Dayananda Sagar College of Engineering

Department of Mathematics

SAMPLING THEORY
Population: A large collection of individuals or attributes or numerical data can be regarded
as population or universe. It is an aggregate of objects, animate or inanimate, under study.
The population may be finite or infinite.

If the population is large, complete enumeration is not possible most of the times because
of the cost involved, time consumed and also in some cases units are destroyed in the
course of inspection (e.g. inspection of crackers). So we take help of sampling.

A finite subset of the population is known as sample.

Size of the population N is the number of objects or observations in the population.

Population is said to be finite or infinite depending on the size N being finite or infinite.

Size of the sample is denoted by n. Sampling is the process of drawing samples from a given
population.

Large sampling: If n≥ 30, the sampling is said to be large

sampling. Small Sampling: If n<30, the sampling is said to be

small sampling. Examples:

1. Population of India is population where as the population of Karnataka is sample.

2. Cars produced in India are the population where as the mantis cars produced in India is
sample.

The statistical constants of the population such as mean (µ), Standard deviation ( σ) etc are
called the parameters. Similarly the constants for the sample drawn from the given
population i.e. Mean (x ) standard deviation (S) etc are called statistics.

Random sampling :

The selection of an item from the population in such a way that each has the same chance
of being selected is called random sampling.

Suppose we take a sample of size n from the finite population of size N. Random sampling is
a technique in which each element has an equal chance of being selected.

Sampling where each member of a population may be chosen more than once is called
sampling with replacement i.e. here the items are drawn one by one and are put back to the
population before the next draw. If N is the size of the finite population and n is the sample
size then we have Nn samples.
Sampling where if a member cannot be chosen more than once it is called sampling
without replacement. Here the items are drawn one by one and are not put back to the
population before the next draw. In this case there will be NCn samples.

Sampling distribution:

Given a population, suppose we consider a set of samples of a certain size drawn from the
population. For each sample, suppose we compute a statistics such as the mean, standard
deviation etc., these statistics will vary from the sample to the other sample, suppose we
group these statistics according to their frequencies and form a frequency distribution. The
frequency distribution so formed is called a sampling distribution.

The standard deviation of sampling distribution is called its Standard error.

The reciprocal of the standard errors is called precision.

Sampling distribution of Means :

Consider a population for which the mean is µ and the standard deviation is  suppose we
draw a set of samples of a certain size n, from this population and find the mean x of each
of these population. The frequency distribution of these means is called a sampling
distribution of means. Let the mean and the standard deviation of sampling distribution of
means be μX¯ and σX¯ respectively.

Suppose the population is finite with size N or random sampling without the replacement
i.e. the items drawn one by one and are not put back to the population before the next
draw. In this case there will be NCn samples and we have

μX¯ = μ and

σ2 X¯ σ2
[
N–n
= N–1
]
n
2 N–n
σ2 X¯ = c σ where c = [ ] is called the finite population correction
n N–1
factor. If N is very large i.e. if the population is infinite or the sampling is finite with
replacement then

c = 1 as N → ∞

σ2
∴ σ2 X¯ = n
So, the mean of sampling distribution is equal to population mean and the corresponding
σ
standard error is where σ is the standard deviation of the population.
√n

If the population is distributed normally with mean µ and S.D. σ, then the mean of all σ
positive random samples of size n are also distributed normally with mean µ and S.E. .
√n

Central Limit Theorem:

This is a very important theorem regarding the distribution of the mean of a sample if the
parent population is non-normal and the sample size is large.

If the variable X has a non-normal distribution with mean μ and standard deviation σ, then
the limiting distribution of
x¯ –μ
Z= , n → ∞ , is the standard normal distribution (i.e, with mean 0 and unit S D)
σ/√n

There is no restriction upon the distribution of X except that it has a finite mean and
variance. This theorem holds well for a sample of 30 or more which is regarded as large.
 Dayananda Sagar College of Engineering
Department of Mathematics

This is the content of central limit theorem.

Statistical Estimation is the method in which the parameters are estimated with the aid of the
corresponding statistics. An estimate of the unknown true or exact value of the parameter or
an interval in which the parameter is to be determined on the basis of sample data from the
population.

1. Confidence interval:

Consider sampling distribution of a statistic S. Suppose S follows normal distribution.

Let μs and σs be the mean and s.d. of the normal distribution.

a. The probability that μs lies in the interval (s-σs, s+σs ) is 68.26%

b. The Probability that μs lies in the interval (s- 2σs ,s+2) is 95.44%
c. The probability that μs lies in the interval (s- 3σs , s +3σs ) is 99.74%

The confidence intervals for μs indicated above are of the form

( s - zcσs , s+ zcσs)

Zc=1 at 68.26% confidence

level Zc=2 at 95.44%

confidence level Zc=3 at

99.74% confidence level

figure
Z% Confidence interval :

The Z% confidence interval for μs if P { ( s - zcσs , s+ zcσs ) } = z%

i.e. Z
= P { - zcσs ≤ μs – s ≤ zcσs }
100

= P { |s–μs | ≤ Z c }
σs
s– μs
= P { |Z| ≤ Zc } , where Z =
σs

then Z
= P { −Zc ≤ Z ≤ zc }
100

= 2 P { 0 ≤ Z ≤ Zc }

= 2 ∅ ( Zc )

Z = 2 ∅ ( Zc ) × 100

Thus Z% confidence interval of μs is the interval ( s - zcσs , s+ zcσs ) , where Zc is the

positive real number.

Confidence limit : The interval ( s - zcσs , s+ zcσs ) is the Z% confidence interval for μs
, then the quantities ( s ± Zcσs ) are called Z% confidence limits. The member Zc is called
the corresponding confidence coefficient or the critical value confidence.

The length of the confidence interval ( s - zcσs , s+ zcσs ) ie 2l = 2 Zcσs is called the error
in the confidence level.

