UNIT-IV

CONCEPT OF ESTIMATION AND TESTING OF

HYPOTHESIS
Introduction:
Basic Definitions:
Population: The set of well defined objects is called a
Population
Example: Set of Natural numbers, whole numbers
Population Size: The no of items or observations of
the Population is called “Population Size” and
It is denoted by” N”
Sample: A finite subset of a population OR A part of
a Population
Sample Size: The no of item or observations of the
sample is called ” Sample Size” and
It is denoted by” n”
According to the sample size sample are TWO types
and they are
1. Small Sample: If the sample size is Less Than
30(n<30) then it is called as Small Sample
2. Large Sample: If the sample size is Greater Than 30
OR Equals To 30( n  30 ) then it is called as Large
Sample
Notations and Symbols
Population Sample
Size N n
Mean  x
Variance 2 S2
Standard  S
deviation(SD)
Proportion P p
P+Q=1 p+q=1
Central Limit Theorem: If x is a mean of random
sample of size n taken from a population having the
mean  and variance  2 , then

x 
Z

n

Where Z is a Normal variate

Standard Error:
The Standard Deviation is the Standard Error of the
distribution

ie, SE  
n
Where SE is Standard Error
Estimator:
The method or rule to determine a population
parameter is called an estimator
Estimations are two type, they are
1.Point Estimation
2. Interval Estimation
 1.Point Estimation: If the population parameter is
being at an exact point that estimation is called
“Point Estimation”
2.Interval Estimation: If the population parameter is
being in an interval then the estimation is called
“Interval Estimation”
Maximum Error of Estimation for Large Sample
( n  30 )
Maximum Error is denoted by ”E”, and it is given
by
 S
E  Z /2 OR E  Z /2
n n
Where  standard deviation of the population
S standard deviation of the sample
n Sample size
Z α/2= Z table value at α/2 level of
significant
Maximum Error of Estimation for Small Sample
(n< 30)
Maximum Error is denoted by ”E”, and it is given
s
by E  t /2
n

Where S standard deviation of the sample

n sample size
t α/2= t table value at α/2 level of significant and
corresponding with (n-1) degrees of freedom(dof)
Confidence Interval for large sample( n  30 )
Confidence interval for large sample with (1-α) level
of significant given by
CI   x  E 

Where E is Maximum Error of Large Sample

 S
E  Z /2 OR E  Z /2
n n
Confidence Interval for small sample(n< 30)
Confidence interval for small sample with (1-α) level
of significant given by
CI   x  E 

Where E- Maximum Error of Small Sample

s
E  t /2
n
 TESTING OF HYPOTHESIS
Hypothesis: A statistical hypothesis is a hypothesis;
Hypothesis is a statement about the parameters of one
or more populations
Testing of Hypothesis:Testing of Hypothesis is a
process for deciding whether to Accept OR Reject the
Hypothesis.
There are Two Types of Hypothesis: they are
1.Null Hypothesis
2.Alternative Hypothesis
1.Null Hypothesis: A Null hypothesis is a
Hypothesis which is tested for possible rejection
under the assumption, that it is TRUE.
Null Hypothesis is denoted by “HO”
2.Alternative Hypothesis : Any hypothesis which is
complimentary to the Null Hypothesis called
“Alternative Hypothesis”
Alternative Hypothesis is denoted by “H1”
Critical Region: A Region in the sample “S” which
amount to rejection of “HO” is called Critical Region
(CR) OR Rejection Region (RR)
In any test the critical region is represented by area
under the probability curve.

Critical Value (CV): The value of test statistic

which separate CRITICAL REGION AND
ACCEPTENCE REGION is called “Critical Value
” or “ Significant Value”
Significant value depended on
1.Level Of Significant (LOS)
2.Alternative Hypothesis(H1)
Alternative Hypothesis (H1) is of three type
1.Right Tailed Test(RTT); H1 :   0
2.Left Tailed Test(LTT); H1 :   0
3.Two Tailed Test (TTT); H1 :   0
Right Tailed Test(RTT)
If the critical region falls right hand side of a
probability curve is known as” Right Tailed Test
(RTT)

Left Tailed Test(LTT)

