Practice all the yellow-marked questions.

You can ignore the red-marked questions.

Measures of Central Tendency:

Ungrouped Data:

A manufacturer of electronic components is interested in determining the lifetime of certain type of

battery. A sample in hours of life , is as follows:

123, 116, 122, 110, 175, 125, 125, 111, 118, 117

i)Find the sample mean, median and mode.

ii) Also Show that the sum of deviation of each value from their mean is zero.

Solution:

x i
i)Mean, x = i =1

123 + 116 + ............. + 117

=
10

=124.2 hours

So the average lifetime of all the batteries is 124.2 hours

ii) The sum of deviation of each value from their mean is zero

Calculation Table

xi x ( xi − x )
123 124.2 -1.2
116 -8.2
122 -2.2
110 -14.2
175 50.8
125 0.8
125 0.8
111 -13.2
118 -6.2
117 -7.2

 (x
i =1
i − x) = 0
 n
So The sum of deviation of each value from their mean is zero,  (x
i =1
i − x) = 0

Median: Since n=10 is even

Arrange the data in ascending order:

110,111, 116, 117, 118, 122, 123, 125, 125, 175

n
Median=Average of and
2

=Average of 5th and 6th observation

118 + 122
= = 120
2

So the median lifetime is 120 hours i.e. The middle most battery has lifetime of 120 hours.

Mode:

Mode= The value in the series xi which has maximum frequency.

x Tally marks fi

110 / 1

111 / 1

116 / 1

117 / 1

118 / 1

122 / 1

123 / 1

125 // 2

175 / 1

Total 10
Since 125 occurs most frequently. So, mode is 125.

Example of Grouped data:

The Lifetime of 40 similar car batteries is as follows:

Lifetime in years Number of batteries

1.5-2.0 2

2.0-2.5 1

2.5-3.0 4

3.0-3.5 15

3.5-4.0 10

4.0-4.5 5

4.5-5.0 3

i)Find the mean, median, and mode

ii)Also show the median and mode graphically

Solution:

Calculation table:

Lifetime in years Number of batteries xi =class midvalue f i xi

fi

1.5-2.0 2 1.75 3.5

2.0-2.5 1 2.25 2.25

2.5-3.0 4 2.75 11

3.0-3.5 15 3.25 48.75

3.5-4.0 10 3.75 37.5

4.0-4.5 5 4.25 21.25

4.5-5.0 3 4.75 14.25

 N=40 n

fx
i =1
i i = 138.5

fx i i
Sample Mean, x = i =1

138.5
=
40

=3.4625 year.

So, the average lifetime of all the batteries is 3.4625 year.

N 40
Median group= = = 20th observation
2 2

Lifetime in years Number of batteries xi =class midvalue Cumulative Frequency

fi

1.5-2.0 2 1.75 2

2.0-2.5 1 2.25 3

2.5-3.0 4 2.75 7

3.0-3.5 15 3.25 22 Median group

3.5-4.0 10 3.75 32

4.0-4.5 5 4.25 37

4.5-5.0 3 4.75 40

N=40

4th group is the Median group

 N
− P.c. f
Median = L + 2 c
fm

40
−7
= 3.0 + 2  0.5
15

= 3.43 years

So, the middle most batteries has lifetime of 3.43 years

Mode: 4th group is the modal group.

Here, 1 = 15 − 4 = 11

2 = 15 −10 = 5

1
Mode= L + c
1 +  2

11
= 3.0 +  0.5
11 + 5

=3.34 year

So the maximum batteries has a lifetime of 3.34 year.

Examples of geometric mean for ungrouped data:

Example: The annual rates of growth of output of a factory in 5 years are

5.0%, 2.5%, 7.5%, 10.0%, 5.0%.

Find the average rates of growth of output for the period.

Solution: So will calculate geometric mean because data are given in percentages.

Let the initial value is 100.

xi log xi

105.0 2.021
 102.5 2.011

107.5 2.031

110.0 2.041

105.0 2.021

Total  log xi = 10.125

n
 log xi 10.125
i =1
We know log GM= =
n 5

=2.025

GM=Antilog (2.025)=105.93

So the average rate of growth of output for the period is =105.93-100=5.93%

Example: In a certain factory a unit of work is completed by A in 4 minute, by B in 5 Minute, C in 6

minute, D in 10 minute, and E in 12 minute. What is the average number of unit of work completed per
minute? Which average do you consider here and why?

Solution:

Since the information are given in unit of time. So we will use harmonic mean.

Calculation Table

xi 1
xi
 A 4 0.25

B 5 0.20

C 6 0.167

D 10 0.100

E 12 0.083

n
1
x
i =1
= 0.8
i

n
We know, HM= =5/0.8=6.25 minute
n 1

i =1 x i

The average number of unit of work completed per minute is 1/6.25=0.16unit.

Relation between AM, GM and HM:

For same data AM  GM  HM

i)Check: 10, 20, 30, 50, 60

Solution:

xi 1 log xi
xi

10 0.10 1.00

20 0.05 1.30

30 0.033 1.47

50 0.02 1.69

60 0.016 1.77

n n n
1
 xi = 170
i =1

i =1 xi
= 0.22  log x =7.23
i =1
i
 n

x i
170
We know AM= x = i =1
= = 34
n 5
n

 log x i
7.23
log GM = i =1
= = 1.446
n 5

GM = Anti log(1.446) = 27.93

n 5
HM = n
= = 22.72
1
x
.22
i =1 i

Since the AM=34, GM=27.93, and HM=22.72, So it is clear that AM  GM  HM.

ii)For same data AM  GM  HM

Profit in Lac Taka Number of companies

25-35 4

35-45 7

45-55 15

55-65 14

65-75 8

75-85 2

Calculation Table

Class interval xi fi f i xi f i log xi

 25-35 30 4 120 5.908 .133

35-45 40 7 280 11.214 .175

45-55 50 15 750 25.485 .3

55-65 60 14 840 24.894 .233

65-75 70 8 560 14.761 .114

75-85 80 2 160 3.806 .025

N=50

We know,

We know,

We know,

Since the AM=54.2, GM=52.65, and HM=51.020, So it is clear that AM  GM  HM

Weighted average:
Ex : In a company, there are 50 skilled workers with daily wage 500 TK,25 semi-skilled workers with
daily wage 400 Tk and 15 unskilled workers with daily wage 300 TK. What is the average daily wages of
the workers?

