 ⇒ s = 5.796 .
n1 + n 2 − 2 8+7−2

X1 − X 2 21.625 − 18.714
t= = = 0.9704 ~ t (n 1 + n 2 − 2 )d.f .
1 1 1 1
s + 5.796 +
n1 n 2 8 7

The tabulated value of t at 5% level of significant for 13 d.f. is 2.16.

Conclusion: The calculated value of t is less than the tabulated value.
Hence, hypothesis H0 is accepted.
i.e., there is no significant difference between the population means, i.e., µ1 = µ 2 .
Therefore, we conclude that the two samples have been drawn from the same
normal population.
Q.No.10.: The two random samples reveal the following data
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 12

Sample No. Size Mean Variance

I 16 440 40
II 25 460 42

Test whether the samples come from the same normal population.
Sol.: A normal populations have two parameters namely the mean µ and the variance

σ 2 . To test whether the two independent samples have been drawn from the same normal
population, we have to test
(i) the equality of means,
(ii) the equality of variance.
Since, the t-test assumes that the sample variances are equal, we first apply F-test.
(i). F-test:

Null hypothesis (H0): σ12 = σ 22 , i.e., two normal population have the same variance.

Alternative Hypothesis (H1): σ12 ≠ σ 22 .

s12
Test statistic: F =
s 22
, (if s 2
1 )
> s 22 .

Given : n1 = 16, n2 = 25, s12 = 40 and s 22 = 42.

n 1s12
s12 n1 − 1 16 × 40 24
∴F = = = × = 0.9752 .
s 22 n 2 s 22 15 25 × 42
n 2 −1
The calculated value of F is 0.9752. The tabulated value of F at 16 − 1, 25 − 1 d.f. for 5%
level of significance.
Conclusion: The calculated value of F is less than the tabulated value of F.
Hence, hypothesis H0 is accepted.
(ii). t-test:
Null hypothesis (H0): µ1 = µ 2 , i.e. the population means are equal.
Alternative Hypothesis (H1): µ1 ≠ µ 2 .

Given n1 = 16, n2 = 25, X1 = 440 and X 2 = 460 .

Test of statistic:

n1s12 + n 2 s 22 16 × 40 + 25 × 42
s2 = = = 43.333 ⇒ s = 6.582 .
n1 + n 2 − 2 16 + 25 − 2
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 13

X1 − X 2 440 − 460
t= = = −9.490
1 1 1 1
s + 6.582 +
n1 n 2 16 25

The tabulated value of t is 9.490.

The tabulated value of t for (n 1 + n 2 − 2)d.f . , i.e.,39 d.f. for 5% level of significant is
1.96.
Conclusion: Since the calculated value of t is greater than the tabulated value.
Hence, hypothesis H0 is rejected.
i.e., there is significant difference between the population means, i.e., µ1 = µ 2 .
Since, there is significance difference between means, and no significant difference
between variance, we conclude that the samples do not come from the same normal
population.
Now let us solve some problems based on Fisher’s z-distribution:

Q.No.11.: Two gauge operations are tested for precision in making measurements. One
operator completes a set of 26 readings with a standard deviation of 1.34 and
the other does 34 readings with a standard deviation of 0.98. What is the level
of significance of this difference?
(Given that for ν1 = 25 and ν 2 = 33 ; z0.05 = 0.305, z0.01 = 0.432).

Sol.: We have n1 = 26, σ x = 1.34 ; n2 = 26, σ y = 0.98 .

n1
∴ s12 = .σ 2x =
26
(1.34)2 ≈ (1.34 )2 and s 22 = n 2 .σ 2y = 34 (0.98)2 ≈ (0.98)2 .
n1 − 1 25 n 2 −1 33
2
 1.34  1
Hence, F =   = 1.8696 and z = log e F = 1.1513 log10 1.8696 = 0.3129.
 0.98  2
Conclusion: Since the calculated value of z is just greater than z0.05 and less than z0.01.
Thus, the difference between the standard deviation is just significant at 5% level and
insignificant at 1% level.
*** *** *** *** ***
*** *** ***
***

Home Assignments
Q.No.1.: Find the value of
a). F0.05 for v1 = 15 and v2 = 7,
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 14

b). F0.95 for 12 and 15 dof,

c). F0.99 for 6 and 20 dof,
d). F0.95 for v1 = 10 and v2 = 20.
Ans.: a). 3.51,
1 1
b). F0.95 (12,15) = = = 0.38 .
f 0.05 (15,12) 2.62

