Question No.

1 Spring 2011
A dice was tossed 144 times and following outcomes were recorded:
Faces 1 2 3 4 5 6
Observed Occurrence 22 23 27 25 26 21
Using chi - square test at 5% level of significance, assess the hypothesis that the dice is fair.
Answer No. 1 Spring 2011
H ; die is fair (p = 1/6)

Ha ; die is not fair (p ≠ / )

Level of significance = ∝ = 0.05

−e
Test Statistics = X =∑
e

Calculations ;
�
� � −
� (w-1) �−

22 24 4 0.1667
23 24 1 0.04167
27 24 9 0.375
25 24 1 0.04167
26 24 4 0.1667
21 24 9 0.375
f −f
∑ = .
f

Calculation of fe

The probability of any outcome of a die is 1/6

fe = (prob.)× total frequency

= 1/6 ×144 = 24(i.e out of 144 throws 24 times of each outcome is expected to occur)

Critical region

We reject Ho if
X cal <X −
∝
− or X cal >X ∝
−

X cal <X . − or X cal > X . −

X cal < 0.831 or X cal > 12.83

Conclusion

We accepted HO as the calculated value falls in acceptance region.

Hence we may conclude that the die is fair

Question No. 2 Spring 2010

A group of people was surveyed about their favorite car. The following results were obtained:

Gender Frequency
Civic Corolla Liana
Male 27 37 11
Female 26 14 5

At 5% level of significance, test the hypothesis that the choice of favourite car is independent of
one's gender.

Answer No. 2 Spring 2010

Table of Observed Values
Gender Frequency
Civic Corolla Liana Total
Male 27 37 11 75
Female 26 14 5 45
Total 53 51 16 120

Table of Expected Values

Gender Frequency
Civic Corolla Liana Total
Male 33 32 10 75
Female 20 19 6 45
Total 53 51 16 120

Observed Values Expected Values

(o-e)2
O e o-e
e
 27 33 -6 1.09
37 32 5 0.78
11 10 1 0.10
26 20 6 1.80
14 19 -5 1.32
5 6 -1 0.17
5.26

Test of hypothesis

H0: The choice of favorite car is independent of one's gender

H1 : The choice of favorite car is associative of one's gender
Level of significance α = 0.05 (5%)

Critical region �2 < �20.05 = 5.991 for (s – 1)(t – 1) = (2 – 1)(3 – 1) = 2 degrees of freedom

Since the calculated value of �20.05 is less than critical value, null hypothesis is accepted and
concluded that the choice of favorite car is independent of one's gender.

Question No. 3 Spring 2010

Tara Electronics claims that its energy saver bulbs have an average life of 6500 hours. A
consumer rights protection agency tested 15 such bulbs to check this claim. It found that the
mean life of 15 bulbs was 6300 hours with a standard deviation of 200 hours. At the 5%
significance level, assess the claim of Tara Electronics. Assume that life of such bulbs has an
approximately normal distribution.

Answer No. 3 Spring 2010

H0 = µ = 6500
H1 = µ ≠ 6500

Tabulated value α14,0.05 = 2.145

x− µ −
Test statistics t = s = = – 3.87
√n √

Since the absolute calculated value of |t| = 3.87 is greater than the tabulated value of t at

5%level of significance against 14 degrees of freedom the null hypothesis is rejected and
concluded that claim manufacture is unjustified.
Question No. 4 Autumn 2009
A cigarette manufacturer claims that the amount of nicotine in each cigarette is 7.5 mg. A
random sample of 40 cigarettes was tested and found to have a mean nicotine content of 7.67 mg
with a standard deviation of 0.6 mg. Test the manufacturer's claim at 5% level of significance.

Answer No. 4 Autumn 2009

Because nicotine should not be excessive, the hypothesis is described as

H0 : µ < 7.5 mg
H1 : µ > 7.5 mg
α = 0.05

x− µ . – . .
Test Statistics = Z = s = . = = √ = 1.79
.
√n

Since the calculated value of Z = 1.79 is greater than the tabulated value of Z = 1.645 against
5% level of significance test being one tailed null hypothesis is rejected and concluded that
manufacturer's claim is not correct.