If the critical region falls Left hand side of a
probability curve is known as” Left Tailed Test (LTT)
Two Tailed Test(TTT)
If the critical region is equally distributed on both
sides of a probability curve is known as” Two Tailed
Test (TTT)”
Type-I and Type –II Errors in Testing of
Hypothesis
There are two types of errors in testing of hypothesis
they are Type-I and Type-II Errors
The Type-I and Type-II Errors are clearly explained
in the following table
Accept HO Reject HO
Null Correct Type-I Error
Hypothesis is decision
True (HO is
True)
Null Type-II Error Correct
Hypothesis is decision
False (HO is
False)

Type-I Error: The Hypothesis is TRUE but the test

is Rejected. In this situation we say that Type-I Error
has been made.
P [Reject HO when HO TRUE] = P [Type-I Error] = 𝛼
Type-II Error: The Hypothesis is FALSE but the
test is Accepted. In this situation we say that Type-II
Error has been made.
P[Accepted HO when HO FALSE]=P[Type-II
Error]=𝛽
Where α, β are the size of Type-I and Type-II Errors
respectively
Working rule for Testing of Hypothesis
This is a five step procedure
Step 1: Null Hypothesis(H0): setup the Null
Hypothesis Ho base on the given
Let it be H0: µ= µo for single mean, where µo is a
specified mean
Let it be H0: µ1= µ2 for difference between two means
Step 2 : Alternative Hypothesis (H1): set up
Alternative Hypothesis H1, so that we could decide
whether we should use One Tailed Test(RTT or LTT)
OR Two Tailed Test(TTT)
Alternative Hypothesis may be one of the following
three conditions
1.Right Tailed Test(RTT); H1 :   0
2.Left Tailed Test(LTT); H1 :   0
3.Two Tailed Test (TTT); H1 :   0
Step 3: Level of Significant(LOS): Set up
appropriate Level of significant depends on reliability
of the estimates and permissible risk
ie, Z tab= Z table value at  % and Type of test (RTT or
LTTor TTT)
Step 4 : Test Statistic: Compute Test Statistic under
Ho
x  E (Z )
Zcal 
SE ( Z )

Where x is sample mean, E(Z) is Population mean,

SE(Z) is Standard deviation of sample
Step 5: conclusion: if Ztab  Zcal , then Accept Null
Hypothesis Ho
Otherwise Reject Null Hypothesis.
Some important table values of Z or Standard
Normal Distribution
% 1% 5% 10%
RTT/LTT 2.33 1.645 1.28
TTT 2.58 1.96 1.645
Working rule for Testing of Hypothesis for single
mean in Large sample
Step 1: Null Hypothesis(H0): setup the Null
Hypothesis Ho base on the given
H0: µ= µo for single mean, where µo is a specified
mean
Step 2 : Alternative Hypothesis (H1): set up
Alternative Hypothesis H1, so that we could decide
whether we should use One Tailed Test(RTT or LTT)
OR Two Tailed Test(TTT)
Alternative Hypothesis may be one of the following 3
conditions
1.Right Tailed Test(RTT); H1 :   0
2.Left Tailed Test(LTT); H1 :   0
3.Two Tailed Test (TTT); H1 :   0
Step 3: Level of Significant(LOS): Set up
appropriate Level of significant depends on
reliability of the estimates and permissible risk
ie Z tab= Z table value at  % and Type of test (RTT or
LTTor TTT)
Step 4 : Test Statistic: Compute Test Statistic under
Ho

x 
Zcal 

n if σ is known

x 
Zcal 
S
n if σ is unknown
Where x is sample mean,  is Population mean, is
Standard deviation of Population, S is SD
Step 5: conclusion: if Ztab  Zcal , then Accept Null
Hypothesis Ho
Otherwise Reject Null Hypothesis.
Problem NO.1: An ambulance services claims that it
takes on an average less than 10 min to reach its
destination in emergency calls. A sample of 36 calls
has a mean of 11 min and variance 16 min to reach
the destination in emergency calls. Test the claim at
95% level of significant?
Solution:
Given that
Sample size n=36 (large sample n>30)
Sample mean x =11
Population variance  2  16,   4
Procedure:
STEP 1: Null Hypothesis(HO): Ambulance services
claim is FALSE
ie , HO: µ=10
STEP 2: Alternative Hypothesis (H1): Ambulance
services claim is TRUE
ie , HO: µ < 10 (LTT)
STEP 3: Level of Significant(LOS):
Z tab= Z table value at 5%and LTT (Left Tailed Test)
Z tab=1.645
STEP 4: Test Statistic: Under the Null Hypothesis
Ho

