CHAPTER MEASURES OF CENTRAL TENDENCY (TOPPERS INSTITUTE)

12.1 CENTRAL TENDENCY

1. Meaning: Central Tendency is defined as the tendency of a given set of observations to
be centred around a single central vaiue or middle value and the single value which
represents a given set of observations is called the measure of central tendency.
2. Different Measures of Central Tendency: The measures of Central Tendency are:
(a) Arithmetic Mean
(b) Median
(c) Partitional Values - Quartiles, Deciles etc.
(d) Mode
(e) Geometric Mean
(f) Harmonic Mean
3. Criteria for an Ideal Measure of Central Tendency:
(a) Properly and Unambiguously defined.
(b) Easy to understand.
(c) Simple to compute.
(d) Based on all the observations.
(e) Have certain desirable mathematical properties.
(f) Should be least affected by the presence of extreme observations.
Arithmetic Mean
A. Meaning: Mean is the most used method to statistically represent and operate with a
given set of data. The Simple Arithmetic Mean is defined as the sum total of the
observations divided by the total number of observations.
B. Computation
1. Discrete Data: For a given set of observations, the AM may be defined as the sum of all
𝛴𝑥
the observation to be divided by the number of observations. 𝑥 = 𝑖
𝑁
𝛴𝑓𝑖 𝑥𝑖
2. Grouped Data: In case of simple frequency distribution (Grouped Data), 𝑥 =
𝑁
3. Continuous Data: AM from Grouped Frequency of Continuous Data (Data represented in
Class Intervals) is given by –
𝛴𝑓𝑖 𝑑𝑖×𝐶 𝑥𝑖 −𝐴
𝑥 =𝐴+ ; where, 𝑑𝑖
𝑁 𝐶

A = Assumed Mean ; C= Class Length

∑𝑊𝑖 𝑋𝑖
4. Weighted AM = Where w, represents the weights assigned to observations.
∑𝑊𝑖

C. Properties of AM:
1. AM of a Constant Series: If all the observations are same (say a), then the AM is also a.
2. Sum of Deviations from AM: The algebraic sum of deviations of a set of observations
from their AM is zero i.e. for unclassified data, ∑(xi - 𝑥̅ ) = 0 and for grouped frequency
distribution, ∑fi(xi - 𝑥̅ ) = 0 .
3. Changes in Origin and Scale: AM is affected due to a change of origin and / or scale
which implies that if the original variable x is changed to another variable y by effecting a
change of origin, say a, and scale say b, of x i.e.
y=a + bx, then the AM of y is given by 𝑦̅ = a + b𝑥̅ .
4. Combined AM: If there are two groups containing n1 and n2 observations, 𝑥̅̅̅1 and 𝑥̅ as the
respective arithmetic means, then the combined AM is given by
𝑛1 𝑥̅1 + 𝑛2 𝑥̅2 ∑𝑛𝑖 𝑥̅𝑖
𝑥̅ = ; 𝑥̅ = for i = 1,2,3,...
𝑛1 +𝑛2 ∑𝑛𝑖

[Refer Illustrations 1 to Illustrations 6]

Median
A. Meaning: Median is the middlemost value when the observations are arranged either in
an ascending or descending order of magnitude. It is a positional average, since the value of
the median is dependant on the position of the given observation.
B. Computation:
1. Discrete Data: Median is the Value of the (n+1) / 2th observation.
2. Grouped Data: Median is the Value of the observation corresponding to the (N+1) / 2 th
frequency.
3. Continuous Data: In case of Continuous Data median is calculated as follows -
𝑁
−𝑚
2
𝑀𝑒𝑑𝑖𝑎𝑛 = 𝐿 + ×𝑐
𝑓

