Central Tendency and Its Measures 39

CHAPTER III 3.1 Arithmetic Mean :

CENTRAL TENDENCY AND ITS MEASURES Arithmetic mean of a set of observations is their sum divided
by the number of observations.
The individual observations of a distribution or a data set are The arithmetic mean may be of two types :
found to have a general tendency to cluster around a certain point, (a) Simple Arithmetic mean
somewhere at the center of the distribution. For example, if we (b) Weighted Arithmetic mean
observe the distribution of the height of a group of students in a class,
(a) Simple Arithmetic Mean : The arithmetic mean x of n
the height of most of the students are close to a certain central value. x 1 + x 2 + ...... + x n 1 n
This tendency of the observations of a distribution to cluster or observations x1, x2, ....., xn is given by x = =  xi
n n i =1
concentrate around the center of the distribution is called central In case of frequency distribution; let f1, f2, ...., fn are the
tendency and its numerical measures are known as the measures of frequencies of x1, x2, .., xn respectively, arithmetic mean is obtained as
n
central tendency.  fi xi
f x + f x + ...... + f n x n i =1 1 n
Different measures of central tendency are : x= 1 1 2 2 = =  fi xi
f1 + f 2 + ........ + f n n N i =1
1. Mean  fi
i =1
(a) Arithmetic mean (AM)
 n 
(b) Geometric mean (GM) where ;  f i = N
 i =1 
(c) Harmonic mean (HM) (b) Weighted Arithmetic Mean : In practice all values of a series
2. Median and quantiles may not carry equal weight or importance. For example, if we want to
3. Mode have an idea of the change in cost of living of a certain group of
The main purpose of measuring central tendency of a people, the simple mean of prices of the commodities consumed by
distribution is to determine such a value, which can be considered to them will not do, since all the commodities are not equally important,
be a representative one. An ideal measure of central tendency should, e.g., rice, sugar and wheat are more important than confectionery
therefore, have the following characteristics : items, coffee, tea, etc.
• It should be rigidly defined Let w1, w2, ............. , wn be the weights attached to the item x1,
• It should be based on all the observations. x2, ..........., xn respectively, the arithmetic mean is computed as-
n
• It should be readily comprehensible and easy to calculate.  wixi
x 1 w 1 + x 2 w 2 ........... + x n w n i =1
x= =
• It should be suitable for further algebraic treatment. w 1 + w 2 + ........... + w n W
• It should be least affected by sampling fluctuations. n
where W =  w i
i =1
40 Methods of Statistics Central Tendency and Its Measures 41

Properties of Arithmetic Mean : n n

 f i ( x i − a ) =  f i ( x i − x ) + ( x − a )
2 2
1. Sum of the deviations of the values of the variable from its i =1 i =1
n n n
arithmetic mean is zero. i.e.,  f i ( x i − x ) = 0 =  f i ( x i − x ) 2 + n ( x − a ) 2 + 2( x − a )  f i ( x i − x )
i =1 i =1 i =1
Proof : n
or,  f i ( x i − a ) 2 = f i ( x i − x ) 2 + n ( x − a ) 2 ; [ f i ( x i − x ) = 0]
n n n i =1
 f i (x i − x) =  f i x i −  f i x
i =1 i =1 i =1 n
  f i ( x i − a ) 2  f i ( x i − x ) 2 ; [Since (x-a)2 is a positive quantity]
n i =1
= Nx − x  f i = Nx − xN = 0
i =1 n n
i.e.,  f i ( x i − x ) 2   f i ( x i − a ) 2 Proved.
Proved. i =1 i =1
2. Arithmetic mean is dependent on change of origin and scale.
4. Mean of Composite Series :
Proof : Let, x1, x2, ........, xn be the values of a variable x. Let us If x i , (i = 1, 2, .... , k) are the means of k-component series of
change the origin to an arbitrary value 'a' and change the scale by sizes ni (i = 1, ......, k) respectively, then the mean x of the composite
x −a series can be obtained by the formula -
dividing by 'h'. The values of the new variable are, u i = i
h
n x + n 2 x 2 + ......... + n k x k
x= 1 1
1 n 1 n  x −a  n 1 + n 1 + ........... + n k
Now, u =  f i u i =  f i  i 
N i =1 N i =1  h 
Proof:
1 1 n 1 1n n  Let x11, x12, ........., x 1n1 be n1 members of the first series,
=  f i (x i − a ) =   f i x i −  f i a
N h i =1 N h i =1 i =1 
x21, x22, ........., x 2n 2 be n2 members of the 2nd series ;
=
1 1
Nx − Na = 1 (x − a) ..................................................................................
Nh h
or, hu = x − a ..................................................................................
 x = a + hu Hence proved. x k1 , x k 2 ,.........., x kn k be nk members of the kth series-