Question No. 5 Spring 2009

a) An auditor claims that 10% of the customers' ledger accounts contain mistakes. A
random sample of 600 accounts was taken to test the accuracy of ledger accounts and
mistakes were detected in 45 accounts. Using 5% level of significance, explain
whether the sample result is consistent with the auditor's claim.
b) The city government has conducted a survey for assessing the inclination of citizens
towards installing backup power arrangements in their homes. It has collected the
following data:

Owners Tenants Total

Generator only 220 200 420
UPS only 160 170 330
Generator + UPS 140 110 250
No backup 180 220 400
Total 700 700 1,400

Using chi-square test at 5% level of significance, assess the hypothesis that installation of backup
power arrangements is independent of the types of residents i.e. owners or tenants.

Answer No. 5(a) Spring 2009

The question relates to test of hypothesis about population proportion where sample size is large
Given P (probability of success) = 45/600 = 0.075 and p = 0.1

Null Hypothesis H0 : p = 0.1

Alternative hypothesis H1 : p ≠ 0.1

P− . – . − .
Test statistics = Z = = = = – 2.33
.
√ √ . .
n

The critical value of |Z| at 5% level of significance is 1.96 which is less than the calculated value
of |Z| = 2.33, the null hypothesis is rejected and concluded that result is not consistent with the
auditor's claim.

Answer No. 5(b) Spring 2009

The expected values from the observed values

Owners (A) Tenants (B) Total

Generator only (C) 210 210 420
UPS only (D) 165 165 330
Generator + UPS (E) 125 125 250
No backup (F) 200 200 400
Total 700 700 1,400

For expected values (AC) = (A)(C) = 700 x 420 = 210

N 1400

Similarly other intersecting value are obtained

i− i
�2 = ∑ i

�2 Table

(o - e)2
Observed value Expected value (o-e) (o-e)2
e
220 210 10 100 0.48
160 165 -5 25 0.15
140 125 15 225 1.80
180 200 -20 400 2.00
200 210 -10 100 0.48
170 165 5 25 0.15
110 125 -15 225 1.80
220 200 20 400 2.00
8.86
H0: The back up arrangement is independent of type of residents
H1: The back up arrangement is not independent but is associated
Level of significance α = 0.05
Degrees of freedom = (n – 1) (m – 1) = (4 – 1) (2 – 1) = 3

Since the calculated value of �2 = 8.86 is greater than the tabulated value of �2 = 7.815 at
5% level of significance against 3 degrees of freedom null hypothesis is rejected and concluded
that there is association between two classifications.

Question No. 6 Spring 2008

A random sample of size 16 has a mean of 53. The sum of squares of deviations of values from
the mean is 150. Assuming the population values are approximately normally distributed, test the
hypothesis that population mean is 56. (Take α = 0.05)

Answer No. 6 Spring 2008

Given mean = 53, n = 16
∑(x – x)2 = 150
H0 : µ = 56
H1 : µ ≠ 56
x− µ − −
Test statistics t = s = . = .
= – 3.8
√n √

∑ �−�
Where s = √ =√ = 3.16
−

Since the absolute calculated value of t = 3.8 is greater than tabulated value of t = 2.31 at 5%
level of rejection against 15 degrees of freedom the null hypothesis is rejected and concluded
that mean value is not 56.

Question No. 7 Autumn 2007

A manufacturer claims that average life of a certain type of product is 270 hours with a standard
deviation of 25 hours. A sample of 12 units when tested showed a mean life of 255 hours. Using
a significant level of 0.05, evaluate the manufacturer's claim.

Answer No. 7 Autumn 2007

Given µ = 270, α = 25, x = 255, n = 12
Null Hypothesis H0 : µ = 270
Alternative Hypothesis H1 : µ ≠ 270
Level of significance α = 0.05
Population standard deviation is given
�− µ −
Test statistics z = � = = –2.08
√� √

Since the absolute calculated value of | z | = 2.08 is greater than the tabulated value of Z
at 5% level of significance which is 1.96 the manufacturer claim is rejected that average life of
the product is 270 hours.

Question No. 8 Spring 2007

When properly adjusted, an automatic machine should produce parts that have a mean diameter
of 25 millimeters (mm). Part diameters are normally distributed. The mean diameter of a sample
of 10 parts is 25.02mm with sample standard deviation = 0.024. Perform a hypothesis test at 5%
level of significance to evaluate whether the machine is working properly.