Where,
L = Lower Class Limit of the Median Class
m = Cumulative Frequency of the Class preceeding the Median Class
f = frequency of the Median Class
c = Class Magnitude of the Median Class
C. Properties of Median:
1. Middle Value: The Median will have 50% of the total items having values less than its
value. Therefore, the other 50% of the Observations will have values greater than the
Median Value.
2. Change of Origin and Scale: If x and y are two variables, to be related by y=a+bx for any
two constants a and b, then the median of y is given by y me = a + bxme
3. Mean Deviation from Median: For a set of observations, the sum of absolute deviations
is minimum when the deviations are taken from the median. This property states that ∑|× i -
A| is minimum if we choose A as the median. Since the total of Absolute Deviations from
Median is the lowest, the Mean of Absolute Deviations from Median is also the Lowest.
[Refer Illustrations 7 and 8]
Partitional Values
1. Partitional values are values dividing a given set of observations into a number of equal
parts. Examples: Quartiles, Deciles, Percentiles.
2. Quartiles:
(a) Meaning: Quartiles are values dividing a given set of observations into four equal parts.
There are totally 3 Quartiles Q1, Q2, Q3.
(b) Computation:
• Discrete Data: The n* Quartile is the Value of the n(N+1) / 4th observation. [Where n is the
Quartile's Number].
• Grouped Data: The nth Quartile is the Value of the observation corresponding to the
n(N+1) / 4th frequency.
• Continuous Data: In case of Continuous Data the Quartiles are calculated as follows -
𝑛𝑁
−𝑚
𝑄𝑛 = 𝐿 + 4 ×𝑐
𝑓
Where,
n = 1,2,3
𝑛𝑁
Relevant Quartile Class- The Class corresponding to the Cumulative Frequency
4

L = Lower Class Limit of the Relevant Quartile's Class

m = Cumulative Frequency of the Class preceeding the Relevant Quartile's Class
f = frequency of the Relevant Quartile's Class
c = Class Magnitude of the Relevant Quartile's Class
(c) Properties:
• Percentage of Values: Each Quartile has the following percentage of the values less than
it.

Quartile Percentage of values less than it

Quartile 1 25% of Values

Quartile 2 50% of Values

Quartile 3 75% of Values

• Change of Origin and Scale: If x and y are two variables, to be related by y=a+bx for any
two constants a and b, then the Qn of y is given by Qny = a + bQnx
3. Deciles:
(a) Meaning: There are Nine Deciles, dividing the set of observations into ten equal parts.
Each deciles have 10% of the total values between it and the previous Decile. ∴ for example,
2nd Decile has 20% of the Total Items less than it, 6th Decile has 60% of the Total Items less
than it.
(b) Computation:
• Discrete Data: The nth Decile is the Value of the n(N+1) / 10th observation. [Where n is the
Decile's Number].
• Grouped Data: The nth Decile is the Value of the observation corresponding to the n(N+1)
/ 10th frequency.
• Continuous Data: In case of Continuous Data, the Deciles are calculated as follows -
𝑛𝑁
−𝑚
𝐷n = 𝐿 + 10 ×𝑐
𝑓
Where,
n = 1,2,3,4,5,6,7,8,9,10
 𝑛𝑁
Relevant Decile Class - The Class corresponding to the Cumulative Frequency
10

L = Lower Class Limit of the Relevant Decile's Class

m = Cumulative Frequency of the Class preceeding the Relevant Decile's Class
f = Frequency of the Relevant Decile's Class
c = Class Magnitude of the Relevant Decile's Class
(c) Change of Origin and Scale: If x and y are two variables, to be related by y=a+bx for
any two constants a and b, then the Dn of y is given by Dny = a + bDnx
4. Percentiles:
(a) Meaning: There are 99 Percentiles dividing the set of observations into One Hundred
equal parts. Each Percentile has 1% of the total values between it and the previous
Percentile. for example, 21st Percentile has 21% of the Total Items less than it, 75 th
Percentile has 75% of the Total Items less than it.
(b) Computation:
• Discrete Data: The nth Percentile is the Value of the n(N+1) / 100th observation. [Where n
is the Decile's Number].
• Grouped Data: The nth Percentile is the Value of the observation corresponding to the
n(N+1) / 100th frequency.
• Continuous Data: In case of Continuous Data the Deciles are calculated as follows -
𝑛𝑁
−𝑚
𝑃𝑛 = 𝐿 + 100 ×𝑐
𝑓
Where,
n = 1,2,3,...100
𝑛𝑁
Relevant Percentile Class - The Class corresponding to the Cumulative Frequency
100

L = Lower Class Limit of the Relevant Percentile's Class

m = Cumulative Frequency of the Class preceeding the Relevant Percentiles's Class
f = Frequency of the Relevant Percentile's Class
c = Class Magnitude of the Relevant Percentile's Class
(c) Change of Origin and Scale: If x and y are two variables, to be related by y=a+bx for
any two constants a and b, then the P n of y is given by Pny = a + bPnx
[Refer Illustrations 9 and 10]
Mode
1. Meaning: Mode is the value around which the items of a distribution are most closely
concentrated. In Simpler Terms, Mode is the value that occurs the maximum number of
times. Therefore it is also called as the Fashionable Average.
2. Types of Distributions: There may be more than one mode for a given distribution.
• Bi-modal distribution - Distribution with two modes
• Multi-modal distribution - Distribution with many modes.
3. Computation:
• Discrete Distribution: The observation which is repeated the most number of times.
• Grouped Distribution: The observation corresponding to the highest Frequency
• Continuous Distribution: The mode is calculated as follows -
𝑓1 −𝑓0
Mo = L + ×C
2𝑓1 −𝑓0 −𝑓2