having means x 1 , x 2 , ...................., x k respectively.

3. The sum of the squares of the deviations of a set of values
from their arithmetic mean is the minimum. Then n1 + n2 + ........ + nk will be the size of the composite series
(x11, x12, ...... , x 1n1 ), (x21, x22, ...... , x 2n 2 ), ......, ( x k1 , x k 2 , ....., x kn k ).
Proof:
Let, x be the arithmetic mean of a set of observations x1, x2, ....., The mean, x of the composite series of size n1 + n2 + ........ + nk
xn with frequencies f1, f2, ............., fn respectively. Now sum of is given by
squares of deviations from an arbitrary value 'a' is
42 Methods of Statistics Central Tendency and Its Measures 43

( x 11 + x 12 + .... + x 1n ) + ( x 21 + x 22 + .... + x 2n ) + ...... + ( x k1 + x k 2 + .... + x kn ) Uses of Arithmetic Mean:

x= 1 2 k
n 1 + n 2 + .......... .. + n
k • It is widely used to calculate average age, average income,
k average price, average salary, average increment, average
 nixi
n x + n 2 x 2 + ............ + n k x k i =1 import and average consumption, etc.
= 1 1 =
n 1 + n 2 + ............ + n k k • It is used to establish the various theories and formulas of
 ni
i =1 Mathematics and also used as an aid in further statistical
1 k k analysis.
x =  n i x i ; where N =  n i Proved.
N i =1 i =1 • It is used in computation of index number.

Mean of first n Natural Numbers : Example 3.1 : The daily wages of a group of farm workers are shown
in the following frequency distribution.
First n natural number are 1, 2, ..........., n.
1 + 2 + ....... + n 1 n (n + 1) n + 1 Daily wages Number of Daily wages Number of
The mean, x = = = (Tk.) workers (Tk.) workers
n n 2 2
n +1 50-55 5 70-75 15
x =
2
55-60 10 75-80 7
Advantages of Arithmetic mean:
60-65 25 80-85 3
• It is rigidly defined.
65-70 35
• It is easy to calculate.
• It is based upon all the observations. Computation of arithmetic mean by direct and indirect method.
• It is suitable for further algebraic treatment. Direct Method :
• It is less affected by sampling fluctuations. Number of
Daily wages
workers Mid value xi fixi
Disadvantages of Arithmetic mean: (Tk.)
fi
• It is affected very much by extreme values. 50-55 5 52.5 262.5
• It cannot be calculated if the extreme class is open. 55-60 10 57.5 575.0
60-65 25 62.5 1562.5
• It is not suitable for extremely skewed distribution.
65-70 35 67.5 2362.5
• It cannot be used if we are dealing with qualitative 70-75 15 72.5 1087.5
characteristics; such as intelligence, honesty, beauty, etc. 75-80 7 77.5 542.5
80-85 3 82.5 247.5
• It cannot be obtained if a single observation is missing or
100 6640.0
lost.
44 Methods of Statistics Central Tendency and Its Measures 45