x  11  10
Zcal   =1.5
 4
n 36
Step 5 :Conlusion: Z tab(1.645) > Z cal(1.5)
Accept HO
ie Reject H1
Ambulance services claim is FALSE
Problem NO.2: A random sample of 64 students
have a mean weight of 70kgs.Can this be regarded as
a sample drawn from the Population with mean
weight 56 kgs and standard deviation(SD) 25 kgs at
99% level of significant?
Solution:
Given that
Sample size n=64 (large sample n>30)
Sample mean x =70 kgs
Population mean µ=56 kgs
Population SD   25
 %  1%
Procedure:
STEP 1: Null Hypothesis (HO): The sample is drawn
from the given Population
ie , HO: µ=56
STEP 2: Alternative Hypothesis (H1): The sample is
NOT drawn from the given Population
ie , H1: µ ≠ 56 (TTT)
STEP 3: Level of Significant(LOS):
Z tab= Z table value at 1%and TTT(Two Tailed Test)
Z tab=2.58
STEP 4: Test Statistic: Under the Null Hypothesis
Ho

x  70  56
Zcal   = 4.48
 25
n 64
Step 5 :Conlusion: Z tab(2.58) < Z cal(4.48)
Reject HO
ie Accept H1 ,
The sample is NOT drawn from the given Population
Problem NO.3: A sample of 400 items is taken from
a Population whose SD is 10.The mean of sample 40.
Test whether the sample has came from a Population
with mean 38. Also calculate 95% Maximum Error
and Confidence Interval for the Population?
Solution: Given that
Sample size n=400 (large sample n>30)
Sample mean x = 40
Population mean µ =38
Population SD σ = 10
 %  5%
Procedure:
STEP 1: Null Hypothesis (HO):: The sample is
drawn from the given Population
ie , HO: µ=38
STEP 2: Alternative Hypothesis (H1):: The sample
is NOT drawn from the given Population
ie , H1: µ ≠ 38 (TTT)
STEP 3: Level of Significant(LOS):
Z tab= Z table value at 5%and TTT(Two Tailed Test)
Z tab=1.96
STEP 4: Test Statistic: Under the Null Hypothesis
Ho

x  40  38
Zcal   =4
 10
n 400
Step 5 :Conlusion: Z tab(1.96) < Z cal(4)
Reject HO
ie Accept H1 , The sample is NOT drawn
from the given Population
Maximum Error:
The given sample is n=400 (large sample)
The Maximum Error for Large sample is

E  Z /2
n
 10 
E   (1.96)
 400 

E  0.98
Confidence Interval:
The given sample is n=400 (large sample)
The confidence interval for Large sample is
CI   x  E 
CI   40  0.98

CI   40  0.98, 40  0.98

CI=[39.02.40.98]
Working rule for Testing of Hypothesis for
Equality of Two Means (OR) Difference between
Two Means in Large Sample
Let x1 and x2 be the sample means of two
independent large random samples of sizes n1 and n2
drawn from two populations having means µ1 and µ2
and Standard Deviations 1 and  2 respectively
Step 1: Null Hypothesis(H0):
Ho: µ1= µ2 OR Ho: µ1- µ2=0
ie The two samples are drawn from same Population
OR
ie There is NO difference between two Population
means
Step 2 : Alternative Hypothesis (H1): Alternative
Hypothesis may be one of the following 3 conditions
1.Right Tailed Test(RTT); H1 :   2
1

2.Left Tailed Test(LTT); H1 :   2

1

3.Two Tailed Test (TTT); H1 :   21

Step 3: Level of Significant(LOS):
Z tab= Z table value at  % and Type of test (RTT or
LTT or TTT)
Step 4 : Test Statistic: compute Test Statistic under
Ho

Zcal 
 x  x      
1 2 1 2

 12  22

n1 n2

Where µ1, µ2 are Population means

σ1, σ2 are Standard deviations of Population
x1 and x2 sample means

Step 5: conclusion: if Ztab  Zcal then Accept Null

Hypothesis Ho
Otherwise Reject Null Hypothesis.