1 1
c). F0.99 (6,20) = = = 0.135135 .
f 0.01 (20,6) 7.4

1 1
d). F0.95 (10,20) = = = 0.36 .
f 0.05 (20,10) 2.77
Q.No.2.: Find the probability that the variance of the first sample will be at least 3 times
as large as that of the second sample if two independent random samples of size
n1 = 7 and n2 = 13 are taken from normal population.
Ans.: 0.05
Q.No.3.: If independent random samples of size n1 = n2 = 8 come from normal
populations having the same variance, what is the probability that either
sample variance will be at least seven times as large as the other.
Ans.:0.01
Q.No.4.:Can we conclude that the two population variances are equal for the following
data of post graduates passed out from a state and ‘private’ university.
State 8350 8260 8130 8340 8070
Private 7890 8140 7900 7950 7840 7920
Ans.: F-ratio is 1.44. The variances are not significantly different.
Q.No.5.: Is there reason to believe that the life expected in south and north India is same
or not from the following data
South 34.0 39.2 46.1 48.7 49.4 45.9 55.3 42.7 43.7
North 49.7 55.4 57.0 54.2 50.4 44.2 53.4 57.5 61.9 56.6 58.2
Ans.: Yes, variances of life expectancy is same for the north and south since

s12 37.843
F= = = 1.603 .
s 22 23.607

Q.No.6.: Two random samples from two normal populations are given below:
Sample I 16 26 27 23 24 22
Sample II 33 42 35 32 28 31
Do the estimates of population variances differ significantly?
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 15

Degrees of Freedom (5, 5) (5, 6) (6, 5)

5% value of F 5.05 4.39 4.95

Ans.: F = 1.49, Not significant.

Q.No.7.: The mean diameter of rivets produced by two firms A and B are practically the
same but their standard deviations are different. For 16 rivets manufactured by
firm A, the SD is 3.8 mm while for 22 rivets manufactured by firm B is 2.9
mm. Do you think products of firm A are the better quality than those of firm
B.
Ans.: Yes.
Q.No.8.: The I.Q.’s of 25 students from one college showed a variance of 16 and those of
an equal number from the other college had a variance of 8. Discuss whether
there is any significant difference invariability of intelligence.
Ans.: F = 2, F(5%) = 1.98, F(1%) = 2.62. Just significant at 5% level and not significant
at 1%.
Q.No.9.: Two samples of 9 and 7 individuals have variances4.8 and 9.6, respectively. Is
the variance 9.6 significantly greater than the variance 4.6?
Ans.: First variance can not be regarded as significantly greater than the second.
Q.No.10.: Tests for breaking strengths were carried out on two lots of 5 and 9 steel wires.
The variance of one lot was 230 and that of other was 492. Is there a significant
difference in their variability?
Ans.: Not significant as F = 2.1 & and F0.05 = 4.15.
Q.No.11.: In two groups of ten children each, the increase in weight due to different diets
during the same period were in pounds
3 7 5 6 5 4 4 5 3 6
8 5 7 8 3 2 7 6 5 7
Is there is a significant difference in their variability?
Ans.: Not significant as F = 2.4 & and F0.01 = 3.2.
Q.No.12.: The daily wages in Rupees of skilled workers in two cities are as follow:
Size of sample of workers S.D. of wages in the sample
City A 16 25
City B 13 32

Ans. Accepted
Q.No.13.: The standard deviation calculated from two random samples of sizes 9 and 13
F-Distribution and Fisher’s z-distribution: Prepared by: Dr. Sunil, NIT Hamirpur (HP) 16

are 2.1 and 1.8, respectively. May the samples be regrated as drawn from
normal populations with the same standard deviation?
Ans. Accepted
Q.No.14.: Two independent samples of size 8 and 9 had the following values of the
variables:
Sample I 20 30 23 25 21 22 23 24
Sample II 30 31 32 34 35 29 28 27 28
Do the estimates of the population variance differ significantly?
Ans. Accepted.
Q.No.15.: From the following two sample values find out whether they have come from
the same population:
Sample I 17 27 18 25 27 29 27 23 17
Sample II 16 16 20 16 20 17 15 21
or
Show, how you would use Fisher’s z-test to decided whether the two sets of
observations 17, 27, 18, 25, 27, 29, 27, 23, 17 and 16, 16, 20, 16, 20, 17, 15,
21 indicate samples from the same universe.
Ans. Rejected.

*** *** *** *** ***

*** *** ***
***