Table for the confidence coefficients Zc for various values of Z

Z Zc Z Zc
50 .6745 90 1.645
55 .7639 95 1.96
60 .843 95.44 2
65 .9259 96 2.05
68.26 1 97 2.195
70 1.041 98 2.33
75 1.15 99 2.58
80 1.277 99.5 2.81
85 1.445 99.74 3
 The confidence interval for the population mean is ( ¯X¯¯ − Z s ¯X¯¯ + Z s
)
c c
√N
,
√N

1. A random sample of size N=100 is taken from a population with standard deviation
σ = 5.1. Given that the sample mean is X¯ = 21.6. Obtain the 95% confidence
interval for the population mean μ
N=100, σ = 5.1 , X¯ = 21.6

Confidence limits for the population mean are

s 5.1
¯X¯¯ ± Z
c √N = 21.6 ± c
√100
Z
For 95% confidence level , Zc = 1.96

∴ ¯X¯¯ ± = 21.6 ± (1.96)

5.1
= 21.6 ± .9996
c √100
s
Z
√N

Statistical Decision

Introduction:

For reaching statistical decisions, we start with some assumptions or guesses about the
populations involved. Such assumptions / guesses, which may or may not be true, are called
Statistical Hypotheses.

In many instances we formulate a statistical hypothesis for the sole purpose of

rejecting or nullifying it. Such Hypotheses are called Null hypotheses and are generally
denoted by H0 .

A statistical hypothesis which differs from a given hypothesis is called an alternative

hypothesis. A hypothesis alternative to the null hypothesis is generally denoted by H1 .

Example: i) Suppose we wish to reach the decision that a certain coin is biased ( that is, the
coin shows more heads tails or vice versa ). To reach this decision, we start with the
hypothesis that the coin is faired ( not biased) with the sole purpose of rejecting it (at the
end). This hypothesis is a null hypothesis.

ii) Consider the situation where the probability of an event is, say, 1/3, according
to some hypothesis. For arriving at some decision, if we make the hypothesis that the
probability is, say, ¼, then the hypothesis we have made is an alternative hypothesis.

Tests of hypothesis and significance:

Procedures which enable us to decide whether to accept or reject a hypothesis or to

determine whether observed samples differ significantly from expected results are called tests
of hypothesis, tests of significance, or rules of decision.
 By an error of judgement, suppose we reject a hypothesis, when it should have been
accepted. Then we say that an error of Type I have been made. On the other hand, suppose
we accept a hypothesis when it should be rejected; in this case, we say that an error of Type II
has been made.

Type I error:

In a hypothesis test, a type I error occurs when the null hypothesis is rejected when it is in
fact true; that is, H0 is wrongly rejected.

For example, in a clinical trial of a new drug, the null hypothesis might be that the new drug
is no better, on average, than the current drug; i.e.
H0: there is no difference between the two drugs on average.
A type I error would occur if we concluded that the two drugs produced different effects
when in fact there was no difference between them.
The following table gives a summary of possible results of any hypothesis test:

Decision

Reject H0 Don't reject H0

H0 Type I Error Right decision

Truth
H1 Right decision Type II Error

A type I error is often considered to be more serious, and therefore it is more

important to avoid than a type II error. The hypothesis test procedure is therefore adjusted so
that there is a guaranteed 'low' probability of rejecting the null hypothesis wrongly; this
probability is never 0. This probability of a type I error can be precisely computed as
P(type I error) = significance level =α

The exact probability of a type II error is generally unknown.

If we do not reject the null hypothesis, it may still be false (a type II error) as the sample may
not be big enough to identify the falseness of the null hypothesis (especially if the truth is
very close to hypothesis).

For any given set of data, type I and type II errors are inversely related; the smaller the risk of
one, the higher the risk of the other.

A type I error can also be referred to as an error of the first kind.

Type II Error:

In a hypothesis test, a type II error occurs when the null hypothesis H0, is not rejected when it
is in fact false. For example, in a clinical trial of a new drug, the null hypothesis might be that
the new drug is no better, on average, than the current drug; i.e.
H0: there is no difference between the two drugs on average.
A type II error would occur if it was concluded that the two drugs produced the same effect,
i.e. there is no difference between the two drugs on average, when in fact they produced
different ones.

A type II error is frequently due to sample sizes being too small.

The probability of a type II error is generally unknown, but is symbolised by β and written
P(type II error) = β

A type II error can also be referred to as an error of the second kind.

Levels of significance:

Suppose that, under a given hypothesis H, the sampling distribution of a statistic S is a

normal distribution with mean μs and standard deviation σs. Then

S–
zs =
μ
----------- (1) is the standard normal variate associated with S, so that for the
σs
distribution of Z the mean is zero and the standard deviation is 1.

Accordingly, for the distribution of z, the z% confidence interval is (-z c ,zc ). This
means that we can be Z% confident that, if the hypothesis H is true, then the value of z will
lie between -zc and zc . This is equivalent to saying that there is (100 – z)% chance that the
hypothesis H is true but the value of Z lies outside the interval (-zc , zc ). If we reject the
hypothesis H on the grounds that the value of Z lies outside the interval (-z c ,zc ), we would be
making a type I error and the probability of making the error is (100-Z)%. Her, we say that
the hypothesis is rejected at a (100 - Z)% level of significance. Thus, a level of significance
is the probability level below which we reject a hypothesis.

In practice, only two levels of significance are employed: one corresponding to Z = 95

and the other corresponding to Z= 99.

The value of the normal variate Z, determinate by using the formula (1) is usually
called the z – score of the statistic S. It is this score that determines the "fate” of a hypothesis
H and is called the test statistic.
Rule of decision:
“Reject a hypothesis H at a (100 – Z)% level of significance if the z – score of the statistic S,
determined on the basis of H, is outside the interval (-z c , zc). Do not reject the hypothesis
otherwise”.

Here the interval (-zc ,zc) is called the interval of test.

Critical Region: The region in which a sample value falling is rejected, is known as critical
region.

Normally, the test statistic, we consider follows normal distribution. Let us look into the
normal curve.

If level of significance is α% then we have critical region as follows

 Dayananda Sagar College of Engineering
Department of Mathematics

Student ‘t’ – distribution

Thus far, we considered sampling distributions on the assumption that they are normal or approximately normal.
As mentioned before, this assumption is valid when the sample size N is large. For small samples, this
assumption is not generally valid. In this section, we consider an important sampling distribution, called the ‘t-
distribution’, which is used in studies involving small samples.
Let N be the sample size, X́ and µ be respectively the sample mean and the population mean, and s be the sample
standard deviation. Consider the statistic -t defined by
X́−µ
t= ( ) √ N−1
s
Suppose w e obtain a frequency distribution of t by computing the value of t for each of a et of samples of size N
drawn from a normal or a nearly normal population. The sampling distribution so obtained is called the

‘Student’s’‘t - distribution’.
Properties of the t Distribution
The t distribution has the following properties:
The mean of the distribution is equal to 0 .
The variance is always greater than 1, although it is close to 1 when there are many degrees of freedom. With
infinite degrees of freedom, the t distribution is the same as the standard normal distribution.