Where,
L = LCB of the modal class, i.e. the class containing the highest frequency
f1= Frequency of the modal class
f0 = Frequency of the pre - modal class
f2 = Frequency of the post modal class
C = Class length of the modal class
4. Change of Origin and Scale: If x and y are two variables, to be related by y=a+bx for any
two constants a and b, then the Mode of y is given by y mo = a + bxmo
[Refer Illustration 11 to 14]
Relationship between Mean, Median and Mode
1. In a Symmetrical Distribution, Mean = Median = Mode.
2. In a Moderately Asymmetrical Distribution, Mode = 3 Median - 2 Mean. This relationship
between Mean, Median and Mode can be used for determining the Mode in case of Bi-Modal
and Multi-Modal Distributions.
[Refer Illustration 12]
Geometric Mean
A. Meaning and Significance: Geometric Mean is the nth root of n terms. It is the best
measure of Central Tendency, for ascertaining the average rate of change over a period of
time.
B. Computation:
1. Discrete Data: If x1, x2, x,…… xn are n values of a variate x, none of them being zero,
then the geometric
mean G is defined as G = (x1 × x2 × x3…………………. × xn)1/n
2. Grouped / Continuous Frequency Distribution: GM is given by
𝔨 𝑓 𝑓 𝑓
𝐺 = (𝑥11 × 𝑥22 × 𝑥33 … … … … … … . .× 𝑥𝑛𝑛 )1/𝑁
𝔣 𝔣
⟹ 𝑙𝑜𝑔 𝐺 = 𝑙𝑜𝑔 (𝑥1 × 𝑥22 × 𝑥3𝔨 3 … … … … … … . .× 𝑥𝑛∩ )1/𝑁
∑𝑛
𝑖=1 𝑓𝑖 𝑙𝑜𝑔𝑥𝑖
⟹ G = Antilog[ ], where N= ∑fi
𝑁

Note: In case of Continuous Distributions, the class mark of the class to be taken as x.
∑𝑊𝑖 𝑙𝑜𝑔 𝑥𝑖
3. Weighted GM = Antilog
∑𝑤𝑖

C. Properties:
1. The Geometric Mean is 0 even if one of the Observations is 0.
2. If all the observations are same, (say a), then the GM of the observation is also a.
3. The product of observations will remain unchanged if the Geometric Mean is substituted in
place of all the observations.
4. GM of the product of two variables is the product of their GM's i.e. if z = xy, then GM of z =
(GM of x) × (GM of y)
5. GM of the ratio of two variables is the ratio of the GM's of the two variables i.e. if z = x/y
𝐺𝑀 𝑜𝑓 𝑥
then GM of z =
𝐺𝑀 𝑜𝑓 𝑦

[Refer Illustration 15]

Harmonic mean
A. Meaning and Significance: For a given set of non-zero observations, harmonic mean is
defined as the reciprocal of the AM of the reciprocals of the observations. It is the most
suitable measure of Central Tendency for finding out average of any ratio.
B. Computation:
1. Discrete Data: If a variable x assumes n non-zero values x1,x2,x3, ,xnthen the HM of x is
given by
𝑛
H=
∑(1/𝑥𝑖 )

2. Grouped Data: For a grouped frequency distribution, we have -

𝑁
H= 𝑓
∑[ 𝑖 ]
𝑥𝑖

∑𝑤𝑖
3. Weighted HM = 𝑤
∑[ 𝑖 ]
𝑥𝑖

C. Properties of HM:
1. If all the observations taken by a variable are same (say x), then the HM of the
observations is also x.
2. If even one of the observations is zero, then the Harmonic Mean is not defined.
3. If there are two groups with nt and n2 observations and Hf and H2 as respective HM's then
𝑛 +𝑛
the combined HM is given by 𝑛11 𝑛22
+
𝐻1 𝐻2

