CPP STATISTICS

` KKP / MYP CENTRE

Frequency Distribution:

Let us consider the marks obtained by 30 students in an exam. The data is as given below:

7, 7, 2, 5, 5, 3, 1, 3, 2, 1
7, 8, 9, 4, 1, 6, 7, 4, 4, 3
2, 9, 5, 1, 1, 6, 3, 7, 9, 1

This representation of data is called Raw Data. This data doesn’t furnish any useful information and is rather
confusing. A better way to express the data is as given below.

Marks (x) Number of

students (f)
1 6
2 3
3 4
4 3
5 3
6 2
7 5
8 1
9 3

The representation of data as above is known as ‘ungrouped frequency distribution’. Here, marks are called the
variable (x) and the number of students against the marks is known as the frequency (f) of the variable.

If the identity of the individuals about whom particular information is taken is not relevant, nor the order in which the
observations arise, then the first real step of condensation is to divide the observed range of variable into a suitable
number of class intervals. For example, in the above case, the data may be expressed as shown below.

Marks Number of Students

1–3 13
4–6 8
7–9 9

The representation of data as above is known as ‘grouped frequency distribution’ (inclusive type of
Classification).

Note:

(a) The classes of the type 1 – 3, 4 – 6, 7 – 9, etc., in which both the upper and lower limits are included are called
inclusive classes.

(b) The following points may be kept in mind for classification of data into grouped frequency.

 The classes should be clearly defined and should not lead to any ambiguity.
 The classes should be exhaustive i.e. each of the given values should be included in one of the classes.
 The classes should be mutually exclusive and non overlapping
 The classes should be of equal width. However, it cannot be rigidly followed.

If the classes are of varying width, the different class frequencies will not be comparable. The ratio of frequency by
the corresponding width of the class interval is called ‘frequency density’.

(c) The difference between the greatest and smallest observations is called the ‘range’.

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
Regd. Off.: 29A, ICES House, Kalu Sarai, Sarvapriya Vihar, New Delhi - 110 016. Ph: 011 - 2651 5949, 2656 9493, Fax: 2651 3942
If we deal with a continuous variable, it is not possible to arrange the data in the class intervals of above type. For
example, let us consider the distribution of age in years. If class intervals are 1 – 3, 4 – 6, 7 – 9…… then the
persons with ages between 6 and 7 years are not taken into consideration. In such cases, we form the class
intervals as shown below.

Age (x) Number of people (f)

1–3 9
3–6 10
6–9 8
9 – 12 3

This form of frequency distribution is known as ‘continuous frequency distribution’ (or) ‘exclusive type of
classification’.

Note: It should be understood that in the above classes, the upper limit of each class are excluded from the
respective classes.

2. Graphical representation of frequency distribution:

Histograms: In drawing the histogram of a given continuous frequency distribution, we first mark off along the x-
axis, all the class intervals on suitable scale. On each class intervals, draw rectangles with heights proportional to
the frequency of the corresponding class interval. The diagram of continuous rectangles so obtained is called
Histogram.

Note: To draw the histograms for an ungrouped frequency distribution of a variable, we shall have to assume that
h h
the frequency corresponding to the variable (x) is spread over the interval x  to x  where h is the jump from
2 2
one value to next. For example, consider the following frequency distribution.

Marks (x) Frequency (f)

1 3
2 7
3 10
4 5
5 4
h
Here, jump h  1   0.5
2

3. Frequency Polygon:

For an ungrouped distribution, the frequency polygon is

obtained by plotting points with abscissa as variable
values and ordinate has corresponding frequencies and
joining the plotted points by means of straight lines.

In case of grouped frequency distribution, the abscissas

of points are mid-values of class intervals.

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
4. Average (or) Measure of central tendency:

(i) Arithmetic Mean

(a) The Arithmetic Mean of a set of observations x1,x 2 ,...xn is denoted by x and it is defined as
x1  x 2  ....x n
x
n
1
x  i
n n i1

x f ,
n
1
(b) In case of ungrouped frequency distribution xi | fi , i  1,2,3,....,n the Arithmetic Mean is given by x  i i
N i1

f
n
where N  i
i1

 f
n n
1
(c) In case of continuous frequency distribution, the arithmetic mean, x  xi fi , where N  i and xi is the
N i1 i1
mid value of class interval.