Note: Table of values of t β ( γ ) at β=0.01∧0.05 levels of

significance for values ofγ ¿1 ¿ 50.

γ t 0.01 ( γ ) t 0.05 ( γ )
1 63.66 12.71
2 9.92 4.30
3 5.84 3.18
4 4.60 2.78
5 4.03 2.57
6 3.71 2.45
7 3.50 2.36
8 3.36 2.31
9 3.25 2.26
10 3.17 2.23
11 3.11 2.20
12 3.06 2.18
13 3.01 2.16
14 2.98 2.14
15 2.95 2.13
16 2.92 2.12
17 2.90 2.11
18 2.88 2.10
19 2.86 2.09
20 2.84 2.09
21 2.83 2.08
22 2.82 2.07
23 2.81 2.07
24 2.80 2.06
25 2.79 2.06
26 2.78 2.06
27 2.77 2.05
28 2.76 2.05
29 2.76 2.04
30 2.75 2.04
Chi - square test

In practice, expected (theoretical) frequencies are computed on the basis of a hypothesis H0. If under this
hypothesis the value of χ 2computed with the use of the formula:
2
n
(f k −e k )2
χ =∑
k=1 ek
It is greater than some critical value χ 2c , we would conclude that the observed frequencies differ significantly
from the expected frequencies and would reject Ho at the corresponding level of significance c. Otherwise; we
would accept it or at least not reject it. This procedure is called the Chi – square ( χ 2) Test of hypothesis or
significance.
Generally, the chi-square test is employed by taking c = 0.05 or 0.01. Tables giving the values of χ 2c for different
values of v are available. Below table gives the value of χ 2c at c = 0.05 and 0.01 levels and for v=1,2,……10.
Table of values of χ 2c (v) for c = 0.05 and 0.01
2 2
v χ 0.05 ( v) χ 0.01 ( v)

1 3.84 6.64
2 5.59 9.21
3 7.82 11.34
4 9.49 13.28
5 11.07 15.09
6 12.59 16.81
7 14.07 18.48
8 15.51 20.09
9 16.92 21.67
10 18.31 23.21

The number of degrees of freedom v is determined by using the formula v = n – m. Here, n is the number of
frequency – pairs (fi ,ei) used in the computation of and m is the number of quantities that are needed (and used)
in the calculation of the expected frequencies ei. if the N= ∑ f iis the only quantity used in the calculation of ei,
then m = 1 so that v = n-1.

Goodness of Fit
When a hypothesis H0 is accepted(or not rejected)on the basis of the Chi- square test, we say that the
expected frequencies calculated on the basis of H0 form a good fit for the given frequencies. When H0 is rejected,
we say that the corresponding expected frequencies do not form a good fit.
It has been mentioned that the sampling distribution of χ 2is approximately identical with the Chi-square
distribution when the expected frequencies are at least equal to 5. Therefore, the Chi- square test is applicable
only if every expected frequency is ≥ 5. If some expected frequencies are less than 5, then some frequencies are
to be clubbed together so that none of the expected frequency is less than 5.
 Dayananda Sagar College of Engineering
Department of Mathematics

Sampling Distribution

Q.No Question
1 a) Explain the following
i) Null hypothesis
ii) Alternative hypothesis
iii) Type I and type II error
iv) Level of significance
v) Standard error
b) A population has mean 75 and standard deviation 12.
a) Random samples of size 121 are taken. Find the mean and standard deviation
of the sample.
b) How would the answers to part a) change if the size of the samples were 400
instead of 121?

Given µ=75, σ=12, n=121

µ x́ = µ =75

σ 12 12
σ x́ = = = =1.09
√ n √ 121 11

n = 400

µ x́ = µ = 75

σ 12
σ x́ = = =0.6
√ n √ 20

So if the size of the samples is changed from 121 to 400. σ x́ decrease from 1.09

to 0.6.
 Dayananda Sagar College of Engineering
Department of Mathematics

2 a) A population has mean 5.75 and standard deviation 1.02.

a) Random samples of size 81 are taken. Find the mean and standard deviation of the
sample.
b) How would the answers to part a) change if the size of the samples were 25 instead of
81?

μ=5⋅ 75,0=1.02
¿ n=81

μ x́=μ=5 ⋅75
σ 1.02
−¿= = =0⋅ 1133
√n 9
σx
¿

μ x́=μ=5 ⋅75
σ 1.02
−¿= = =0⋅ 204
√n 95
σx
¿

changes¿
μ x́ remains same but the σ−¿
x 0.1133 ¿ 0.204

b) The weights of 1500 ball bearings are normally distributed with a mean of 635 gms and S.D of
1.36gms. If 300 random samples of size 36 are drawn from this population, determine the
expected mean and S.D of the sampling distribution of means if sampling is done a) with
replacement b) without replacement.
 Here N=1500
µ = 635 σ= 1.36 n= 36
a) Expected Mean μ x́=µ=635
2

σx
−¿=
√ σ
n
=¿ ¿¿
1.36/6=0.227
b) Expected Mean μ x́=µ=635
2

σx
−¿=
√ σ
n
=¿
σ
.
N −n
√
√ n N −1
¿¿

(1.36)2
√ 36
¿ √ 0.05
×
1500−36
1500−1 √
¿ 0.224

3 a) A population consists of 4 numbers 3, 7, 11, 15.

a) Find the mean and S.D. of the sampling distribution of means by
considering samplings of size 2 with replacement.
b) If n, n denotes respectively the population size and sample size, σ and
σx¯ respectively denotes population S.D. and S.D. of the sampling
distribution of means without replacement
2
−¿=
σx √ σ
n
=¿
σ
.
N −n
√
√ n N −1
¿¿

i.
ii. μx¯ = μ where μx¯ is the mean of this distribution and μ is the
population mean.