D. Relationship between AM, GM and HM:

• AM = GM = HM only where all the values are identical.
• AM > GM > HM, where the observations are all positive and unequal quantities.
[Refer Illustration 16 to 18]
 Illustrations
Illustration 1: Calculation of Simple Arithmetic Mean - Discrete Series
Following are the daily wages in rupees of a sample of 9 workers: 58, 62, 48, 53, 70, 52, 60,
84, 75, Compute the mean wage.
Solution: Let x denote the daily wage in rupees.
Then as given, x1 = 58, x2 = 62, x3 = 48, x4 = 53, x5 = 70, x6 = 52, x7 = 60, x8 = 84, x9 = 75.
∑9
𝑥 (58+62 + 48 + 53 + 70 + 52 + 60 + 84 + 75) 562
𝑋̅ = 𝑖=1 𝑖 = = = 62.44.
9 9 9

Illustration 2: Construction of Frequency Distribution and ascertaining of Averages

Following are the salaries of 20 workers of a firm expressed in thousand rupees: 5,17,12, 23,
7,15, 4,18,10, 6,15, 9, 8,13,12, 2,12, 3,15,14. The firm gave bonus amounting to Rs. 2000,
Rs. 3000, Rs. 4000, Rs. 5000 and Rs. 6000 to the workers belonging to the salary groups
1000 - 5000, 6000 -10000 and so on and lastly 21000 - 25000. Find the average bonus paid
per employee.
Solution:
Computation of Average bonus

Salary in 000's No of workers (f) Bonus in fx

Tally Mark. (2)
Rs. (1) (3) Rupees (x) (4) (5) = (3) × (4)

1-5 |||| 4 2000 8000

6-10 |||| 5 3000 15000

11 - 15 |||| ||| 8 4000 32000

16-20 || 2 5000 10000

21-25 | 1 6000 6000

Total ∑f= 20 ∑fx = 71000

∑𝑓𝑥 71000
Hence, the average bonus paid per employee = = Rs. 3550
∑𝑓 20

illustration 3: Calculation of Simple Arithmetic Mean - Grouped Discrete Series

Compute the mean weight of a group of BBA students of St. Xavier’s College from the
following data:

Weight in kgs. 44-48 49-53 54-58 59-63 64-68 69-73

No. of Students 3 4 5 7 9 8

Solution:
Computation of mean weight of 36 BBA students

Weight in kgs. No. of Student (f) Mid-Value (x) fx

(1) (2) (3) (4) = (2) x (3)

44-48 3 46 138
 49-53 4 51 204

54-58 5 56 280

59-63 7 61 427

64-68 9 66 594

69-73 8 71 568

Total ∑f = N = 36 - ∑fx = 2211

∑𝑓𝑥 2211
We know that Arithmetic Mean x = = = 61.4167 Kgs.
𝑁 36

Illustration 4: Calculation of Simple Arithmetic Mean - Grouped Discrete Series Find he AM

for the following distribution:

Class
350-369 370-389 390-409 410-429 430-449 450-469 470-489
Intervals

Frequency 23 38 58 82 65 31 11

Solution: Since the figures involved in the given frequency are large, we take the help of the
Assumed Mean to Compute the Arithmetic Mean. The Assumed Mean is taken as the mid-
value of the Middle Most class of the Distribution in case of an odd number of class intervals
and any one of the two middle most mid-values for an even number of class intervals. The
class length is taken as C. Suitable Approximations may be made to avoid non-integer
values.
𝑥−𝐴 𝒙−𝟒𝟏𝟗.𝟓𝟎
Class Interval Frequency (f) Mid-Value (xi) 𝑑 = = fd
𝑐 𝟐𝟎
(1) (2) (3) (4) (5) = (2) × (4)

350-369 23 359.50 -3 -69

370-389 38 379.50 -2 -76

390-409 58 399.50 -1 -58

410→29 82 419.50 0 0

430-449 65 439.50 1 65

450-469 31 459.50 2 62

470-489 11 479.50 3 33

Total ∑f = 308 A=419.50 ∑fd = -43

Note: A stands for Assumed Mean

∑𝑓𝑑 (−43)
The required AM is given by 𝑥̅ = A + x C = 419.50 + ×20 = 419.50 - 2.7922
𝑁 308
=416.7078
Illustration 5: Reverse Working - Find the Missing Frequencies
Given that the mean height of a group of students is 67.45 inches. Find the missing
frequencies for the following incomplete distribution of height of 100 students.