Example (1): Find Arithmetic Mean of following frequency distribution:

x 1 : 2 3 4 5 6 7
f 5 : 9 12 17 14 10 6
 fi xi 299
Solution: x 
 fi 73

Example (2): Find the Arithmetic Mean of the following table:

Marks : 0-10 10-20 20-30 30-40 40-50 50-60

No. of Students : 12 18 27 20 17 6

Solution:

Marks No. of Students Mid values of fi xi

fi  Class intervals
 xi 
0-10 12 5 60
10-20 18 15 270
20-30 27 25 675
30-40 20 35 700
40-50 17 45 765
50-60 6 55 330
N  100
 f x  2800
i i

2800
Arithmetic Mean x   28
100
Note:

(a) Shifting of origin:

x f
n
1
The Arithmetic Mean x  i i
N i1
Let di  xi  A, where A is a fixed constant for all i  1,2....n . Then xi  di  A
n n n n
1 1 A 1
 A.M., x 
N 
i1
fi  di  A  
N 
i 1
fidi 
N i1
fi  A 
N fd
i 1
i i

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
Example (3): Find the Arithmetic Mean of the data given in Example (2) by shifting the origin.

Solution:

Marks No. of Students Mid values of di  x i  A fi di

fi  Class intervals
 xi 
0-10 12 5 – 20 – 240
10-20 18 15 – 10 – 180
20-30 27 25 (A) 0 0
30-40 20 35 10 200
40-50 17 45 20 340
50-60 6 55 30 180
N  100
 f d  300
i i

Let A  25
300
 Arithmetic Mean  25   28
100
xi  A
(b) Change of Scale: Let di  , for all i  1,2,...n , where A is any constant and h is length of each class.
h
h
Then, x  A 
N f d
i i

Example (4): Find the Arithmetic Mean of the data given in Example (2) by change of scale.

Solution:

Marks No. of Students Mid values of xi  A fi di

di 
fi  Class intervals h
 xi 
0-10 12 5 –2 – 24
10-20 18 15 –1 – 18
20-30 27 25 (A) 0 0
30-40 20 35 1 20
40-50 17 45 2 34
50-60 6 55 3 18
N  100
 f d  30
i i

Let A  25 and h  10
h 10  30 
Then, x  A 
N  f d  25 
i i
100
 28

(c) Let the mean of m variables x1, x 2 ,...x m be x and the mean of y1,y 2 ,....yn be y . Then the mean of combined
mx  ny
data x1, x 2 .....xm ,y1,y 2 ....yn is equal to .
mn

Example (5): The average salary of a male employee is Rs 520 and that of a female is Rs 420. The mean salary of
all the employees is Rs. 500. Find the percentage of male and female employees.

Solution: Let the number of males and females be m and f respectively.

m  520   f  420 
Therefore, 500   m  4f
mf
 Males percentage = 80,
Females percentage = 20

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
Merits and demerits of A.M.:

Merits: (a) It is clearly defined.

(b) It is based up on all the observations.

Demerits: (a) A.M. cannot be obtained if a single observation is missing.

(b) It is affected very much by extreme values.
(c) It leads to wrong conclusions if the details of the data from which it is computed are not given.

(ii) Geometric Mean (G.M):

(a) The Geometric Mean of n observations x1, x 2 ,..xn is given by G   x1x 2 x3 .....xn 
1/ n
.
(b) In case of ungrouped frequency distribution xi | fi , i  1,2,...n the Geometric Mean is given by
n
 
1/ N
G  x1f1 x 2f 2 ...xnfn , where N  f .
i1
i

n
1
 logG 
N  f log x
i 1
i i

 
1/ N
(c) In case of grouped frequency distribution Geometric Mean is given by G  x1f1 x 2f 2 ...xnfn , where
n
N f
i1
i and x1,x 2 ,......xn are mid-values of the class intervals.

Note: Let G1 be the G.M. of x1, x 2 ,...x m and G2 be the G.M. of y1,y 2 ,....yn . Let G be the G.M. of combined data
x1, x 2 ,....xm ,y1,y 2 ....yn .
1/  m n  mlogG1  nlogG2
Then, G   x1x 2 ....xm y1y 2 y3 ...yn 
1/  m n 

 G  G1m Gn2   logG 
mn

Merits and demerits of G. M:

Merits: (a) It depends on all observations.

Demerits: (a) G.M. becomes zero if one of the observations is zero.
(b) Sometimes G.M. becomes imaginary if some of the observations are negative.