1 1 1 n 
x= f i x i = (6640.0) = Tk. 66.40  GM = Anti log   log x i 
N 100  n i =1 
 Average daily wage is Tk. 66.40
In case of frequency distribution, when f1, f2, ...... , fn be the
Indirect Method : frequencies of x1, x2, .........., xn respectively, then

Daily Number of New variable

Mid value x − 67.5 n
wages workers
xi ui = i
fiui GM= N x 1f1 .x f22 . .......... x fnn ; where N =  f i
(Tk.) fi 5 i =1
50-55 5 52.5 -3 -15 1 n 
55-60 10 57.5 -2 -20 = Anti log   f i log x i 
 N i =1 
60-65 25 62.5 -1 -25
65-70 35 67.5 0 0 Advantages of Geometric Mean :
70-75 15 72.5 1 15
75-80 7 77.5 2 14 • It is rigidly defined.
80-85 3 82.5 3 9 • It is based upon all the observations.
100 -22 • It is not affected much by sampling fluctuations.
xi − a • It is suitable for further algebrical treatments.
New variable, ui = ; where, a = 67.5 and h = 5
h • In measuring rate of change it is the most suitable average.
1 1
Now, u = f i u i = (−22) = -0.22 Disadvantages of Geometric Mean :
N 100
 x = a + hu = 67.5 + 5(-0.22) = 66.40 • It cannot be computed where there is any negative or zero
values in the series.
 Average daily wage is Tk. 66.40
• It is not easy to understand and to calculate for persons
3.2. Geometric Mean (GM) : having very weak mathematical skills.
Geometric mean of a set of n non-zero positive observations is • It cannot be computed when the extreme classes of the
the nth root of their product. The GM of n non-zero positive values frequency distribution are open.
x1, x2, ............, xn of a variable x is given by • The value of GM may not be found in the series.
1
Uses of Geometric Mean :
x 1 .x 2 . ......... x n = (x 1 .x 2 . ............. . x n )
n
GM =n
• Geometric mean is used to find the average of ratios, rate of
1
1 n population growth, rate of interest, average of percentages.
log (GM) = log (x 1 .x 2 . ........... x n ) =
n
 log x i
n i =1 • It is used in the construction of index numbers.
46 Methods of Statistics Central Tendency and Its Measures 47

Example 3.2 : 3.3 Harmonic Mean (HM) :

Rate of increase of yield of a new wheat variety compared with a Harmonic mean of a set of non-zero observations is the
local variety in 10 selected agricultural farms are given below - reciprocal of the arithmetic mean of the reciprocals of the given
values. Harmonic mean of n non-zero observations x1, x2, ......... , xn is
Rate of increase of yield (%) Number of farm
given by
0-5 1
1 1 n
5-10 2 HM = = =
1 1 1 1  1 n 1 n 1
10-15 4  + + ......... +   
n  x1 x 2 xn  n i =1 x i i =1 x i
15-20 2
1 1 n 1
20-25 1  = 
HM n i =1 x i
If f1, f2, ........, fn are respectively the frequencies of x1, x2, ......, xn
For computation of geometric mean, we construct the following table non-zero observations; then
Rate of change of Frequency Mid 1 N
log xi filogxi HM = =
yield (%) fi value xi 1 n n
 (f i / x i )  (f i / x i )
0-5 1 2.5 0.39794 0.39794 N i =1 i =1

1 1 n
5-10 2 7.5 0.87506 1.75012  =  (f i / x i )
HM N i =1
10-15 4 12.5 1.09691 4.38764
15-20 2 17.5 1.24304 2.48608 Advantages of Harmonic Mean :
20-25 1 22.5 1.35218 1.35218 • It is rigidly defined.
• It is based upon all the observations.
filogxi
fi = N =10 • Sampling fluctuation is less.
= 10.37396
• It is not affected much by extreme values
1 
GM = Anti log  f i log x i  Disadvantages of Harmonic Mean :
N 
1  • It cannot be computed where there is any zero values in the
= Anti log  (10.373996) = Anti log (1.037396)
10  series.
= 10.9 (Approx.) • It is not easily understood and difficult to compute.
• It is very complex for further algebrical treatments.
 The average rate of change of yield of the new variety of wheat
• It cannot be computed if the extreme classes of the
is 10.9% frequency distribution are open.
48 Methods of Statistics Central Tendency and Its Measures 49