a) Population mean µ=1/4(3+7+11+15)=9

2 1 2 2 2 2
Population Variance σ = 4 {( 3−9 ) + ( 7−9 ) + (11−9 ) + (15−9 ) }=20

a) Let us consider the sample of Size 2 with supplement. They are follows

(3,3), (3,7), (3,11),(3,15),

(7,3), (7,7), (7,11),(7,15),

(11,3),(11,7),(11,11),(11,15),

(15,3),(15,7),(15,11),(15,15)
 Sampling means are as follows

3,5,7,9, 5,7,9,11, 7,9,11,13, 9,11,13,15

The frequency distribution of the sampling means is as follows

X: 3 5 7 9 11 13 15

f:1 2 3 4 3 2 1

f i xi 144
μ x́ =∑ = =9
fi 16

f i xi
σ 2x́ =∑ -( μ x́ )2
fi
=1456/16-92
=10
Thus μ x́ =9
❑
σ x́ =√ 10
b) Let us consider the sample without replacement they are as follows
(3,7),(3,11),(3,15),(7,11),(7,15),(11,15)
The sampling means are 5,7,9,9,11,13
μ x́=1/6(5+7+9+9+11+13)=9
μ x́=µ
1
σ 2x́ =
{(5−9 )2 + ( 7−9 )2 + ( 9−9 )2 +( 11−9 )2 + ( 13−9 )2 }
6
40 20
¿ =
6 3

σ 2 N −n
Consider n
x [
N −1 ]
20 4−2
¿
2
x [ ]
4−1
σ2 N −n
¿ x
n [ ]
N −1
2
= σ x́
b) Certain tubes manufactured by a company have mean life time of 800 hours and S.D of
60hours. Find the probability that a random sample of 16 tubes from the group will have a
mean life time a) between 790 hours and 810 hours b) less than 785 hours c) more than 820
hours d) between 770 hours and 830 hours.

3 b) By data µ=800, σ=60, n=15

σ 60
Therefore σ x́ = = =15
√n 4
x́−μ x́−800
z= = N ( 0,1 )
σ x́ 15
a ¿ P ( 790< x́< 810 )
790−800 810−800
¿p ( 15
<z<
15 )
¿ P ( 0⋅ 67< z <0 ⋅67 )
¿ p ( 0 ⋅67< z <0 ⋅67 )
¿ 0. 7486

b ¿ P ( ź<785 )
785−800
(
¿ p z<
15 )
¿ P ( z ←1 )
¿ P ( z >1 )
¿ 0.1587

¿ P ( x́>820 )
c¿ 820−800 ¿=P ( z>1 ⋅33 ) ¿=0.0918¿
(
¿ P z>
15 )
 d)

¿ P ( 770< x́ <880 )
770−880 830−800
p ( 15
<
15 )
¿ P (−2< z< 2 )
¿ 2 P ( 0< z<2 )
¿ 0 ⋅9 772

4 a) A prototype automotive tire has a design life of 38500 miles with S.D. of 2500 miles. Five such
tires are manufactured and tested. On the assumption that the actual population S.D. is 2500
miles, find the probability that the sample mean will be less than 36000 miles. Assume that the
distribution of lifetimes of such tires is normal.

4a) For simplicity we use units of thousands of miles

Then μ x́ = Mean of sample mean

μ=38⋅5 Thousands of miles

σ 2 ⋅5
σ x́ = = =1 ⋅11803 thousands of miles
√n √5
 These normally distributed
¿ P ( x́ <86 )

x́ is also the following normal distribution

¿ P ( x́ <86 )
36−μ x́
(
¿ P z<
σ x́ )
¿ P ( z ←2 ⋅24 )

= 0.0125

That is, if the time perfom ad designed, there is only about a 125% chance
that the average of a sample of this size would be so low

b) An automobile battery manufacturer claims that its midgrade battery has a mean life of 50
months with a S.D. of 6 months. Suppose the distribution of battery lives of this particular
brand is approximately normal. a) On the assumption that the manufacturer claims are true,
find the probability that a randomly selected battery of this type will last less than 48 months.
b) On the same assumption, find the probability that mean of a random sample of 36 such
batteries will be less than 48 months.

Given µ=Mean life of a battery=50 months

σ = Standard deviation= 6 months
Let X : life of a battery

x N ( μ , σ2)
x−μ
=2 N ( 0,1 )
σ
¿ p ( x< 48 )
48−μ
(
¿ P z<
σ )
¿ p ( z←0.33 )
¿ 0 ⋅3707
 n = Sample Size =36
Therefore, Sample Mean = μ x́ =¿ µ =50

σ 6
σ x́ = = =1
√ n √36
μ x́
x́ −¿ =P ( x́ <48 )
σ x =z N ( 0,1 ) ¿
48−50
¿ P z<( 1
⋅ )
¿ P ( z ←2 )
¿ 0.0228

5 a) The weights of 1500 ball bearings are normally distributed with a mean of 635 gms and S.D. of
1.36 gms. If 300 random samples of size 36 are drawn from this population. In the case of
random sampling with replacement, find how many random samples would have their mean
a)between 634.76gms and 635.24 gms, b) greater than 635.6 gms, c)less than 634.5 gms or
more than 635.24 gms

X= wt of ball bearing
N-pop Size =1500
µ= 635 gm
σ= 1.36 gm
No of random sample =300
n= sample size=36
X N ( μ , σ2)
x́ N ( μ x́ , σ 2x́ )
μ x́ =μ=635
σ 1 ⋅36
σ x́ = = =0.2267
√n 6
x́−635
z= N ( 0,1 )
0 ⋅2267
i) The no of random samples have their mean between 634.75 g & 635.24 g

= No.Of samples x p{634.76<X<635.24}

= 300 x P[ 1.0587 <z< 1.0587]

=300 x 0.8554

=256.62

=257

ii) No of Samples that have there mean greater than 635.6 g

= 300 x P[X>635.6]
= 300 x P [ Z> 2.6467]
= 300 x [0.5-0.4960]
= 300 x 0.004
=1.2
=1
 d)No of Samples with mean less than 634.5 g or more than 635.24g

=300 x [P[ x́ <634.5 ¿+ P[ x́> 635.24]¿

=300 x [p[z<-2.21]+P[z>1.06]]

= 300 x [0.0136+0.1446]

= 300 x[ 0.1582]

=47.46

b) 500 ball bearings have a mean weight of 142.30 gms and S.D. of 8.5 gms. Find the probability
that a random sample of 100 ball bearings chosen from this group will have a combined weight
a) between 14061 and 14175 gms b) more than 14460 gms

Population size =N=500

µ=142.3 gms
σ= 8.5 gms
sample size =n=100
x: wt of ball bearing
n is large by central limit theorem
x́−μ x́
N ( 0,1 )
σ x́
μ x́ =μ=142 ⋅3 gm
8 ⋅5
σ x́ = =0 ⋅85 gm
√n
 x́−142⋅ 3
z= N ( 0,1 )
0 ⋅ 85

i) P[ the combined wt of the group lies between 140.61 gms &141.75 gms]