Answer No.8 Spring 2007

Given µ = 25, x = 25.02, s = 0.024, n = 10
Null-Hypothesis H0: µ = 25
Alternative Hypothesis H1: µ ≠ 25
Level of significance α = 0.05
Degrees of freedom = 10 – 1 = 9
�− µ . −
Test statistics = � = . = 2.64
√� √

Since the calculated value of t = 2.64 is greater than tabulated value of t at 5% level of
significance against 9 degrees of freedom = 2.262 the null hypothesis is rejected and concluded
that machine is not working properly.

Question No. 9 Autumn 2006

It is claimed that by using an automatic filling machine, the average amount of chips in each bag
is 100 grams. A random sample of 36 bags of chips showed a mean weight of 101.25 grams with
a standard deviation of 2.5 grams. Is the machine operating properly at 0.05 level of
significance?

Answer No. 9 Autumn 2006

Given µ = 100, X = 101.25, s = 2.5, n = 36
Null Hypothesis H0: µ = 100 gms
Alternative Hypothesis H1 : µ ≠ 100
Level of significance α0.05 = 1.96
 �− µ . −
Test statistics Z = � = . = 3
√� √

Since the calculated value of Z = 3 is greater than tabulated value at 95% confidence level
Z = 1.96, the null hypothesis is rejected and it is concluded that average amount of filling
machine is not 100 grams.

Question No. 10 Spring 2005

A firm believes that the mean cost to process a sales order is Rs. 132.50. The cost controller fears
that the average cost of processing is more than that. A random sample of 100 orders has a
sample mean of Rs. 133.5 and standard deviation of Rs. 5. Level of significance = 0.05

Required:

i) State the Null Hypothesis and Alternative Hypothesis.

ii) Which test distribution (Z or t) will be used and why?
iii) Whether the 'Ho' will be rejected or not; give reason for your conclusion.

Answer No. 10 Spring 2005

i) Null Hypothesis H0: µ < 132.5
Alternative Hypothesis H1 : µ > 132.5
�− µ
ii) Test statistics Z = � as n is more than 30
√�
Significant level = 0.05
. − .
Z= = 2.00
√

iii) The calculated value of (Z = 2) is greater than the tabulated value of (Z = 1.645) at
5% level of significance, the null hypothesis is rejected accepting the claim of cost
controller that mean cost of processing a sale order is greater than 132.5.

iv) p – value Probability of rejecting H0 when it is true

�− µ
Z= �.�
= where µ = 132.5, x = 133.5, σ = 5
. − .
Z= = 0.2
P(0 < Z < 0.2) = 0.0793
P(0.2 < Z < ∞) = 0.5 – P(0 < Z < 0.2) = 0.5 – 0.0793 = 0.4207
 Hence probability of rejecting H0 when it is true is 0.4207
Question No. 11 Autumn 2005
I.Q of students at a college are normally distributed with a mean = 112 and a standard deviation
= 12. Professor Hameed believes that IQ of students in his Statistics class do not have a mean IQ
of 112. To prove this point, a random sample of 5 students from Statistics class is chosen. Their
IQ Scores were found to be 115, 124, 131, 129 and 120. Test the hypothesis that the means IQ of
the students of the class is the same as that of the population of students at the college. Use
α = 0.05.

Answer No. 11 Autumn 2005

Total
X 115 124 131 129 120 619

∑�
X= = = 123.8

Null Hypothesis H0: µ = 112

Alternative Hypothesis H1: µ ≠ 112

The population standard deviation is given, test statistic is Z

�− µ
Test statistics = Z = � = α = 0.05
√�

. −
Z= = 2.20
√

Because the calculated value of Z = 2.2 is greater than the tabulated value of Z = 1.96 the null
hypothesis is rejected concluding that mean value is not 112 on the basis of mean values of
sample.

Question No. 12 Autumn 2004

The personnel manager of a firm is concerned about 'absenteeism'. He decided to sample the
records to determine if absenteeism is distributed evenly throughout the six days week. The
sample results are:

Weekday Monday Tuesday Wednesday Thursday Friday Saturday

Number Absent 12 9 11 10 9 9

a) What is the expected frequency for each day?

b) How many degrees of freedom are there?
 c) What is the chi-square value at 1 % level of significance?
d) Using the chi-square test of significance, compute � 2.
e) Is the null hypothesis rejected at 1% level of significance?