(iii) Harmonic Mean (H.M):

n n
(a) The Harmonic Mean of n observations x1, x 2 ,x3 ...x n is given by H.M.   .
1 1 1 n
1

x1 x 2
 .... 
xn i1
xi
(b) In case of ungrouped frequency distribution xi | fi , i  1,2,.....n , the Harmonic Mean is given by
n
N
H.M.  n
fi
, where N  f . i


i1
xi
i1

n
N
(c) In case of grouped frequency distribution, Harmonic Mean is given by H.M.  n
fi
, where N   f and i


i1
xi
i1

x1,x 2 ,......xn are mid-values of the class intervals.

Examples (6): A student starts from his house and reaches the college at speed of 10 kmph and back from college
to home at speed of 15 kmph. Find average speed.
Solution:
2x 2 t1
Vavg    12km / h x
t1  t 2 1 1 A B

v1 v 2
t2

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
Merits and Demerits of H.M:

Merits: It depends on all the observations.

Demerits: It is not easily understood and difficult to compute.

(iv) Median: Median of a distribution is the value of the variable which divides it into two equal parts.

(a) In case of ungrouped data, if the number of observations is odd then median is the middle value after values
have been arranged in ascending or descending order.

In case of even number of observations, there are two middle terms and median is obtained by taking A.M. of
middle two terms. In fact, any value lying between the middle two values can be taken as the median.

Eg: The median of 7, 4, 5, 9, 10 is 7.

The median of 7, 6, 3, 8, 10, 12 is 7.5

(b) In case of discrete (ungrouped) frequency distribution, median is obtained by considering the cumulative
frequencies. The steps for calculating median are given below.

f .
n
N
Step (1): Find , where N  i
2 i1
N
Step (2): See the less than cumulative frequency (l.c.f) just greater than .
2
Step (3): The corresponding value of x is the median.

Example (7): Find median of following data:

x: 1 2 3 4 5 6 7 8 9
f: 8 10 11 16 20 25 15 9 6

Solution:

x f c.f
1 8 8
2 10 18
3 11 29 N
4 16 45 Here,  60
2
5 20 65 Therefore, Median = 5
6 25 90
7 15 105
8 9 114
9 6 120
N=120

(c) In case of grouped frequency distribution the median is given by the following formula:
h N 
Median      c 
f  2 
where,  is the lower limit of the median class
h is the length of the median class
f is the frequency of the median class

f .
n
c is the cumulative frequency of the class preceding the median class and N  i
i1

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
Proof:

Class Frequency c.f

Intervals
x1  x 2 f1 F1
x 2  x3 f2 F2

xk1  xk fk1 (c) Fk1

(I) xk  xk 1 fk (f) Fk

xn  xn1 fn Fn
N  fi

Let xk  xk 1 be median class.

Now if we assume that the variable values are uniformly distributed over the median class.
Then, from the figure,

slopeof AB  slope of AC
N
F
Fk  Fk 1 2 k 1
 
h x  xk
N
c
f hN 
  2  x     c
h x f 2 

Example (8): Find median of following distribution:

Class interval : 20-30 30-40 40-50 50-60 60-70

Frequency : 3 5 20 10 5

Solution:
Class Frequency c.f
Intervals
20-30 3 3
30-40 5 8
40-50 20 28
50-60 10 38
60-70 5 43
N = 43

N
Here ,  21.5 ,   40 , h  10 , f  20 and c  8
2
10
 Median  40  21.5  8  46.75
20

Merits and demerits of median:

Merits: (a) it is clearly defined.

(b) it is easy to calculate.
(c) it is not effected by extreme values.

Demerits: (a) In case of even number of observations, median can’t be determined exactly.
KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
 (b) It does not depend on all the observations.

Uses: Median is the only average to be used while dealing with qualitative data which cannot be measured
quantitatively but still can be arranged in ascending or descending order of magnitude. For example, average
intelligence or average honesty.

(v) Mode: Mode is the value which occurs most frequently in a set of observations and around which other
observations of the set cluster densely.

Eg: The mode of 2, 2, 9, 1, 3, 3, 5, 6, 8, 9 is 2

Merits and demerits:

Merits: (a) it is easy to calculate.

(b) it is not affected by extreme values.

Demerits:
(a) Mode is ill-defined. For example, the data 1, 2, 7, 2, 3, 4, 5, 3 have two modes. This kind of data is called
bimodal.
(b) It is not based upon all the observations

Uses: Mode is an average to be used to find the ideal size.