Height in inches 60-62 63-65 66-68 69-71 72-74

No. of Students 5 18 - - 8

Solution: Let x denote the height and fi and f2 as the two missing frequencies.
∴ (5+18+8)+ f1+ f2 = 100
31 + f1 + f2 = 100
⟹ f1 + f2 = 69 ... (1)
𝑥𝑖 −𝐴 𝒙𝒊 −𝟔𝟕
Class Interval Frequency (fi) Mid-Value (xi) 𝑑= = fidi
𝑐 𝟑
(1) (2) (3) (4) (5) = (2) × (4)

60-62 5 61 -2 -10

63-65 18 64 -Is -18

66-68 fi 67 0 0

69-71 f2 70 1 f2

72-74 8 73 2 16

Total ∑f=N = 31 + f1 + f2 A = 67 - ∑fd = –12+f2

𝑥̅ = 67.45 [Given]
∑𝑓𝑑
A+[ × 𝑐] = 67.45
𝑁
−12 + 𝑓2
67 + [ × 3 ] =67.45
100

(-12 + f2) x 3 = (67.45 - 67) x 100

-12 + f2 = 15
f2 = 27
On substituting 27 for f2 in (1), we get
fi + 27 = 69 ⟹ f2 = 42
The missing frequencies are 42 and 27.
Illustration 6: Combined Mean - Calculations
The mean salary for a group of 40 female workers is ' 5200 per month and that for a group of
60 male workers is ' 6800 per month. What is the combined salary?
Solution: As given m = 40, n2 = 60, 𝑥̅ 1 = ' 5200 and 𝑥̅ 2 = ' 6800 hence, the combined mean
salary per month is
𝑛1 𝑥1+n2 𝑥2 (40×𝑅𝑠.5200)+(60×𝑅𝑠.6800)
𝑥= = = ′6160.
𝑛1 +𝑛2 40+60

Illustration 7: Median
Compute the median for the distribution as given in Illustration 4.
Solution:
Computation of Median

Less than
Exclusive
Class Interval (1) Frequency (fj) (3) Cumulative
Classes (2)
Frequency (4)

350-369 349.50 - 369.50 23 23

370-389 369.50 - 389.50 38 61

390→09 389.50 - 409.50 58 119 m

410-429 409.50 - 429.50 82 201 Median Class

430-449 429.50 - 449.50 65 266

450-469 449.50 - 469.50 31 297

470-489 469.50 - 489.50 11 308

Total Total 308

𝑁
−𝑚
2
Median = 𝐿 + ×𝑐
𝑓
𝑁 302
= = 154
2 2

Since the 154th items lies in the class 409.50-429.50, the same is chosen as the Median
Class. Note: The Class 409.50- 429.50 contains the Items 120 to 201.
Where, L is the Lower Class Limit of the Median Class = 409.50 m is the cumulative
Frequency of the Class preceeding the Median Class= 119 f is the frequency of the Median
Class = 82 c is the Class Magnitude of the Median Class = 20
154−119
Median = 409.50 + ×20
82

= 409.50 + 8.5366
= 418.0366 (Note: This Final answer should lie within the Median Class)
Illustration 8: Median - Reverse Working
Find the missing frequency from the following data, given that the median mark is 23.

Mark 0-10 10-20 20-30 30-40 40-50

No. of students 5 8 ? 6 3

Solution: Let the missing frequency by f

Exclusive Classes Frequency (fi) Less than Cum.

(1) (2) Freq.(3)

0-10 5 5

10-20 8 13 m
 20-30 f 13+f Median Class

30-40 6 19+f

40-50 3 22+f

Total
𝑁
−𝑚
2
Median = 𝐿 + ×𝑐
𝑓
𝑁
= 22 + f 2
2

Where, L is the Lower Class Limit of the Median Class = 20

m is the cumulative Frequency of the Class preceeding the Median Class= 13
f is the frequency of the Median Class = f
c is the Class Magnitude of the Median Class = 10
Therefore,
22+𝑓
−13
2
23 = 20 + × 10
𝑓
22 + 𝑓−26
3= ×5
𝑓

3f = 5f – 20 ; 2f = 20 ; f = 10 ; The missing frequency is 10.