=P[140.61/n<x,141.75 /n]

=P[140.61<x<141.75]

=p[-1.988<Z<-0.647]

=0.2345

ii) P[THE COMBINED WT OF THE GROUP IS MORE THAN 144.60

GMS]

=P[X>144.6]

=P[Z>2.71]

= 0.0034

6 a) The mean and S.D of the maximum loads supported by 60 cables are 11.09 tonnes and 0.73
tonnes respectively. Find a) 95% b) 99% confidence limits for mean of the maximum loads of
all cables by the company.
 6A) By data
X max load supported by a cable then
x́=11.09 , σ =b 0.73 , n=60
a) 95% confidence limits for the mean of maximum loads are given by

x ± 196 ( σ / √ n )
± 11.09 ±1 ⋅ 96 ( 0 ⋅73 / √ 60 )
á , 1.09 ±0 ⋅18
Limits are 10.91 tonner & 11.27 tonner

b) 99% confidence limits for the mean of maximum loads are given by

= x́ ± 2 ⋅58 ( σ / √ n )
¿ 11.09 ± 2⋅58 ( 0 ⋅73 / √ 60 )
¿ 11.09 ± 0 ⋅ 24
Limits are 10.85 tonners∧11.33 tonners

b) A sample of 900 men is found to have a mean height of 64inch. If this sample has been
drawn from a normal population with standard deviation 20 inch, find the 99% confidence
limits for the mean height of the men in the population.

6 b) Given n=900
Let x: ht
x́=64 inch , σ=20 inch
99% confidence limits for the mean is given by
¿´x ± 2 ⋅58 ( σ / √ n )
¿ 64 ± 2⋅58 ( 20 /30 )
¿ 64 ± 2⋅58 ( 0.6667 )
¿ 64 ± 1⋅720086
The limits are 62.279914 & 65.720086

7 a) A sample of 5000 students in a college was taken and their average height was found to be
62.5Kg with a standard deviation of 22kg. Find the 95% confidential limits of the average
weight of the students in the entire University.
 7 a) n=50000, x́= Average wt=62.5 kg
Σ= 22 kg
a) 95% Confidence limits for the mean of maximum loads are given by
x́ ± 1.96 ( 22/ √ n )
¿ 62 ⋅5± 1 ⋅ 96 ( 22/ √ 5000 )
¿ 62 ⋅5± 0 ⋅6098
Limits are 61.8902∧63.1098

b) Systolic blood pressure of 566 males was taken. Mean BP was found to be 128.8mm and SD
13.05mm. Find 95% confidence limits of BP within which the populations mean would lie.
7 b) Let x: systolic blood pressure
n= 566 , x́=128.8 mm , σ =13.05mm

x́ ± 1.96 ( σ / √ n )
¿ 129 ⋅8 ±1 ⋅ 96 ( 13.05/ √ 566 )
¿ 128 ⋅8 ±1.07506
Limits are 127.72494∧129.87506

8 a) Standard deviation of blood sugar level in a population is 6 mg%. If population mean is not
known, within what limits is it likely to lie if a random sample of 100 has a mean of 80mg%?

8 a) x : blood sugar level

n= 100, x́=80 mg , σ=6 mg
95% Confidence limits for the mean is given by
x́ ± 1.96 ( σ / √ n )
= 80 ± 1.176
The limits are 78.824 & 81.176

b) To know the mean weights of all 10 year old boys in Delhi a sample of 225 was taken. The
mean weight of the sample was found to be 67 pounds with s.d. of 12 pounds. What can we
infer about the mean weight of the population?
 8 b) x : wt of a 10 years old boy in Delhi
n = 22.5 x́=67 pounds , σ =12 pounds
95% Confidence limits of x is x́ ± 1.96 ( σ / √ n )
= 67 ± 1.96 ( 0.8 )
= 67 ± 1.568
The limits are 65.432 pounds
& 68.568 pounds
We can interfere that pop mean wt lies in (65.432, 68.568)
99% Confidence limits of x is
x́ ± 2.58 ( σ / √n )
= 67 ± 2.58 ( 0.8 )
= 67 ± 2.064
The limits are 64.936 & 69.064 pounds
We can interfere that pop mean wt lies in ( 64.936, 69.064) pound with 99 %
confidence

9 a) The mean and S.D of the diameters of a sample of 250 rivet heads manufactured by a company
are 7.2642 mm and 0.0058mm respectively. Find (a) 99% (b) 95% confidence limits for the
mean diameter of all the rivet heads manufactured by the company.
X: Diameter of rivert read
n = 250 , x́=7.2642 mm , σ=0.0058mm

95% Confidence limits of x is= 7.2642 ± 1.96 ( 0.0058/ √ 250 )

= 7.2642 ± 0.00072
The limits are 7.26348 mm & 7.26492 mm
 99% Confidence limits of x is x́ ± 2.58 ( σ / √n )
= 7.2642 ± 2.58 ( 0.0058/ √ 250 )
= 7.2642 ± 0.00095
The limits are 7.26325 mm & 7.26515 mm

b) Spring break can be a very expensive holiday. A sample of 80 students is surveyed, and the
average amount spent by students on travel and beverages is $593.84. The sample standard
deviation is approximately $369.34.Construct a 95% confidence interval for the population
mean amount of money spent by spring breakers.