Answer No. 12 Autumn 2004

a) The absenteeism is evenly distributed hence the probability of absenteeism for
each day in 6 days is 1/6

The total frequency =12 + 9+11+10 + 9 + 9 = 60

The expected frequency = 60 x 1/6 =10

b) Degrees of freedom
There are 6 days so degrees of freedom is

c) n – 1 = 6 – 1 = 5
The value of x2α0.01 at 5 degrees of freedom is 16.812
d)

Observed Expected (o – e)2

Days o–e (o – e) 2
Value (0) Value (e) e
Monday 12 10 2 4 0.4
Tuesday 9 10 -1 1 0.1
Wednesday 11 10 1 1 0.1
Thursday 10 10 0 0 0
Friday 9 10 -1 1 0.1
Saturday 9 10 -1 1 0.1
8 0.8

−�
�2 = ∑ �

e) Since calculated value of �2 = 0.8 against tabulated value of �2 = 16.812 the null
hypostasis is accepted concluding that absenteeism is evenly distributed.

Question No. 13 Autumn 2003

i) Given the following data, test at 0.05 level whether there is any relationship between
the level of education and social activities of an individual?

Education Social Activity

 Level
Above Average Average Below Average
Graduation 20 10 10
Matriculation 30 50 80
Primary 10 60 130

ii) Interpret your result.

Answer No. 13 Autumn 2003

i) Null Hypothesis H0: The given attributes are independent
Alternative Hypothesis H1: The given attributes are associated
α = 0.05
Degrees of freedom = (c – 1)(r – 1) = 2 x 2 = 4

�−�
Test statistics = �2 = ∑ �

Rejection Region � 2 > � 2 α (c – 1) (r – 1)

� 2 > � 2 0.05 (4)
� 2 > 9.45

(Q – e)2
Q E Q–e (Q – e) 2
e
20 6 14 196 32.67
30 24 6 36 1.50
10 30 -20 400 13.33
10 12 -2 4 0.33
50 48 2 4 0.08
60 60 0 0 0.00
10 22 - 12 144 6.55
80 88 -8 64 0.73
130 110 20 400 3.64
58.83

Expected Values Table

Above Below
Average Average Average Total
A B C
D Graduation 6 12 22 40
E Matriculation 24 48 88 160
F Primary 30 60 110 200
Total 60 120 220 400
From the original table

(A)(D) 60 x 40
AD = = =6
Total 400

(A) (E) 60 x160

AE = = = 24
Total 400

(B)(D) 120x40
BD = = = 12
Total 400

(B)(E) 120x160
BE = = = 48
Total 400

Rest of the table will automatically be completed

Since �2 is greats than � α, the null hypothesis is rejected and concluded that education level has
association with social activities.

Question No. 14 Spring 2003

Investigating the success of its interviewers, a firm finds that 176 out of 225 interviews
attempted by trained interviewers are successfully completed of 310 interviews attempted by
untrained interviewers, only 188 are successfully completed. Determine whether these data
provide sufficient evidence at the 5% level of significance to indicate a relationship between the
training status of interviewers and the outcome of interview attempted.

Answer No. 14 Spring 2003

176
Given p1 = = 0.78
225
188
P2 = = 0.61
310

Null hypothesis H0: p1 = p2

Alternative hypothesis H1 : p1 ≠p2
α = 0.05

Rejection Region H0 |Sample Z | > 1.96

 176 + 188
pe = = 0.68
225 + 310

S(p1 – p2) =√ � � + = √ . . + = 0.041

− . − . .
Z=� = = = 4.15
− . .

Since |sample Z = 4.15| > 1.96 we reject null hypothesis and conclude that proportion of success
by trained interviewers is different from proportion of untrained interviewers.

Question No.15 Autumn 2001

To judge certain safety features of a car, an engineer must know whether reaction time of drivers
to a given emergency situation has a standard deviation of 0.010 second or whether it is greater
than 0.010 second. What can be concluded at 0.05 level of significance if he gets s = 0.014
second for a sample size of n = 15.