Problem No:1 Average marks scored by 32 Boys

students is72 with a SD of 8 ,while that for 36 Girls
average marks are 70 with SD of 6. Can this indicate
that Boys performance is better than Girls at 95%
Level of Significance?
Solution : Given that
Boys Girls
Sample size 32(n1) 36(n2)
Sample mean 72( x1 ) 70( x2 )
Variance 64( 12 ) 36( 2 2 )
SD 8( 1 ) 6( 2 )
Population µ1 µ2
Given that  %  5%
Procedure:
Step 1: Null Hypothesis(H0):
Ho: µ1= µ2 OR Ho: µ1- µ2=0
ie Boys performance is equal to Girls performance
OR
Boys and Girls performances are equal
Step 2 : Alternative Hypothesis (H1): H1: µ1>µ2;
Right Tailed Test(RTT);
ie Boys performance is better than Girls
Step 3: Level of Significant(LOS):
Z tab= Z table value at 5% and Right Tailed
Test (RTT)
Z tab=1.645
Step 4 :Test Statistic: Compute Test Statistic under
Ho

Zcal 
 x  x      
1 2 1 2
Z 
 72  70    0 
 12  22
cal
64 36
 
n1 n2 32 36

Zcal  1.154

Step 5: Conclusion: Z tab (1.645) > Z cal (1.154),

Accept Null Hypothesis Ho
Reject Alternative Hypothesis H1.
ie Boys performance is NOT better than Girls
Problem No:2 Two types of new Cars produced in
USA are tested for petrol consumption. One sample is
consisting of 42 Cars with average mileage 15 kmpl,
while the other sample of 80 Cars with average
mileage 11.5kmpl with Population variances σ12=2
and σ22=1.5. Test whether there is any significance
difference between two types of Cars according to
petrol consumption?
Solution :Given that
Type I car Type II car
Sample size 42(n1) 80(n2)
Sample mean 15( x1 ) 11.5( x2 )
Variance 2 ( 12 ) 1.5( 2 2 )
Population µ1 µ2
Assume α% = 5%

Procedure:
Step 1: Null Hypothesis(H0):
Ho: µ1= µ2 OR Ho: µ1- µ2=0
ie There is NO difference between two types of Cars
according to petrol consumption
Step 2 : Alternative Hypothesis (H1):
H1: µ1≠µ2; Two Tailed Test (TTT);
ie There is a difference between two types of Cars
according to petrol consumption
Step 3: Level of Significant(LOS):
Z tab= Z table value at 5% and Two Tailed Test(TTT)
Z tab=1.96
Step 4 : Test Statistic: Compute Test Statistic under
Ho

Zcal 
 x  x      
1 2 1
Z 2

15  11.5   0 
Zcal  13.585
 
2 2 cal
2 1.5
1
 2 
n1 n2 42 80

Step 5: Conclusion: Z tab (1.96)< Z cal (13.585),

Reject Null Hypothesis Ho, Accept Alternative
Hypothesis H1.
ie There is a difference between two types of Cars
according to petrol consumption
Problem No:3 The means of two large sample of size
1000 and 2000 are 67.5 and 68 respectively .Can the
samples be regarded as drawn from the same
Population of SD 2.5 at 95% LOS?

Solution :Given that

I St sample IInd
sample
Sample size 1000(n1) 2000(n2)
Sample 67.5( x1 ) 68( x2 )
mean
SD 2.5(both sample sd’s are same) 2.5
Variance 6.25( 12 ) 6.25( 2 2 )
Population µ1 µ2
Assume α% = 5%
Procedure:
Step 1:Null Hypothesis(H0):
Ho: µ1= µ2 OR Ho: µ1- µ2=0
ie The two sample are drawn from same Population
Step 2 : Alternative Hypothesis (H1):
H1: µ1≠µ2; Two Tailed Test(TTT);
ie The two sample are NOT drawn from same
Population
Step 3: Level of Significant(LOS):
Z tab= Z table value at 5% and Two Tailed Test
(TTT)
Z tab=1.96
Step 4 : Test Statistic: Compute Test Statistic under
Ho

Zcal 
 x  x      
1 2 1
Z 2

 67.5  68    0 
Zcal  5.163
 
2 2 cal
6.25 6.25
1
 2 
n1 n2 1000 2000

Step 5: Conclusion: Z tab (1.96) < Z cal (5.163)

Reject Null Hypothesis Ho,
Accept Alternative Hypothesis H1.
ie The two sample are NOT drawn from same
Population
Problem No:4 A company claims that it bulbs are
superior those of its main competitor. A study showed
that a sample of 40 bulbs have a mean life of 647 hrs
of continuous use with a SD of 27 hrs ,while a sample
40 bubs made by its main competitor had a mean life
of 638 hrs of continuous use with a SD of 31hrs. Test
the claim at 5% Level of Significance? Also find
Maximum Error and Confidence Interval at 5% los?