X: Money spent by a student in a spring break.

n = 80 , x́=593.84 , σ =369.34
95% Confidence limits of x is
x́ ± 1.96 ( σ / √ n )
= 593.84 ± 1.96 ( 369.34/ √ 80 )
= 593.84 ± 80.9352
The limits are 512.9048 & 674.7752

10 a) 400 items are sampled from a normally distributed population with a sample mean x¯ of 22.1
and a population standard deviation(σ) of 12.8. Construct a 95% confidence interval for the
true population mean.
n = 400 , x́=22.1 , σ=12.8
95% Confidence limits of x is
x́ ± 1.96 ( σ / √ n )
= 22.1 ±1.96 ( 12.8 / √ 400 )
= 22.1 ± 1.2544
The limits are 20.8456 & 23.3544
 b) The mean and S.D. marks of a sample of 100 students are 67.45 and 2.92 respectively. Find
(a) 95% (b) 99% confidence intervals for estimating the marks of the population.

n = 100 , x́=67.45 , σ =2.92

95% Confidence limits of x is
x́ ± 1.96 ( σ / √ n )
= 67.45 ±1.96 ( 2.92 / √100 )
= 67.45 ± 0.57232
The limits are 66.87768 & 68.02232
99% Confidence limits of x is
x́ ± 2.58 ( σ / √n )
= 67.45 ±2.58 ( 2.92 / √ 100 )
= 67.45 ± 0.75336
The limits are 66.69664 & 69.20336

11 a) A machine is expected to produce nails of length 3 inches. A random sample of 25 nails gave
an average length of 3.1 inch with standard deviation 0.3. Can it be said that the machine is
producing nails as per specification?(t0.05 for 24 d.f. is 2.064)

a) Let x length of nail

µ pop mean of x =3 inch
n= 25
s = sample std deviation =0.3

x́=3 inch
t = x́−µ/s ¿
= 0.1/0.3 √ 25=1.67−2.064=t 0.05 , df
Thus the hypo that the machine is producing nails are per specifications is accepted
At 5% level of significance.
 b) Ten individuals are chosen at random from a population and their heights in inches are found to
be 63,63,66,67,68,69,70,70,71,71. Test the hypothesis that the mean height of the universe is
66 inches. ( t0.05=2.262 for 9 d.f.)

11 b) x: height of an individual in inches

µ =66 inchs
n= 10
∑ xi
x́= =67 ⋅ 8
n
1
s2= Σ ( x ;− x́ )2
( n−1 )
¿ 9 ⋅067
s=3.011
x́−μ
t= ⋅√n
s
678−66
¿ √ 10=1.89<2.262
3⋅ 0/1
¿ t 0.05 ∕ 9 ⅆf
12 a) A certain stimulus administered to each of 12 patients resulted in the following increases of
blood pressure: 5,2,8,-1,3,0,-2,1,5,0,4,6. Can it be concluded that the stimulus will in general be
accompanied by an increase in blood pressure. (t0.05 for 11 d.f. is 2.2)
Given n=12
X: increase in blood pressure
∑ xi
x́= = [5+2+8-1+0-2+1+5+0+4+6]/12
n
¿ 31 / 12=2.5833
1
s2= Σ ( x ;− x́ )2
( n−1 )
1 1
=
( n−1 ) (
∑ ζ 2i − ( ∑ x i )2
n )
1 1
¿
11(185− ( 31 )2
12 )
= 9 ⋅3−38
s=3.088
μ=0
x́−μ
t= √n
s
¿ 2 ⋅8979>2 ⋅201
11ⅆf

= t ❑0.05
The hype is rejected it with 95% confidence we can say that the stimulus in general is accompanied with
Increase in bp.
 b) A machinist is making engine parts with axle diameter of 0.7 inch. A random sample of 10
parts shows mean diameter 0.742 inch with a standard deviation of 0.04 inch. On the basis of
this sample, would you say that the work is inferior? ( t0.05=2.262 for 9 d.f.)

x: axle diameter
n=10
x́=0.742
S = 0.04
µ = 0.7

x́−μ
t= ⋅√n
s
= 0.042/0.04 √ 10
= 3.3204> 2.262
9ⅆf
= t ❑0.05
The hypo is rejected 0.7 on the basis of sample we can say that the work is interior

σs
13 a) Show that 95% confidence limits for the mean µ of the population are x¯ ± t0.05.

If t0.05 is the tabulated value of t for n-1 degrees of freedom at 5% 1.0.3

⇒ P[1 t 1 > t0.05 ] = 0.05

⇒ P[1 t 1 ≤ t0.05 ] = 1 - 0.05 = 0.95

i.e with 95% confidence

1 t 1 ≤ t0.05

x́−μ
⇒ | σs
. √ n ≤ t0.05
|
x́−μ
⇒ - t0.05 ≤ . √ n ≤ t0.05
σs

σst σs
⇒- 0.05
≤ x́−μ ≤ t0.05
√n √n
σst σs
⇒ x́ - 0.05
≤ μ≤ x́ + t0.05
√n √n
σst
∴ 95% confidence limits of μ is x́ ± 0.05

√n
 b) A random sample of 10 measurements of the diameter of a sphere gave a mean of 12 cm and
standard deviation 0.15 cm. Find 95% confidence limits for the actual diameter. ( t 0.05=2.262
for 9 d.f.)
13 b)
X = diameter of the sphere
n = 10; x́=12 cm; S = 0.15 cm
∴ 95% confidence limits are
S
x́ ± t 0.05
√n
12
i.e 12 ± (2.262)
√ 10
i.e 12 ± 0.1073
i.e 11.8927, 12.1073

14 a) A random sample of 10 boys had the following I.Q.: 70, 120, 110, 101, 88, 83,95, 98, 107,100.
Do these data support the assumption of a population mean I.Q. of 100 at 5% level of
significance (t0.05=2.262 for 9 d.f.)

14 a)
X = IQ of a boy
X́ = (70 + 120 + 110 + 101 + 88 + 83 + 95 + 98 + 107 + 100) / 10
= 972 / 10
= 97.2

1
S2 = ¿
9
= 203.7333

∴ S = 14.2735
2.8
∴1t1= √ 10 = 0.6203 < 2.262
14.273
∴ Hypothesis that the population mean IQ of 100 at 5% level of significance is accepted
b) A random sample of size 25 from a normal population has the mean 47.5 and s.d 8.4. Does this
information refute the claim that the mean of the population is 42.1.( t0.05=2.064 for 9 d.f.)
14 b)
n = 25; X́ = 47.5; S = 8.4
µ = 42.1
´
X−μ
∴ t= . √n
s

47.5−42
= . √ 25
8.4

= 45 / 14
= 3.2143 > 2.064 = t0.05 ; d.f = 24
∴ Hypothesis is rejected at 5% 1.0.3
15 a) A process for making certain bearings is under control if the diameter of the bearings have the
mean 0.5 cm. What can we say about this process if a sample of 10 of these bearings has a
mean diameter of 0.506 cm. and S.D. of 0.004cm?( t0.05=2.262 for 9 d.f.)
15 a)
X = diameter of bearing
µ = 0.5; n = 10; X́ =0.506 ; s = 0.004