Answer No. 15 Autumn 2001

Null Hypothesis H0: σ2 < 0.0001
Alternative Hypothesis H1 : σ2 > 0.000196
Level of significance: α = 0.05
− s
Test statistics = �2 =
σ
Degrees of freedom = n – 1 =15 – 1 = 14

(15 – 1)(0.000196)
Observed value �2 = = 27.44
0.0001

Since observed value at 14 degrees of freedom = 27.44 is greater than tabulated value of 23.68,
we reject null hypothesis and conclude that there is a significant difference between the two
variances.

Question No. 16 Spring 2001

The test scores of students of Zone A and Zone B yielded the following results

Sample size Mean Score Sample Std Dev

Zone A 150 47.3 7.3
Zone B 100 45.4 6.9
If population scores are normal test the hypothesis that there is no difference between mean
scores at significance level of α = 0.05

Answer No 16 Spring 2001

n1 = 150, x1 = 47.3, S1=7.3
n2 = 100, x2 = 45.4, S2 = 6.9
Null Hypothesis H0: µ 1 = µ 2
Alternative Hypothesis H1 : µ 1 ≠ µ 2
Critical level α = 0.05
� −� � −� . .
Test statistics Z = = = = = 2.08
√ . . √ . + . .
√
� � +
+
� �

The tabulated value of Z at 5% rejection is 1.96 which is less than calculated value of Z = 2.08,
hence null hypothesis is rejected and is concluded that there is a significant difference between
two means.

Question No. 17 Autumn 2000

Assuming x to be normally distributed and given that x = 52, s = 7 and n = 20. Test the
hypothesis that σ = 9 at α = 0.05

Answer No. 17 Autumn 2000

Null Hypothesis H0: σ = 9
Alternative Hypothesis H1 : σ ≠ 9
Critical level α = 0.05
− �
Test statistics = �2 = �

�2 = 19(7)2/(9)2 = 11.49

Rejection Region �2 < �2 0.025 (19) and �2 > �2 0.975 (19)

�2 < 32.85 and �2 > 8.91

Conclusion: The value of x2 falls in the acceptance region so mull hypothesis is accepted i.e.
σ=9

Question No. 18 Autumn 1999

A drug company makes tables is considered out of control if the standard deviation of the tablet
weight exceeds 0.120 mg. A random sample of 15 tablets taken during a routine periodic check
produced a sample standard deviation of 0.190 mg. At the 0.5 levels, is the tablet production
process out of control?
Answer No. 18 Autumn 1999
Given σ = 0.0120 mg, n = 15, s = 0.0190 mg
Level of significant α = 0.05

At one sided 95% confidence upper bound interval

− � − .
�2 = = = 35.10
� .

Tabulated �2 with 14 degrees of freedom at 95% confidence level is 23.68. Hence it is concluded
that tablets production process is out of control.

Question No. 19 Section II September 1998

A random sample of size 400 is drawn from a normal population. The sample means and
standard deviation are 5.2 and 3.7 respectively.

i) Estimate standard error of mean.

ii) Test at 5% level of significance that population mean is not more than 5.

Answer No. 19 Section II September 1998

Given x = 5.2, S.D = 3.7, n = 400

SD SD
i) Standard error S = = = 0.185
√ √
ii) Null hypothesis H1 : µ > 5.0

�−µ . − .
Test statistics Z = � = . = = 1.08
.
√� √

Level of significance α = 5%

It is a one tailed test the tabulated value of Z at 5% is = 1.645.

Since the calculated value of Z is less then observed, the null hypothesis is accepted
and it is concluded that sample value is not more than 5.

Question No. 20 Section II April 1998

Define the terms.
i) Type I error
ii) Level of significance

Answer No. 20 Section II April 1998

 i) TYPE I ERROR
The committing of error by rejecting the null hypothesis when it is true on the basis of
given observations is called type I or a error.

ii) LEVEL OF SIGNIFICANCE

The probability of rejecting a null hypothesis when it is true is called level of
significance.