Illustration 9: Positional Averages - Quartiles, Percentiles and Deciles
Following are the wages of the labourers: 82, 56, 90, 50,120, 75, 75, 80,180, 65. Find Qi, D6
and P82.
Solution:

Step 1: Arranging the wages in an ascending Step 1: Arrange the wages in an ascending
order, we get 50, 56, 65, 75, 75, 80, 82, 90, order
120, 180.
Step 2: Deciles divide the entire distribution
Step 2: into 10 equal parts, Decile 6 has 60% of the
items less than it.
The Three Quartiles divide the entire
distribution into 4 quarters. Quartile 1 has l/4th ∴ D6 = (10 + 1) × 6 th value 10
10
of the Total Values less than it.
th
(𝑛+1) = 6.60 value
∴ 𝑄1 = 𝑡ℎ value
4 = 6th Value + 0.60 (difference between the 7th
(10+1) & 6th values)
= th value
4
= (80 + 0.60 × 2) = 81.20
= 2.75th value
= 2nd value+ 0.75(difference between the 3 rd &
2nd values)
= [56+0.75+x(65 - 56)] = 62.75
Step 1: Arrange the wages in ascending order.
Step 2: Percentiles divide the distribution into 100 equal parts. The 82 nd Percentile has 82%
of the Total Items less than its
value.
82
P82 = (10 + 1) × th value
100
= 9.02th value = 9th value + 0.02 x difference between the 10th and the 9th values
= (120 + 0.02 x 10) = 120.20

Illustration 10: Positional Averages - Quartiles, Percentiles and Deciles Following distribution
relates to the distribution of monthly wages of 100 workers.

Wages in Rs. Less than More than

500 - 699 700-899 900- 1099 1100-1499
500 1500

No. of
5 23 29 27 10 6
Workers:

Compute Q3, D7 and P23.

Solution; Converting the above classes into exclusive Classes we get,

Exclusive Less than Cum.

Frequency (fj) (2)
Classes (1) Freq. (3)

Less than 499.50 5 5

499.50-699.50 23 28

699.50-899.50 29 57 m

Quartile
899.50-1099.50 27 84
Class

1099.50-1499.50 10 94

More than
6 100
1499.50

Total 100
 𝑛𝑁
−𝑚 𝑛𝑁
4
𝑄𝑛 = 𝐿 + ×𝑐 10
−𝑚
𝔨 𝐷𝑛 = 𝐿 + ×𝑐
𝑓
𝑛𝑁 3(100) 300
= = = 75 𝑛𝑁 7(100) 700
4 4 4 𝐹𝑜𝑟 𝐷7− = = = 70
10 10 10
th
Since the 75 item lies in the class interval th
899.50-1099.50, the same is the Quartile's Since the 70 item lies in the class interval
Class. Note: The Quartile Class 899.50- 899.50-1099.50, the same is the Decile
Class. Note: The Decile class 899.50-
1099.50 contains items between 57 and 84 L
1099.50 contains items between 57 and 84.
is the Lower Class Limit of the Quartile Class
= 899.50 m is the cumulative Frequency of the L is the Lower Class Limit of the Decile Class
Class preceeding the Quartile Class= 57 = 899.50 m is the cumulative Frequency of
the Class preceeding the Decile Class= 57
f is the frequency of the Quartile Class = 27 c
is the Class Magnitude of the Quartile Class = f is the frequency of the Decile Class = 27 c is
200 the Class Magnitude of the Decile Class =
75−57 200
∴ Q3 = 899.50 + × 200 = 𝑅𝑠.1032.83
27 70−57
𝐷7 = 899.50 + × 200 = 𝑅𝑠.995.7963
27

𝑛𝑁
−𝑚
𝑃𝑛 = 𝐿 + 100 ×𝐶
𝑓
𝑛𝑁 23(100)
For 𝑃23 − = = 23
100 100

Since the 23rd item lies in the class 499.50-699.50, the same is chosen as the Percentile
Class.
L is the Lower Class Limit of the Decile Class = 499.50. Note: The items 6 to 28 lies in the
Percentile class 499.50-699.50 m is the cumulative Frequency of the Class preceeding the
Decile C!ass= 5 f is the frequency of the Decile Class = 23 c is the Class Magnitude of the
Decile Class = 200
23−5
P23 = [499.50 + × 200] = 𝑅𝑠. 656.02
23

Illustration 11: Computation of Mode Compute mode for the following distribution -

Class
350-369 370-389 390-409 410-429 430-449 450-469 470-489
Intervals

Frequency 23 38 58 82 65 31 11

Solution: Converting the above classes to Exclusive classes we get -

Computation of Mode

Class Intervals Frequency

349.50 -369.50 23

369.50 - 389.50 38

389.50 - 409.50 58(f0)

409.50 - 429.50 82(f1)

 429.50 - 449.50 65(f2)