Solution : Given that

 I St IInd
sample(claimer) sample(competitor))
Sample size 40(n1) 40(n2)
Sample 647( x1 ) 638( x2 )
mean
SD 27( σ1) 31(σ2)
Variance 729 ( 12 ) 961 ( 2 2 )
Population µ1 µ2
Assume α% = 5%
Procedure:
Step 1: Null Hypothesis(H0):
Ho: µ1= µ2 OR Ho: µ1- µ2=0
ie Two types of bulbs are having same length of life
Step 2 : Alternative Hypothesis (H1):
H1: µ1>µ2; Right Tailed Test (RTT);
ie The company bulbs are superior than the
competitor regarding length of life
Step 3: Level of Significant(LOS):
Z tab= Z table value at 5% and Right Tailed
Test (RTT)
 Z tab=1.645
Step 4 : Test Statistic: Compute Test Statistic under
Ho

Zcal 
 x  x      
1 2 1
Z 2

 647  638    0 
 
2 2 cal
729 961
1
 2 
n1 n2 40 40

Zcal  1.38

Step 5: Conclusion: Z tab (1.645) > Z cal (1.38)

Accept Null Hypothesis Ho,
Reject Alternative Hypothesis H1.
ie The company bulbs are NOT superior than the
competitor regarding length of life
Maximum Error:
The given sample is n1=40 and n2=40 (large sample)
The Maximum Error for Large sample is
  12   22
E  Z /2
n1  n2

729  961
E 1.96
40  40
E  4.5961

Confidence Interval:
The confidence interval for Large sample is
CI   ( x1  x2 )  E 

CI   (647  638)  4.5961

CI  9  4.5961
CI   4.4039,13.5961
 PROPORTIONS
Sample Proportion: A large random sample of size n,
out of n sample observations x observations are
having a certain Attribute (ie, character or quality)
The Sample Proportion of a certain Attribute is
denoted by p and
x
it is given by p=
n
Population Proportion is denoted by P
And P+Q=1
Maximum Error for Single Proportion
Maximum Error for single Proportion with (1-α) level
PQ
of significant given by E  Z /2
n

Confidence Interval for Single Proportion

confidence interval for large sample with (1-α) level
of significant given by CI   P  E 
PQ
where E  Z /2
n
Maximum Error for difference between two
Proportion
Maximum Error for difference between two
Proportion with (1-α) level of significant given by
1 1
E  Z /2 PQ   
n n 
 1 2 

Confidence Interval for difference between two

Proportion
Confidence Interval for difference between two
Proportion with (1-α) level of significant given by

 
CI   p1  p2  E 

1 1
Where E  Z /2 PQ   
n n 
 1 2 

n1 p1  n2 p2 x1  x2
P OR P 
n1  n2 n1  n2

P+Q=1
Working Rule of Testing of Hypothesis for Single
Proportion:
Step 1: Null Hypothesis(H0): Ho: P=P0
ie The Population has a specified Proportion P0
Step 2 : Alternative Hypothesis (H1):
Alternative Hypothesis may be one of the following 3
conditions
1.Right Tailed Test(RTT);H1: P > P0
2.Left Tailed Test(LTT); H1: P < P0
3.Two Tailed Test (TTT); H1: P≠ P0
Step 3: Level of Significant(LOS):
Z tab= Z table value at α% and Type of test (RTT or
LTTor TTT)
Step 4 : Test Statistic: Compute Test Statistic under
Ho

pP
Zcal 
PQ
n
Where P is Population Proportion, P+Q=1;Q=1-P
p is sample proportion
n is sample size
Step 5: Conclusion: If Ztab  Zcal , then Accept Null
Hypothesis Ho
Otherwise Reject Null Hypothesis.
Problem NO.1: An experience had shown that 20%
of a manufactured product is of top quality. In one
day production of 400 articles, only 50 are top quality.
Test the hypothesis at 95% Level of Significance?
Also find Maximum Error and Confidence Interval at
5% los?
Solution :
Given that
Population proportion P = 20%=20/100=0.2
P+Q=1
Q=1-P=1-0.2=0.8
P=0.2,Q=0.8
n=400,x=50
sample proportion p=x/n=50/400=0.125
Procedure:
Step 1: Null Hypothesis(H0):
Ho: P=20% OR 0.2
ie 20% of a manufactured product is of top quality.
Step 2 : Alternative Hypothesis (H1):
P≠ 20% OR P=0.2 Two Tailed Test (TTT);
ie 20% of a manufactured product is NOT of top
quality.
Step 3: Level of Significant(LOS):
Ztab= Z table value at 5% and Two Tailed Test (TTT)
Step 4 : Test Statistic: Compute Test Statistic under
Ho