Uses of Harmonic Mean : Proof:

• The harmonic mean is used when observations are made in Let the two non-zero positive observations be x1 and x2.
terms of work done per hour, speeds (kilometers covered per x1 + x1
By definition, A = ; G = x 1 .x 2 = ( x 1 .x 2 ) ½
2
hour), quantity of things purchased per taka etc.
1 1 2x 1 x 2
and H = = =
Example 3.3 : 
1 1 1  x 1 + x 2 x 1 + x2
 + 
The frequency distribution of profit per share of 10 companies 2  x1 x 2  2x 1 x 2
are given below - i) Since any square quantity is always non-negative,

Profit per share (Tk). 0-5 5-10 10-15 15-20 20-25  ( x1 − x 2 )2  0

No. of companies 1 2 4 2 1 or, x1 + x2 - 2 x1x 2  0
x1 + x 2
To calculate harmonic mean, we construct the following table : or,  x1x 2 ................................ (1)
2
Profit per share (Tk.) Frequency fi Mid value xi f i / xi  A G ................................ (2)
Again from equation (1)
0-5 1 2.5 0.4000
x 1 + x2  2 x 1 x 2
5-10 2 7.5 0.2667
2 x1x 2
10-15 4 12.5 0.3200 or, 2 x 1 x 2  x 1 + x 2 or, 1
x1 + x 2
15-20 2 17.5 0.1143
Multiplying both sides by x1 x 2 we get
20-25 1 22.5 0.0444
2 x1x1 x 2 x 2
Total 10 1.1454  x1x 2
x1 + x 2
N 10 2x 1 x 2
HM = = = 8.73 or,  x1x 2
f i / x i 1.1454 x1 + x 2
 The average profit per share is Tk. 8.73 or, H  G i.e., G  H ................................ (3)
Relationship among AM, GM and HM: From (2) & (3) it follows that A  G  H. Proved.
◙ For two non-zero positive observations: x 1 + x 2 2x 1 x 2
(ii) A.H = x = x1x 2
i) A  G  H 2 x1 + x 2
ii) AH = G2; where A = Arithmetic mean, = ( x1x 2 )2 =G 2

G = Geometric mean and H = Harmonic mean. A.H = G2 Proved.

50 Methods of Statistics Central Tendency and Its Measures 51

◙ For n non-zero positive observations: 1 1 1

As x1, x2, ....... , xn are non-zero positive, , , ......., are
A G  H x1 x 2 xn
also non-zero positive and in this case also A  G
Proof. :
1 1 1 1  1 1 1
Let x1, x2, x3, ..........., xn be n non-zero positive observations. that is,  + + ....... +   n x x .......x
n  x1 x 2 xn  x1 x 2 xn
Let, di = xi - A
1 1
xi = A + di ............... (1) or, 
1 H n x 1 x 2 . ........ x n
By definition, G = (x 1 .x 2 ...............x n )
n
1 1
or,   G  H ................................ (3)
1 n H G
 logG =  log x i From (2) and (3) it follows that
n i =1
1 n A  G H Proved.
=  log( A + d i ) ; From (1)
n i =1 3.4 Median :
1 n 1 n
=  log A(1 + d i / A ) =  log log A + log (1 + d i / A ) The median of a distribution is the value of the variable which
n i =1 n i =1
divides the distribution into two equal parts if arranged in order of
n n
 log A +  log ( 1 + d i / A )
1 1
= magnitude. Median is the value such that the number of observations
n i =1 n i =1
above it is equal to the number of observations below it. Thus median
1 n d i (d i / A ) 2 1
= log A +  { − . }; is the middlemost value of an ordered set and as such a positional
n i =1 A 2 (1 + d i / A) 2
average.
[For 0    1]
Expanding log (1 + d i / A ) in ascending power of (d i / A ) by
In case of ungrouped data, when the number of observations, n is
n +1
Taylor's expansion method and assuming the higher order terms odd, th observation in the series will be the median. Again,
negligibly small, we get 2
1 1 1 (d i / A ) 2 1 when the number of observations, n is even, median will be the
log G = log A +  di −  x
nA n 2 (1 + d i / A)2 arithmetic mean of
n n 
th and  + 1th observation in the series.
Since, di = (xi-A) = 0; as A = Arithmetic mean 2 2 