449.50 - 469.50 31

469,50 - 489.50 11

Going through the frequency column, we note that the highest frequency i.e. f x is 82. Hence,
f0 = 58 and f2 = 65. Also the modal class i.e. the class against the highest frequency f 0 is 410
- 429.
𝑓1 −𝔨0
Mo = 𝐿 + ×𝑐
2𝑓1 −𝑓0 −𝑓2

Where L is the LCB of the Modal Class = 409.50 fi = Frequency of the Modal Class = 82 f 0 =
Frequency of the Class preceeding the Modal Class = 58 f2 = Frequency of the Class
succeeding the Modal Class = 65 c = Equal Class Intervals = 20
82−58
Mo = 409.5 + × 20 = 421.21
(2×82)−58−65

Note: The Final Answer should fall within the Modal Class. Here 421,21 belongs to the
modal class 410 - 429.
Illustration 12: Computation of Mode
For a moderately skewed distribution of marks in statistics for a group of 200 students, the
mean mark and median mark were found to be 55.60 and 52.40. What is the modal mark?
Solution:
For an moderately assymetrical distribution, the following empirical relationship exists
between Mean, Median and Mode . Mean - Mode = 3(Mean - Median), Therefore, Mode = 3
Median - 2 Mean.
Mean = 55.60 and Median = 52.40, we get the modal mark as
Mode = 3 × Median - 2 × Mean
= 3 × 52.40 - 2 × 55.60 = 46.
Illustration 13: Computation of Mean, Median and Mode
Find the mean and median from the following data:

Marks: Less than 10 Less than 20 Less than 30 Less than 40 Less than 50

No. of
5 13 23 27 30
Students:

Also compute the mode using the approximate relationship between mean, median and
mode.
Solution:
Computation of Mean Marks for 30 Students

Marks Class Cumulative

No. of Students (f) Mid - Value (x) fx
Interval Frequency
(2) (3) (4) = (2) x (3)
(1) (5)

0-10 5 5 25 5
 10-20 13 - 5 = 8 15 120 13

20-30 23 - 13 = 10 25 250 23

30-40 27 - 23 = 4 35 140 27

40-50 30 - 27 = 3 45 135 . 30

Total ∑f = N =30 - 670

∑𝑓𝑥 670
(a) The Mean Mark is given by 𝑥̅ = = = 22.33
𝑁 30

(b) Median:
𝑁
−𝑚
2
Median = ×𝑐
𝑓
𝑁 30
= = 15, Since items 13 to 23 Iie in the class 20‐30, the same is the Median CIass.
2 2

𝑚 = 13, 𝑓 = 10, 𝑐 = 10
15−13
Median = 20 + × 10 = 22
10

(c) Mode = 3 Median - 2 Mean

Mode = 3 × 22 - [2 × 22.33] = 21.34
Illustration 14: Computation of Mode, Median of related series
(a) If y = 2 +1.50x and mode of x is 15, what is the mode of y?
(b) If y = 25X -10, the mean of x is 20, what is the mean of y?
Solution:
The Mean, Median and Mode are affected both by origin (Multiplication / Division) and by
Scale (Addition / Subtraction). Therefore,
1- Ymode = 2 + 1.50 X [Xmode]
Ymode = 2 + (1.50 × 15)= 24.50
2. Ymean = 25[Xmean] – 10
Ymean = 25[20] - 10 = 490.
Illustration 15: Computation of Geometric Mean (a) Find the GM of 3, 6 and 12.
Solution: As given x1 = 3, x2 = 6, x3 = 12, n= 3.
For a discrete series If x1, x2/ x,…….. xn are n values of a variate x, none of them being zero,
then the geometric mean G is defined as G = (x1 × x2 × x3…………….. × xn)1/n
∴ G = (3 × 6 × 12)1/3 = (63)1/3 = 6.
(b) Find the GM for the following distribution:

x: 2 4 8 16

f: 2 3 3 2

Solution:
G = (xf1 × x2f2 × x3f3 × ×4f4)1/N
= (22 × 43 × 83 × 162)1/10
= (2)2'50 = 4 √2 =5.66
G can also be found out using the following procedure,
∑𝑛
𝑖=1 𝑓𝑖 𝑙𝑜𝑔𝑥𝑖
G = Antilog[ ], where N= ∑fi
𝑁