pP 0.125  0.20

Zcal  Zcal 
PQ (0.2)(0.8)
n 400
Zcal  3.75

Step 5: Conclusion: Ztab(1.96) <Zcal(3.75), then

Reject Null Hypothesis HO.
Accept Alternative Hypothesis H1
 20% of a manufactured product is NOT of top
quality
Maximum Error for Single Proportion
Maximum Error for single Proportion with (1-α) level
of significant given by
PQ
E  Z /2
n
(0.2)(0.8)
E  1.96
400
E  0.0392

Confidence Interval for Single Proportion

confidence interval for large sample with (1-α) level
of significant given by

CI   P  E  where E  Z /2 PQ =0.0392
n

CI   0.2  0.0392
CI   0.2  0.0392,0.2  0.0392
CI   0.1608,0.2392
Problem NO.2: 40 people were attacked by a disease
and only 36 survived. Will you reject the hypothesis
that the survival rate if attacked by this disease 85% in
favour of the Alternative Hypothesis that is more.
Test the hypothesis at 95% Level of Significance?
Solution: Given that
Population Proportion P = 85%=85/100=0.85
P+Q=1, Q=1-P=1-0.85=0.15
P=0.85, Q=0.15
n=40, x=36
Sample Proportion p=x/n=36/40=0.9
Procedure:
Step 1: Null Hypothesis(H0): Ho: P=85% OR 0.85
ie the survival rate if attacked by this disease 85%
Step 2 : Alternative Hypothesis (H1): P> 85% OR
P>0.85 Right Tailed Test (RTT);ie the survival rate if
attacked by this disease is more than 85%.
Step 3: Level of Significant(LOS):
 Ztab= Z table value at 5% and Right Tailed Test
(RTT)
Ztab=1.645
Step 4 : Test Statistic: Compute Test Statistic under
Ho

pP
Zcal 
PQ
n

0.9  0.85
Zcal 
(0.85)(0.15)
40
Zcal  0.885

Step 5: Conclusion: Z tab 1.645 >Z cal (0.885), then

Accept Null Hypothesis HO.
Reject Alternative Hypothesis H1
 The survival rate if attacked by this disease is NOT
more than 85%.
Problem NO.3: In a hospital 480 Female and 520
Male babies were born in a week. Do these figures
confirm the hypothesis that Male babies and Females
babies are born in equal number?
Solution:
Let p is a sample proportion of Female babies
Total sample babies n=480+520=1000
X= no of Female babies=480
Sample proportion p=x/n=480/1000=0.48
In general Male babies and Female babies are equally
distributed
Therefore Population proportion of Female babies is
P=50%=0.5
P+Q=1
Q=1-P=1-0.5=0.5
n=1000
P=0.5, Q=0.5
p=0.48, assume α%=5%
Procedure:
Step 1: Null Hypothesis(H0): Ho: P=50% OR 0.5
ie Male babies and Females babies are born in equal
number
Step 2 : Alternative Hypothesis (H1): P≠ 50% OR
P≠0.5 Two Tailed Test (TTT);
ie Male babies and Females babies NOT are born in
equal number
Step 3: Level of Significant(LOS):
Z tab= Z table value at 5% and Two Tailed Test
(TTT)
Z tab=1.96
Step 4 : Test Statistic: Compute Test Statistic under
Ho