and (d i / A )2 .
1 For computing median from frequency distribution we first need
is a positive quantity
(1 + d i / A ) 2
to identify the median class (class which contains the median). If the
 log G = log A + 0 - a positive quantity total frequency is N, the class having cumulative frequency equal to,
 log A = log G + a positive quantity or immediately higher than N/2 will be the median class.
 log A  log G For frequency distribution, the formula for computing the
AG ............................................... (2) median is
52 Methods of Statistics Central Tendency and Its Measures 53

N 1) Quartiles (2) Deciles and (3) Percentiles.

− Fm
'

Me = Lm + 2 x h
fm 3 quartiles : Qi (i = 1, 2, 3); devide the whole distribution
into four equal parts
where,
9 Deciles : Dj (j = 1, 2,..., 9); devide the whole distribution
Lm = lower limit of the median class.
into 10 equal parts.
N = total frequency
99 Percentiles : Pk (k = 1, 2, ..., 99); devide the whole
fm = frequency of the median class
'
distribution into 100 equal parts.
Fm = cumulative frequency of the pre-median class
Computation of quantiles from frequency distribution is very
h = length of median class. much similar to that of median. We first need to identify the
Advantages of Median : corresponding quantile class. The classes having cumulative
• It is rigidly defined. frequencies equal to or immediately higher than iN/4, jN/10 and
• It is easily understood and easy to compute. kN/100 are respectively the ith quartile class, the jth decile class and
the kth percentile class.
• It is not influenced by extreme items.
• It can be calculated for distribution with opened classes. For frequency distributions the quantiles are computed as -
• It can be used in defining the median of attributes. iN
− Fi'
Qi = Li + 4 x h ; i = 1, 2, 3
Disadvantages of Median : fi
• It is not based upon all the observations. jN
− Fj'
• It is not suitable for further algebraic treatment. Dj = Lj + 10 x h ; j = 1, 2, .............., 9
fj
• It is affected much by the sampling fluctuation.
kN
− Fk'
Uses of Median : 100
Pk = Lk + x h ; k = 1, 2, .............., 99
• It is used in case of both quantitative and qualitative data. fk
• It is used for calculating the typical value in problem i, j, k indicate the order of quartiles, deciles and percentiles
concerning wages, distribution of wealth etc.
respectively; Fi' , Fj' and Fk' are respectively the cumulative frequencies
3.5 Quantiles : Quantiles also are some positional or location of class preceding the ith quartile, jth decile and kth percentile
measures of the distribution. Quantiles are those values in a series, classes; h is the corresponding class interval.
which divide the whole distribution into a number of equal parts when
the series is arranged in order of magnitude of observations. The It may be mentioned that
following are the quantiles that are used in Statistics - Q2 = D5 = P50 = Me ; Q1 = P25 ; Q3 = P75 ; D6 = P60 etc.
54 Methods of Statistics Central Tendency and Its Measures 55