Question No. 21 Section II April 1998

A machine is set to fill cereal boxes with a mean weight of 500 grams of cereal per box. The
standard deviation is known to be 125 grams. A random sample of 25 filled boxes is taken and
the mean weight of cereal per box is computed as 540 grams. Is there reason to believe that
machine is overfilling the boxes, thereby increasing the costs, Used α = 0.05 (Z0.05 = 1.645)

Answer No. 21 Section II April 1998

Given µ = 500, x = 540
n = 25, S.D = 125

Null Hypothesis H0: µ = 500

Alternative Hypothesis H1 µ > 500
Level of significance Z005 = 1.645

x− µ −
Test statistics Z = σ = = 1.60
√n √

Since the calculated value of Z is less than the tabulated value of Z, the null hypothesis is
accepted. It is concluded that there is no significant difference and machine is not overfilling the
boxes.

Question No. 22 Section II 1997

A manufacturer claims that the average life of a certain type of a product is at least 150 hours. It
is known that the standard deviation of this type of product is 20 hours. A consumer wishes to
test the manufacturer's claim and accordingly tests 16 products. The sample mean is 144 hours.
Perform a test to reject the manufacturer's claim at 5% level of significance.
[Z0.05 = 1.645 and Z0.025 = 1.96]

Answer No. 22 Section II 1997

Given µ = 150, S.D = 20, x = 144, n = 16
Z0.05 = 1.645 and Z0.025 = 1.96
The population mean and standard deviation are known. Z test will be used even if sample size is
less than 30.

Null hypothesis H0 : µ > 150

Alternative hypothesis H1 : µ < 150
x− µ −
Test statistics Z = SD = = – 1.2
√n √

The absolute calculated value of Z is less than the tabulated value of Z, the null hypothesis is
accepted and it is concluded that there is no significant difference between the population mean
and sample mean. The claim of manufacturer is accepted.

Question No. 23 Section II April 1997

The yield of corn in kg./hectre is normally distributed with standard deviation of 120 kg. For a
random sample of six plots the yields are: 1430, 1190, 1280, 1270, 1310 and 1380 kg.
Test the hypothesis that mean yield is greater than 1350 kg at 5% level of significance. Given
(Z0.05 = 1.645 and t(5,0.05) = 2.015

Answer No. 23 Section II April 1997

Null Hypothesis H0: µ < 1350
Alternative Hypothesis H1 = µ > 1350
Level of significance α = 0.05 or 5%

x− µ − − ×√
Test statistics Z = SD = =Z= = – 0.816
√n √

The Calculated absolute value (0.816) is less than Z = 1.645 at 5% level of significance, we
accept the null hypothesis that which means yield is less than or equal to 1350 kgs.

Question No. 24 Section II November 1996

An employment exchange officer claims that the applications received in his office are processed
in an average of at most 5 working days.

A sample of 8 applications are randomly taken, and these were processed in 7, 6, 7, 9, 3, 3, 8 and
5 days.

Assuming that the processing time is normally distributed; test the validity of the officers claim
at 0.01 level of significance.

Given:
 Critical values of t
d.f One Tail Area
0.005 0.01
7 3.499 2.998
8 3.355 2.896
9 3.250 2.821

Answer No. 24 Section II November 1996

Given n = 8 and other data items are

x x2
7 49
6 36
7 49
9 81
3 9
3 9
8 64
5 25
48 322

∑x 48
x= = =6
n 8

∑x
s=√ −
[∑x − ]=√ [ − ]=√ =√ . = 2.2

Null hypothesis H0: µ < 5

Alternative Hypothesis H1: µ > 5

x− µ − √ .
Test statistics t = s = . = = = 1.3
. .
√n √

t against 7 degrees of freedom at α0.01 (one tailed test) = 2.998;

Since tabulated value of t is greater than calculated value we accept our null-hypothesis claiming
that the time to process an application is 5 days and the claim of officer is correct.

Question No. 25 Section II (Part 1) May 1996

The time a patient must wait before being attended by a physician in a hospital has been
observed for each of 100 randomly selected patients.

The sample mean is 70.3 minutes and the sample standard deviation is 28.2 minutes.

Test the hypothesis, at 1% level of significance, that the mean waiting time in the hospital is not
more than 65 minutes.