´
X−μ
∴ t= . √n
s

0.006
= . 10
0.004 √

= 4.743 > 2.262

∴ The hypothesis is rejected at 5% 1.0.3

i.e the process is not under control

b) A machine is supposed to produce washers of mean thickness 0.12cm. A sample of 10 washers

was found to have a mean thickness of 0.128cm and standard deviation 0.008. Test whether
the machine is working in proper order at 5% level of significance.. ( t0.05=2.262 for 9 d.f.)
15 b)
X = thickness of washer

µ = 0.12; n = 10 ; X́ =0.128 ; s = 0.008

´
X−μ
∴ t= . √n
s

0.008
= . 10
0.008 √
= 3.1623 > 2.262 = t0.05 , 9 df
∴ The hypothesis is rejected i.e the machine is not working in proper order at 5% 1.03

16 a) The number of automobile accidents per week in a certain community are as follows:
12,8,20,2,14,10,15,6,9,4. Are these frequencies in agreement with the belief that the accident
conditions were the same during this 10 week period ?
16 a)
Total number of accidents in 10 week period = 100
∴ number of accidents expected per week = 100/10 = 10
Oi = 12 8 20 2 14 10 15 6 9 4
Ei = 10 10 10 10 10 10 10 10 10 10

∴ X2 = ∑ (Oi−Ei)2
Ei
1
= [4+ 4+100+64 +16+0+ 25+16+1+36]
10
266
=
10
= 26.6 > 16.92 = X 20.05 ; 9 df

∴ Hypothesis that number of accident conditions were same during this 10 week period is
rejected at 5% > 1.03
 b) A sample analysis of examination results of 500 students was made. It was found that 220
students had failed, 170 had secured third class 90 had secured second class and 20 had secured
first class. Do these figures support the general examination result which is in the ratio 4:3:2:1
for the respective categories. (χ2 0.05=7.81 for 3 d.f.)

16 b)
Let us take hypothesis that these figures support to the general result in the ratio 4:3:2:1
The repeated frequencies are

4 3 2 1
x 500 ; x 500 ; x 500 ; x 500
10 10 10 10
i.e 200; 150; 100; 50

∴ We have
Observed frequency Oi: 220 170 90 20
Expected frequency Ei: 200 150 100 50

∴ X2 = ∑ (Oi−Ei)2 = 13.67 > X 20.05 ; 3 df

Ei

∴ The hypothesis is rejected that is the figures don’t support to the general result in the ratio
4:3;2:1 with 5% 1.03

17 a) The following figures show the distribution of digits in numbers chosen at random from a
telephone directory. Test whether the digits may be taken to occur equally frequently in the
directory.(χ2 0.05=16.92 for 9 d.f.)

Digits 0 1 2 3 4 5 6 7 8 9
Frequency 1026 1107 997 966 1075 933 1107 972 964 853

17 a)
Digits 0 1 2 3 4 5 6 7 8 9
Frequency 1026 1107 997 966 1075 933 1107 972 964 853

We have total = 10000

Number of digits = 10
∴ number of each digits if equally distributed = 1000 = Ei
 Digit Ei Oi (Ei-Oi)2
0 1000 1026 676
1 1000 1107 11449
2 1000 997 9
3 1000 966 1156
4 1000 1075 5625
5 1000 933 4489
6 1000 1107 11449
7 1000 972 784
8 1000 964 1296
9 1000 853 21609

2 ∑ ( Ei−Oi)2
∴X =
Ei

= 58542 / 1000
= 58.542 > X20.05 , 9 df = 16.92
∴ The hypothesis is rejected at 5% = 10.8

b) Fit a Poisson distribution for the following data and test the goodness of fit given that
(χ2 0.05=7.81 for 3 d.f.)
x 0 1 2 3 4
f 122 60 15 2 1

17 b)
x 0 1 2 3 4
f 122 60 15 2 1

μ=
∑ fixi
∑ fi

0+60+30+6 +4
=
200

= 0.5

∴ We have μ = m = 0.5 for Poisson distribution

mx e−m
∴ P(m) =
x!
 f(x) = 200 x P(x)

200 x 0.5 x e−0.5

=
x!

1213 x e−0.5
=
x!

∴ f(0) = 121; f(1) = 61; f(2) = 15; f(3) = 3; f(4) = 0

∵ the last of the expected freq. is 0

So we shall club it with the previous one.

∴ Oi = 122 60 15 2+1 =3

Ei = 121 61 13 3

1 1
X2 = + + 0 + 0 = 0.025 < 7815 = X 20.05
121 61

∴ The Poisson distribution gives a good fit.

18 a) The number of accidents per day (x) as recorded in a textile industry over a period of 400 days
is given below. Test the goodness of fit in respect of Poisson distribution of fit to the given data
(χx2 0.050=9.491 for 42d.f) 3 4 5
f 173 168 37 18 3 1

18 a)
x 0 1 2 3 4 5
f 173 168 37 18 3 1

We have m=
∑ fixi = 0.7825
∑ fi
m x e−m 0.782x e−0.782
P(x) = =
x! x!
∴ f(x) = 400 x P(x)
(182.9) 0.7825x
=
x!
∴ f(0) = 183
f(1) = 143
f(2) = 56
 f(3) = 15
f(4) =3
f(5) =0
∵ the last of the expected freq. is 0, we shall club it with the previous one.
∴ Oi 173 168 37 18 (3+1 = 4)
Ei 183 143 56 15 (3+0 = 3)
2
∴ X2 = ∑
(Oi−Ei)
= 12.297
Ei
= 12.3 > 9.49 = X20.05
∴ with 5% > 1.0.3 we say that
Poisson fit is not good

b) In experiments on pea breeding, the following frequencies of seeds were obtained:

Round & yellow Wrinkled & yellow Round & green Wrinkled & green Total
315 101 108 32 556
Theory predicts that the frequencies should be in proportions 9:3:3:1. Examine the
correspondence between theory and experiment.(χ2 0.05=7.815 for 3 d.f.)

18 b)
Round & yellow Wrinkled & yellow Round & green Wrinkled & green Total
315 101 108 32 556

Total seeds = 556

We have expected frequencies of seeds of type
9
Round yellow = X 556 = 312.75
16
= 313
3
Wrinkled yellow = X 556 = 104.25
16
= 104
3
Round and green = X 556 = 104.25
16
= 104
1
Wrinkled and Green = X 556 = 34.75
16
= 35

Oi 315 101 108 32

Ei 313 104 104 35

2 ∑ (Oi−Ei)2
∴X =
Ei
2
2 32 4 2 32
= + + +
313 104 104 35
 = 0.0128 + 0.0865 + 0.1538 + 0.257

= 0.5103 < 7.815 = X20.05 3 df

∴ The correspondence between theory and experiment is good

19 a) 200 digits were chosen at random from a set of tables. The frequencies of the digits are shown
below. Use the chi square test to assess the correctness of the hypothesis that the digits were
distributed in equal number in the tables from which these were chosen..(χ2 0.05=16.92 for 90
d.f.)