Graphical Location of Median and Quantiles : frequent value of a distribution. In the case of frequency distribution,
Median, quartiles, deciles and percentiles can be located from mode is given by
ogive; the necessary steps are briefly discussed below : f 0 − f1
Mo = L + x h
i) An ogive is drawn and the position in the Y-axis are marked 2f 0 − f1 − f 2
for different partition values (e.g., N/2 for median, N/4 for 1st
quartile, 4N/10 for 4th decile etc.) where, L = lower limit of modal class
ii) From the corresponding points in the Y-axis, a line parallel f0 = frequency of modal class
to the X-axis is drawn which intersects the ogive at certain f1 = frequency of pre-modal class
point. f2 = frequency of post-modal class
iii) From the corresponding point of intersection mentioned [The class which corresponds to the maximum frequency is the
above, a perpendicular is drawn on the X-axis; the foot of model class]
the perpendicular is the desired partition value. The whole Advantages of Mode :
process is illustrated in figure 3.1 below : • It is easy to understand and easy to calculate
N • It is not affected by extreme values.
• It can be located graphically.
Cumulative Frequency

3N
/4 Disadvantages of Mode :
• It is not rigidly defined - a distribution may have more than
one mode.
N
/2
• It is not based upon all the observations.
• It is not suitable for further algebraic treatment.
N
/4 Uses of Mode :
• Mode is used to find the ideal size, e.g., in business
0 Q1 Me Q3 forecasting, Meteorological forecast on weather condition, in
Upper limits of class interval the manufacture of ready-made garments, shoes, etc.

Fig. 3.1: Location of median and quantiles. Graphical Location of Mode :

Mode can graphically located in two ways :
3.6. Mode (Mo) :
a) Using frequency curve.
Mode of the distribution is that value of the variate for which the
b) Using the histogram.
frequency is the maximum. In other words, mode is the highest
56 Methods of Statistics Central Tendency and Its Measures 57

a) From the peak of the frequency curve, a perpendicular is drawn

on the X-axis; the foot of the perpendicular indicates the mode
A B
(shown in figure 3.2) :
D

Frequency
C
Frequency

Mo
Class intervals
Mo
Mid-values of class intervals. Fig. 3.3: Location of mode from the histogram.
Fig. 3.2: Location of mode from frequency curve

b) Mode can be located more accurately from the histogram; the

steps are the following :

i) The rectangles corresponding to the modal group, the pre-modal

group and the post-modal group are considered. A straight line is
drawn connecting the left vertical point (say A) of the modal
group rectangle and the left vertical point (say D) of the post
modal group rectangle. Similarly the right vertical point (say B)
of the modal group rectangle and the right vertical point (say C)
of the pre-modal group rectangle are connected.

ii) From the point of intersection of AD and BC, a perpendicular is

drawn on the X-axis; the foot of the perpendicular indicates
mode.
58 Methods of Statistics Central Tendency and Its Measures 59

Comparison Among the Measures of Central Tendency Example 3.4 :

Find the median, lower and upper quartiles, 4th decile, 70th
Criteria AM GM HM Me Mo
percentile and mode for the following distribution :
Not Class: 50-60 60-70 70-80 80-90 90-100 100-110 110 and
Rigidly Rigidly Rigidly Rigidly
Definition rigidly over
defined defined defined defined
defined Frequ- 5 9 13 20 19 9 5
Values must Values ency:
Data No No No
be nonzero must be i) Draw the ogive and locate Me, Q3, D4 and P70
restriction restriction restriction restriction
and positive nonzero ii) Draw the Histogram and locate mode
Compu- Slightly Slightly Solution :
Easy Easy Easy
tation difficult difficult
Class Frequency c.f.
Based 50-60 5 5
upon all
Yes Yes Yes No No 60-70 9 14
observa-
tions 70-80 13 27
Effect of 80-90 20 47
Less Less Not Not
extreme Not affected
values
affected affected affected affected 90-100 19 66
100-110 9 75
Sampling
fluctua- Little Little Little Much Much 110 and over 5 80
tion N = 80
Graphical Not Not Not
Possible Possible Here, N = 80
location possible possible possible
Further Computation of Median :
Not Not Not
algebrical Possible Possible
possible possible possible N 80
treatment = = 40th observation lies in the class (80-90)
2 2
 (80-90) is the median class
From the above comparison, it is clear that arithmetic mean is
the best measure of central tendency. N / 2 − Fm
' Lm = 80
 Me = Lm + x h N/2 = 40
fm
'
Fm = 27
40 − 27
= 80 + x 10 fm = 20
20
h = 10
= 86.5
60 Methods of Statistics Central Tendency and Its Measures 61