X f Log x f Logx

2 2 0.3010 0.6020

4 3 0.6020 1.8062

8 3 0.9030 2.7093

16 2 1.2041 2.4082

∑f = 10 7.5258
7.5258
G = Antilog = Anti Iog0.7526 = 5.657
10

Illustration 16: Computation of Harmonic Mean Answer the following questions -

(a) Find the HM for 4,6 and 10. (b) Find the HM for the following data
Solution: x 2 4 8 16 f 2 3 3 2
𝑛 𝑁
HM= Solution: HN = 𝔣
∑(1/𝑥𝑖 ) 𝛴 𝑖
𝑥𝑖
3
HM = 1 1 1 10
+ +
4 6 10 HM = 2 3 3 2 = 4.36
+ + +
2 4 8 12
3
=
0.25+0.17+0.10

= 5.77

(c) Compute the Average Speed for the (d) Compute AM, GM, and HM for the
Following Data - numbers 6, 8,12, 36. Solution:
6+8+12+36
Speed AM = = 15.50
4
25 40 80 60
(km/h) GM = (6 × 8 × 12 × 36)1/4
4
Distance 𝐻ℕ = 1 1 1 1 = 9.93
+ + +
20 30 30 20 6 8 12 36
(Kms)

Solution: In case of Ratios such as speed,

Harmonic Mean is the best Average.
Taking Speed as the Variate and the
respective distance travelled as the Frequency,
we obtain the HM as follows –

Speed (x) Distance (f) 1/x f(1/x)

25 20 0.0400 0.8000
 40 30 0.0250 0.7500

80 30 0.0125 0.3750

60 20 0.0167 0.3333

100 2.2583
𝑁 100
HM= 𝔣 = = 44.2804 Note: The distance
𝛴 𝑖 2.2583
𝑥𝑖 of 100

kms has been covered at an Average Speed of

44.2804 kms/hr

Illustration 17: Computation of Weighted Averages

Find the weighted AM and weighted HM of first n natural numbers, the weights being equal
to the squares of the Corresponding numbers.
Solution:
X 1 2 3 ……… n
………….
W 12 22 32 n2

𝛴𝑤𝑖 𝑥𝑖
Weighted AM =
𝛴𝑤𝑖
2
𝑛(𝑛2 + 1)
1 × 12 + 2 × 22 + 3 × 32 … ⋯ ⋅⋅ 𝑛 × 𝑛 2 13 + 23 + 33 + ⋯ … … . 𝑛 3 [ ]
2
= = =
12 + 22 + 32+ ⋯..+𝑛 2 12 + 22 + 32 + ⋯ . . +𝑛 2 𝑛(𝑛 + 1)(2𝑛 + 1)
6
3𝑛(n + 1)
=
2(2𝑛 + 1)

𝛴𝑤𝑖
Weighted HM = 𝛴𝑤𝑖
𝛴( )
𝑥𝑖

12 + 22 + 32 + ⋯ … … .⋅ 𝑟1𝑧 12 + 22 + 3𝑍 + ⋯ . .⋅ +𝑛 2
= =
12 22 32 𝑛2 1 + 2 + 3 + ⋯ … . +𝑛
+ +3 + ⋯ …
1 2 𝑛
𝑛(𝑛 + 1)(2𝑛 + 1)
6 2𝑛 + 1
= =
𝑛(𝑛 + 1) 3
2
Illustration 18: Relation between AM. GM and HM
Answer the following Questions -
Given two positive numbers a and b, prove The AM and GM for two observations are 5
that ab=G2. Does the result hold for any set of and 4 respectively.
observations? (G - Geometric Mean)
Find the two observations.
Solution: For two positive numbers a and b,
Solution: If a and b are two positive
we have,
observations then as given,
𝑎+𝑏
AM = 5, ∴ =5
𝑎+𝑏 2
𝐴= 𝐺 = √𝑎𝑏
2 a + b = 10 ... (1)
2 2𝑎𝑏
And 𝐻 = 1 1 = GM = √𝑎𝑏 = 4 ; ab = 16
+ 𝑎+𝑏
𝑎 𝑏
𝑎+𝑏 2𝑎𝑏 ∴ b = 16/a ... (2)
Thus A × 𝐻 = × = 𝑎𝑏 = 𝐺 2
2 𝑎+b
Subs (2) in (1) you get,
No, this results holds for only two positive 16
observations or if the observations are in a + 𝑎 = 10
arithmetical progression.
a2 - 10a + 16 = 0
a2 - 8a -2a + 16 = 0
(a - 8)(a-2) = 0
a = 8;2
Sub. in (1) we get
Where a = 8, b=2
Where a = 2, b=8
ͬ∴Two observations are 8 and 2.