Answer No. 25 Section II (Part 1) May 1996

Given x = 70.3, S.D = 28.2, µ = 65, n = 100
The test statistic used will be Z

Null hypothesis H0: µ < 65

Alternative hypothesis H1 : µ > 65,

x− µ . − . ×√
Z= SD = . = = 1.87
. .
√n √

Level of significance = 0.01 Critical Region |Z| > α 0.01

It is one tailed test and the value of α at 1% level is 2.33

As the calculated value of Z is less then tabulated value the null hypothesis is accepted; that the
mean waiting time in the hospital is not more than 65 minutes.

Question No. 26 Section II May 1996

The mean length of time required to perform a certain task on assembly line has been established
at 15.5 minutes, with a standard deviation of 3 minutes. A random sample of 9 employees is
taught a new method. After the training period, the average time these 9 employees take to
perform the task is 13.5 minutes. Do these results provide sufficient evidence to indicate that the
new method is faster than old? Use α = 0.05. Assume that the time required to perform the task
are normally distributed.

Answer No. 26 Section II May 1996

Given µ = 15.5, S.D = 3, Sample Mean x = 13.5, n = 9
The question relates to test of hypothesis
Null Hypothesis H0: µ > 15.5
Alternative Hypothesis H1 : µ < 15.5

Significance level α = 0.05

 x− µ . − .
Test statistics Z = SD = = – 2.00
√n √

The value of |Z|at 5% level is 1.645. The absolute calculated value of Z is greater than 1.645, we
reject H0 and accept H1 and conclude that the new method is faster than older.

Question No. 27 Section II May 1996

Define:
i) Type I error
ii) Type II error.

Answer No. 27 Section II May 1996

i) When testing the hypothesis the rejection of null hypothesis if it is true is called a
error or type 1 error.
ii) The acceptance of null hypothesis when it is false or rejection of alternative
hypothesis when it is true is (3 error or type II error.

Question No. 28 Section II (Part 1) November 1995

A random sample of 49 observation produced the following sums:

∑x = 20.7
∑x2 =10.9

Test the null hypothesis µ > 0.47 at 1% level of significance.

Answer No. 28 Section II (Part 1) November 1995

Null Hypothesis H0 : µ > 0.47
Alternative Hypothesis H1 : µ < 0.47
Level of significance = 0.01
x− µ
Test of statistics Z = S
√n
Critical region Z < α 0.01

Given ∑x = 20.7
This is a test with known mean and unknown variance

∑� .
x= = = 0.42
∑� ∑� . .
S.D = √ − =√ − = √ . − . = 0.21
 . − .
Z=
. /√

The tabulated value at α 0.01 = 2.33

The tabulated value of (Z = 2.33) is greater than the absolute calculated value i.e 1.67, we accept
our null hypothesis and conclude that µ > 0.47

Question No. 29 Section II (Part 1) April 1995

A machine is set to produce nails with a mean length of 5 cm. A random sample of 150 nails
gives x = 5.1 cm and S = 0.2 cm.
[Given Z0.005 = 2.576]

Answer No. 29 Section II (Part 1) April 1995

Given µ = 5 cm, x = 5.1 cm, S = 0.2cm. N = 50, α = 0.05
Null Hypothesis H0: µ = 5

x− µ . −
Test statistics Z = S = . = 3.535
√n √
Critical Region (–Z < 2.575 <+ Z)

The calculated value is more than tabulated value of Z, so H0 is rejected and concluded that
machine is not working properly.

Question No. 30 Section II April 1995

A manufacturer of bolts claims that the means length is 4.500 inches with a standard deviation
of 0.020 inches. A random sample of 16 bolts yields a mean of 4.512 inches. Do these data
provide sufficient evidence to indicate that the true mean length is greater than manufacturer's
claim? Assume that the dimensions are normally distributed. Use a 0.01 level of significance:

Answer No. 30 Section II April 1995

This question relates to test of hypothesis

Where µ = 4.5, x = 4.512, α = 0.020, n = 16

Null hypothesis H0 : µ < 4.5
Alterative Hypothesis H1: µ > 4.5
Level of significance α = 0.01 or 1%
Critical region = Z > α 0.01
 x− µ . − . .
Test statistics Z = σ = . = x 4 = 2.4
.
√n √

The tabulated value of Z at 0.01 level 2.575 which is greater than the calculated value of Z. We
accept our null hypothesis that the mean length of bolts is equal to 4.5.