Computation of Quartiles : Computation of Percentiles :

N 80 70N 70(80)
= = 20th observation lies in the class (70-80) = = 56th observation lies in the class (90-100)
4 4 100 100
 (70-80) is the lower quartile (Q1) class  (90-100) is the 70th percentiles (P70) class
N / 4 − F1' L1 = 70 70N
− F70
' L70 = 90
 Q1 = L1 + x h N/4 = 20
f1  P70 = L70 + 100 x h 70N
= 56
F1' = 14 f 70 100
20 − 14
= 70 + x 10 fi = 13 56 − 47
'
F70 = 47
13
h = 10 = 90 + x 10
19 f70 = 19
= 74.62 (app.)
= 94.74 (app.) h = 10
3N 3(80)
Again, = = 60th observation lies in the class (90-100)
4 4
 (90-100) is the upper quartile (Q3) class Computation of Mode :
3N
Here (80-90) is the modal class because maximum frequency
− F3'
L3 = 90
(20) lies in that class
 Q3 = L3 + 4 x h 3N
= 60
f3 f 0 − f1
10  Mo = L + x h
F3' 2f 0 − f1 − f 2
60 − 47 = 47 L = 80
= 90 + x 10
19 f3 = 19 20 − 13 f0 = 20
= 80 + x 10 f1 = 13
= 96.84 h = 10 2 x 20 − 13 − 19
f2 = 19
= 88.75 h = 10
Computation of Deciles :
4 N 4(80)
= = 32th observation lies in (80-90)
10 4
 (80-90) is the 4th deciles (D4) class
4N L4 = 80
− F4'
 D4 = L4 + 10 x h 4N
= 32
f4 10
32 − 27 F4' = 27
= 80 + x 10
20 f4 = 20
= 82.50 h = 10
62 Methods of Statistics Central Tendency and Its Measures 63

(i) Graphical location of quantiles from ogive : Exercises

90 1. The rice yield (in kg) from a number of small plots are grouped
80 with common class interval of 2 kg. in the table below; the x
values are midvalues of the classes.
Cumulative Frequency

70
Yield(x) 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0
60 D4=83.0
No. of 4 15 20 47 63 88 59 35 15 18 10 5
50 plots (f)
Me=86.5
40
Compute arithmetic mean, geometric mean and harmonic mean
30
P70 =94.0 and verify the relationship A.M.  G. M.  H.M.
20
Q3 =97.0 2. The frequency distribution below gives the cost of production of
10 sugar in different holdings :
0 Cost No. of Cost No. of Cost No. of
60 70 80 90 100 110 120 (Tk.) holdings) (Tk.) holdings) (Tk.) holdings)
Upper limits of class interval 10-14 11 22-26 45 34-38 20
14-18 27 26-30 35 38-42 15
ii) Location of mode from histogram. 18-22 42 30-34 30

25

Compute :
20
(a) A. M., G. M. and H. M.
Mo =89.0 (b) Median, Q1, D4, Q3 and P80.
Frequency

15

(c) Mode
10
(d) Draw a histogram and locate the mode
5 (e) Draw a frequency curve and locate the mode.
(f) Draw an ogive and locate median, Q1, D4, Q3 and P80.
0

0 50 60 70 80 1 90 100 110 120 3. A set of 20 observations gives arithmetic mean 45 units and
Class interval
another set of 30 observations has arithmetic mean 60. The two
set are combined; find the arithmetic mean of the combined set.