Solution: Let , = no.of Skilled worker=50

W2= no. semi-skilled worker=25

W3= no. of unskilled worker=15

Weighted Average =

50  500 + 25  400 + 15  300

=
50 + 25 + 15

=438.89TK

So the workers earn on average 438.89 Taka daily.

 Measures of Dispersion (Maths)

Example of ungrouped sample data:

The government awarded grants to the agricultural department of 9 universities to test the yield
capabilities of two new varieties of wheat. The yields in kilograms per plot are recorded as
follows:
University 1 2 3 4 5 6 7 8 9
Variety
1 38 23 35 41 44 29 37 31 38
2 45 25 31 38 50 33 36 40 43
Compare the variability in the yield capability using an appropriate measure. Yield capability of
which variety shows more consistency?

Solution:
To compare the variability in the data set, standard deviation and coefficient of variation is the
most appropriate measure.
Variety 1:
−
Sample mean, x =
 xi = 38 + 23 + ...... + 38 = 316 = 35.11 kg
n 9 9
Calculation Table
xi x i2
38 1444
23 529
35 1225
41 1681
44 1936
29 841
37 1369
31 961
38 1444
x i =316 x 2
i =11430

 (x i − x)
2

Sample Standard Deviation, s = i =1

n −1
 n

x
i =1
2
i − nx 2
=
n −1
11430 − 9  35.112
=
9 −1
=6.47 kg

s
Coefficient of variation, CV=  100
x
6.47
= 100
35.11
=18.43%

Variety 2:
−
Sample mean, x =
x i
=
45 + 25 + ...... + 43 341
= = 37.89 kg
n 9 9

Calculation Table:
xi 45 25 31 38 50 33 36 40 43  x =341
i

xi2 2025 625 961 1444 2500 1089 1296 1600 1849  x =13389
2
i

 (x i − x)
2

Sample Standard Deviation, s = i =1

n −1
n

x
i =1
2
i − nx 2
=
n −1
13389 − 9  37.89 2
=
9 −1
=7.75 kg

s
Coefficient of variation, CV=  100
x
7.65
= 100
37.89
 =20.19%
Interpretation: Since the CV of yield capability of variety 1 is 18.43% and the CV of yield
capability of variety 2 is 20.19%. There is more variation in the distribution of variety 2 relative
to the distribution of variety 1.So distribution of variety 2 is more variable and distribution of
variety 1 is more stable and consistent.
Example of grouped sample data:
The following data represent the electricity cost for a sample of 50 two bedroom apartment and a
sample of 50 three bedroom apartment in a city of Newmexico during the month of May last
year:

Electricity cost Two bedroom Three bedroom

No of apartments No of apartment
120-140 2 3
140-160 7 4
160-180 20 10
180-200 10 25
200-220 5 7
220-240 6 1

Compare the variability in the electricity costs using an appropriate measure. Electricity cost of
which type of apartment shows more consistency?

Solution:
To compare the variability in the data set, standard deviation and coefficient of variation is the
most appropriate measure.

Calculation Table for two bedroom apartment

Electricity Cost No of Apartments Class Midvalue f i xi f i xi2
fi xi

120-140 2 130 260 33800

140-160 7 150 1050 157500
160-180 20 170 3400 578000
180-200 10 190 1900 361000
200-220 5 210 1050 220500
220-240 6 230 1380 317400

Total: N=50 fx i i = 9040 fx 2

i i = 1668200
Sample Mean, x =
fx i i

N
9040
=
50
= 180.8$

Sample Standard Deviation, s =

fx 2
i i − N x2
N −1
1668200 − 50 180.8 2
=
50 − 1
= 26.25$
s
Coefficient of Variation, CV=  100
x
26.25
= 100
180.8
= 14.52%

Calculation Table for three bedroom apartment

Electricity Cost No of Class Midvalue f i xi f i xi2
Apartments xi
fi
120-140 3 130 390 50700
140-160 4 150 600 90000
160-180 10 170 1700 289000
180-200 25 190 4750 902500
200-220 7 210 1470 308700
220-240 1 230 230 52900
Total: N=50 fxi i = 9140 fx 2
i i = 1693800

Sample Mean, x =
fx i i

N
9140
=
50
= 182.8$
Sample Standard Deviation, s =
fx 2
i i − N x2
N −1
1693800 − 50  182.82
=
50 − 1
= 21.67$

s
Coefficient of Variation, CV=  100
x
21.67
=  100
182.8
= 11.85%
Interpretation: Since the CV of electricity cost for sample of two bedroom apartment is 14.52%
and the CV of electricity cost for sample of three bedroom apartment is 11.85% . It is clear that
there is more variation in the distribution of two bedroom sample relative to the distribution of
three bedroom sample. So distribution of two bedroom sample is more variable and distribution
of three bedroom sample is more stable and consistent.

Example: Distribution of test Score in a course:

Mean Test Score=200
Standard deviation=40
Distribution of year of service in a course:
Mean year of service=20 years
Standard deviation=2 years
Compare the variability of the two distribution using an appropriate measure.
Do yourself