pP
Zcal 
PQ
n
 0.48  0.5
Zcal 
(0.5)(0.5)
1000
Zcal  1.26

Step 5: Conclusion: Z tab (1.93) >Z cal (1.26), then

Accept Null Hypothesis HO.
Reject Alternative Hypothesis H1
 Male babies and Females babies are born in equal
number
Test for Equality of Two Proportions
OR
Working Rule for Testing of Hypothesis for
Defference between Two Proportions:
Let p1 and p2 be the proportions of two large samples
of sizes n1 and n2 drawn from two populations having
the population proportions P1 and P2 respectively
NOTE: if two samples are drawn from the same
population, the population proportions are equal.
ie P1 = P2
Step 1: Null Hypothesis(H0):
Ho: P1 = P2 OR Ho: P1 - P2=0
ie the two samples are drawn from same Population
OR
ie there is NO difference between two Population
Proportions
Step 2 : Alternative Hypothesis (H1): Alternative
Hypothesis may be one of the following 3 conditions
1.Right Tailed Test(RTT);H1: P1 > P2
2.Left Tailed Test(LTT); H1: P1 < P2
3.Two Tailed Test (TTT); H1: P1 ≠ P2
Step 3: Level of Significant(LOS):
Ztab= Z table value at α% and Type of test (RTT or
LTTor TTT)
Step 4 : Test Statistic: Compute Test Statistic under
Ho

( p1  p2 )  ( P1  P2 )
Zcal 
1 1
PQ(  )
n1 n2

Where is sample mean

P1 and P2 are Population Proportions
And p1 and p2 are Sample Proportions
n1 and n2 are sample sizes
 n1 p1  n2 p2 x x
P or 1 2
n1  n2 n1  n2

Step 5: Conclusion: if Ztab  Zcal ,

then Accept Null Hypothesis Ho

Otherwise Reject Null Hypothesis Ho.
Problem No:1 Random samples of 400 men and 600
women were asked whether they would like to have a
flyover near their residency . 200 men and 325
women were in favour of the proposal. Test the
Hypothesis that Proportion of men and women in
favour of the proposal are same at 5% Level of
Significance? Also find Maximum Error and
Confidence Interval at 5% level?

Solution :

Given that
Men sample size n1=400
Favourable men to the proposal
x1=200
Men proportion of the proposal
p1=x1/n1=200/400=0.5
Women sample size n2=600
Favourable women to the proposal x2=325
Women Proportion of the proposal
p2=x2/n2=325/600=0.541
Pooled Population Proportion
x x 200  325
P 1 2

n1  n2 400  600

P=0.525
P+Q=1
Q=1-P=1-0.525=0.475
Procedure:
Step 1: Null Hypothesis(H0):
Ho: P1= P2 OR Ho: P1- P2=0
ie Proportions of men and women in favour of the
proposal are same
Step 2 : Alternative Hypothesis (H1): H1: P1≠ P2
Two Tailed Test(TTT);
ie Proportions of men and women in favour of the
proposal are NOT same
Step 3: Level of Significant(LOS):
Ztab= Z table value at 5% and Two Tailed Test(TTT)
Z tab=1.96
Step 4 : Test Statistic: Compute Test Statistic under
Ho

( p1  p2 )  ( P1  P2 )
Zcal 
1 1
PQ(  )
n1 n2

(0.5  0.541)  (0)

Zcal 
1 1
(0.525)(0.475)(  )
400 600
Zcal  1.24

Step 5: Conclusion: Z tab (1.96) > Z cal (1.24),

Accept Null Hypothesis Ho
Reject Alternative Hypothesis H1.
Ie  Proportions of men and women in favour of the
proposal are same
Maximum Error for difference between two
Proportion
Maximum Error for difference between two
Proportion with (1-α)=5% level of significant given
1 1
by E  Z /2 PQ   
n n 
 1 2 

 1 1 
E  1.96 (0.525)(0.475)   
 400 600 
E  0.0281

Confidence Interval for difference between two

Proportion
Confidence Interval for difference between two
Proportion with (1-α)=5% level of significant given
 
by CI   p1  p2  E 

CI   0.5  0.541  0.0281

CI   0.041  0.0281
CI   0.041  0.0281,  0.041  0.0281
CI   0.0691, .0129
Problem No2: A company wanted to introduce a
new plan of work and a survey was conducted for this
purpose .one sample of 500 workers in one group
62% favoured the new plan of work and the other
group of 900 workers, 41% were against the new plan
of work . Is there any significance difference between
the two groups in their attitude towards the new plan
at 5% Level of Significance?
Solution: Given that
First group sample size n1=500
First group favoured proportion p1=62%=0.62
Second group sample size n2=900
Second group favoured proportion p2=1-against
proportion=1-0.41=0.59
Population Proportion
n1 p1  n2 p2 500(0.62)  900(0.59)
P 
n1  n2 500  900