Digit 0 1 2 3 4 5 6 7 8 9
Frequency 18 19 23 21 16 25 22 20 21 15
19 a)

We have total digits = 200

No. of digits = 10

No. of each digit if equally distributed = 20

Digit 0 1 2 3 4 5 6 7 8 9
Oi 18 19 23 21 16 25 22 20 21 15
Ei 20 20 20 20 20 20 20 20 20 20
(Oi – Ei)2 4 1 9 1 16 25 4 0 1 25

10
( Oi−Ei ) 2
X2 = ∑ = 86/20 = 4.2 < 16.92
i=1 Ei

∴ X20.05 = 9

∴ Hypothesis is accepted i.e digits were distributed in equal number in the tables from which

these were chosen.

b) Fit a Poisson distribution to the following data and test for its goodness of fit at level of
significance 0.05..(χ2 0.05=7.82 for 3 d.f.)
x 0 1 2 3 4
f 419 352 154 56 19

19 b)
x 0 1 2 3 4
fi 419 352 154 56 19
xfi 0 352 308 168 76

∴ m=
∑ fixi = 904/1000 = 0.904
∑ fi
∴ e-m = 0.4049

e−m mxi
∴ P(xi) =
x!

∴ Ei = 1000 x P(xi)

∴ Ei = 404.9 366 165.4 49.8 11.3

406.2 366 165.4 49.8 12.6

∴ We adjust in 1st to last value equalls

∴ Obj of X2 is reduced again by 1

∴ X2 = ∑ ( Ei−Oi)2
Ei

= 0.403 + 0.536 + 0.786 + 3.251

= 5.748 < X20.05 =7.72

∴ Poisson fit is a good fit

20 a) A pair of dice are shown 360 times and the frequency of each sum is indicated below. Would

you say that the dice are fair on the basis of the chi square test at 0.05 level of significance?.(χ2

=18.31 for 10 d.f.)

0.05

Sum 2 3 4 5 6 7 8 9 10 11 12
Frequency 8 24 35 37 44 65 51 42 26 14 14

20 a)

1
When a pair of dice is thrown, each sample point will have some Pred =
36

1
P[sum=2] = P (1,1) =
36

2
P [Sum=3] = P (1,2) + P (2,1) =
36

3
P [Sum=4] = P (1,3) + P (3,1) + P (2,2) =
36

4 5 6 5
Similarly, P[Sum=5] = , P[Sum=6] = , P[Sum=7] = ; P[Sum=8] = ;
36 36 36 36

4 3 2 1
P[Sum=9] = ; P[Sum=10] = ; P[Sum=11] = ; P[Sum=12] = ;
36 36 36 36

Ei = Expected freq. that sum = i = P[Sum=i] x 360

∴ We have

Sum 2 3 4 5 6 7 8 9 10 11 12
Oi 8 24 35 31 44 65 51 42 86 14 14
Ei 10 20 30 40 50 60 50 40 30 20 10

4 16 25 9 36 25 1 4 16 36 16
∴ X2 = + + + + + + + + + +
10 20 30 40 50 60 50 40 30 20 10

= 74.83 < X20.05 < 18.31

∴ The fit is good with 5% = 10.8

b) In 1000 extensive sets of trials for an event of small probability, the frequencies f0 of the

number x of successes proved to be:.(χ2 0.05=11.07 for 5 d.f.)

Fit a Poisson distribution to the data and test the goodness of fit.

x 0 1 2 3 4 5 6 7
f0 305 365 210 80 28 9 2 1

20 b)

Total number of trails = 1000

x 0 1 2 3 4 5 6 7
f0 305 365 210 80 28 9 2 1
xfo 0 365 420 240 112 45 12 7

∴ m=
∑ xf = 1201 / 1000 = 1.201
∑f
∈m m x 0.3009 m x
∴ P(x) = =
x! x!

x 0 1 2 3 4 5 6 7
Ei=P(x) 300.9 361.37 217.004 80.37 26.08 6.26 1.25 0.2
x 1000 ≃301 ≃561 ≃ 217 ≃ 87 ≃ 26 ≃6 ≃1 ≃0

∵ the last expected value is 0, we shall add it with the previous one

∵ the total freq. ∑ Ei = 999,

we adjust in the 1st and last value equating f(0) = 301.5 = f(6) + f(7) = 1.5

∴ We have,

Oi 305 365 210 80 23 9 3

Ei 301.5 361 217 87 26 6 1.5

 12.25 16 49 49 4 9 2.25
X2 = + + + + + +
301.5 361 217 87 26 6 1.5

= 4.0276 < 11.07 = X20.05 = 5

With 5% = 10.8; we can say that Poisson fit is good

 Multiple Choice Questions

Question Question
Number
1. Sample error of sample mean with µ as the population mean , σ as the
standard deviation and sample size n is given by
(a) σ (b) σ/n (c) σ/√n (d) none of these
2. A hypothesis is true, but is rejected. Then this is an error of type
(a) II (b) I (c) both I & II (d) none of these
3. A hypothesis is false but accepted, then there is an error of type
(a) II (b) I (c) both I & II (d) none of these
4. The finite population correction factor is
(a) n-N/N-1 (b) N-n/N-1 (c) N-1/N-n (d) none of these
5. The probability distribution of a statistic is called distribution.
(a)Normal (b)Binomial (c) Sampling (d) none of these
6. Sample is a subset of
(a)Data (b) group (c) population (d) distribution
7. Any numerical value computed from population is called
(a)Statistic (b) bias (c) sampling error (d) parameter
8. In a random sampling, the probability of selecting an item from the population is
(a) Unknown (b ) undecided (c) known (d) zero
9. In sampling with replacement ,a sampling unit can be selected
(a )Only once (b) more than once (c) less than once (d) none of above
10. Sampling in which a sampling unit cannot be repeated more than once is called
(a) Sampling without replacement (b) simple sampling
(c) Sampling with replacement (d) none of above