P=0.606 P+Q=1, Q=1-P=1-0.606=0.394

Procedure:
Step 1: Null Hypothesis(H0):
Ho: P1= P2 OR Ho: P1- P2=0
ie there NO significance difference between the these
two groups in their attitude towards the new plan
Step 2 : Alternative Hypothesis (H1): H1: P1≠ P2
Two Tailed Test(TTT);
ie there a significance difference between the these
two groups in their attitude towards the new plan
Step 3: Level of Significant(LOS):
Ztab= Z table value at 5% and Two Tailed Test(TTT)
Ztab=1.96
Step 4 : Test Statistic: Compute Test Statistic under
Ho

( p1  p2 )  ( P1  P2 ) (0.62  0.59)  (0)

Zcal  Zcal 
1 1 1 1
PQ(  ) (0.606)(0.394)(  )
n1 n2 500 900
Zcal  0.915

Step 5: Conclusion: Z tab (1.96) > Z cal (0.915),

Accept Null Hypothesis Ho
Reject Alternative Hypothesis H1.
ie  There NO significance difference between the
these two groups in their attitude towards the new
plan
Problem No 3: In an investigation on Machine
performance the following results are obtained

No of Units No of Defectives
Machine-I 375 17
Machine-II 450 22
Test whether there any significance difference
between performances of two Machines at 5% LOS?

Solution:
Given that
First Machine sample size n1=375
First Machine no of defectives x1=17
First Machine Proportion of defectives
p1=x1/n1=17/375=0.0453
Second Machine sample size n2=450
Second Machine no of defectives x2=22
Second Machine Proportion of defectives
p2=x2/n2=22/450=0.0488
Population Proportion
n1 p1  n2 p2 375(0.0453)  450(0.0488)
P 
n1  n2 375  450

P=0.0473
P+Q=1
Q=1-P=1-0.0473=0.9527
Procedure:
Step 1: Null Hypothesis(H0):
Ho: P1= P2 OR Ho: P1- P2=0
ie there NO significance difference between
performances of two Machines
Step 2 : Alternative Hypothesis (H1):
H1: P1≠ P2 Two Tailed Test(TTT);
ie there a significance difference between
performances of two Machines
Step 3: Level of Significant(LOS):
Z tab= Z table value at 5% and Two Tailed Test(TTT)
Z tab=1.96
Step 4 : Test Statistic: Compute Test Statistic under
Ho

( p1  p2 )  ( P1  P2 )
Zcal 
1 1
PQ(  )
n1 n2

(0.0453  0.0488)  (0)

Zcal 
1 1
(0.0473)(0.9527)(  )
375 450
Zcal  0.2430

Step 5: Conclusion: Z tab (1.96) > Z cal (0.2430),

 Accept Null Hypothesis Ho
Reject Alternative Hypothesis H1.
ie  There NO significance difference between
performances of two Machines
Problem No 4: Among the items produced by a
factory out of 500, 15 were defective. In another
sample out of 400, 20 were defective .Test whether
there is any significance difference between two
Population Proportions at 1% level of significance?
Solution:

Given that
First factory sample size n1=500
First factory no of defectives x1=15
First factory Proportion of defectives
p1=x1/n1=15/500=0.03
Second factory sample size n2=400
Second factory no of defectives x2=20
Second factory Proportion of defectives
p2=x2/n2=20/400=0.05
Pooled Population Proportion
n1 p1  n2 p2 500(0.03)  400(0.05)
P 
n1  n2 500  400

P=0.038
P+Q=1
Q=1-P=1-0.038=0.962
Procedure:
Step 1: Null Hypothesis(H0): Ho: P1= P2 OR Ho: P1-
P2=0
ie there NO significance difference between two
Population Proportions
Step 2 : Alternative Hypothesis (H1): H1: P1≠ P2
Two Tailed Test(TTT);
ie there a significance difference between two
Population Proportions
Step 3: Level of Significant(LOS):
 Z tab= Z table value at 1% and Two Tailed Test(TTT)
Z tab=2.58
Step 4 : Test Statistic: Compute Test Statistic under
Ho

( p1  p2 )  ( P1  P2 )
Zcal 
1 1
PQ(  )
n1 n2

(0.03  0.05)  (0)

Zcal 
1 1
(0.038)(0.962)(  )
500 400
Zcal  1.5

Step 5: Conclusion: Z tab (2.58) > Z cal (1.5),

Accept Null Hypothesis Ho
Reject Alternative Hypothesis H1.
ie  There NO significance difference between two
Population Proportions