Note: (a) For a symmetrical distribution mean, median and mode coincide.
For example, for the following data Mean = Median = Mode = 5

x f

1 1

2 3

3 5

4 3

5 1

(b) If the distribution is moderately asymmetrical Mean, Median and Mode obey the following empirical
relation.
Mode  3 Median   2 Mean 

5. Partition values: These are the values which divide the series into a number of equal parts.

(i) The three points which divide the series into 4 equal parts are called quartiles. The first, second and third points
are known as first, second and third quartiles, respectively. The first quartile ( Q1 ) is the value which exceed 25% of
observations and is exceeded by 75% of observations. The second quartile ( Q2 ) is the value which exceed 50% of
observations and is exceeded by 50% of observation. The third quartile ( Q3 ) is the value which exceed 75% of
observations and is exceeded by 25% of observation.

(ii) The nine points which divide the series into ten equal parts are called deciles. The first, second, …., ninth
deciles are represented by D1,D2 ,.....D9 respectively.

(iii) The ninety nine points which divide the series into hundred equal parts are called percentiles. The first, second,
percentiles are represented by P1,P2 ,..... respectively.

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
Example(9) : Calculate the median, quartiles, 4th decile and 27th percentile of following data:

x: 0 1 2 3 4 5 6 7 8
f: 1 9 26 59 72 52 29 7 1

Solution:
N
(i)  128  Median  4
2
x f c.f
0 1 1 N
(ii)  64  Q1  3
1 9 10 4
2 26 36 N
 128  Q2  4
3 59 95 2
4 72 167 3N
 192  Q3  5
5 52 219 4
6 29 248
4N
7 7 255 (iii)  102.4  D4  4
10
8 1 256
N  256 27N
(iv)  69.12  P27  3
100

Mode:

A   x k ,fk 1  , B   xk 1,fk 
D   xk ,fk  , C   xk 1,fk 1 
fk  fk 1
y  fk 1   x  xk 
h
f f
y  fk  k 1 k  x  xk 
h
x  xk
 fk  fk 1   2fk  fk 1  fk 1 
h
h  fk  fk 1 
 x  xk 
2fk  fk 1  fk 1
h  fk  fk 1 
 Mode   
2fk  fk 1  fk 1

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
2. Measures of Dispersion

1. Measure of dispersion:
Consider the series (i) 7, 8, 9 (ii) 8, 8, 8 (iii) 4,8,12

In all these cases, we see that number of observations is 3 and the mean is 8. If we are given that, the mean of 3
observations is 8, we cannot form an idea as to whether it is the average of first series or second or third series or of
any other series of 3 observations whose sum is 24. Thus, we see that measure of central tendency is inadequate to
give us a complete idea of the distribution. They must be supported and supplemented by some other measures.
One such measure is dispersion.

The following are the measures of dispersions:

(i) Range,
(ii) Quartile deviation (or) semi-interquartile range,
(iii) Mean deviation,
(iv) Standard deviation, and
(v) Root mean square deviation.

Range: The range is the difference between two extreme observations of the distribution. If A and B are the greatest
and smallest observations respectively in a distribution, then its range is A-B.

Range is the simplest, but a crude measure of dispersion. Since it is based on two extreme observations, it is not at
all a reliable measure of dispersion.

1
(ii) Quartile Deviation: Quartile deviation Q is given by Q   Q3  Q1  , where Q1 and Q3 are the first and third
2
quartiles of the distribution respectively.

Quartile deviation is definitely a better measure than range as it makes use of 50% of the data but since it ignores
the other 50 % of the data, it cannot be regarded as a reliable measure.

(iii) Mean Deviation: Consider n observations x1, x 2 ,x3 ,.....xn . Then mean deviation from the average A (usually
n
1
A.M, median or mode) is given by
n  x A .
i 1
i

 1
n n
In case of frequency distribution xi | fi , i  1,2,......,n , Mean deviation =
i 1
fi xi  A  , where N 

 i1
fi . 
N   
Since mean deviation is based on all observations, it is better measure of dispersion than quartile deviation.

(iv) Standard Deviation: The standard deviation of n observations x1, x 2 ,x3 ...x n is denoted by  , and it is given by

 x1  x    x 2  x   ....  xn  x 
2 2 2
1
x  x
2
  i , where x  AM of x1,x 2 .....xn .
n n

In case of frequency distribution xi | fi , i  1,2,....n standard deviation is given by

n n
1
 fi  xi  x  , where N  f .
2
 i
N i 1 i1

The square of standard deviation is called variance ( 2 )

n n
1
 fi  xi  x  , where N  f .
2
2  i
N i1 i1

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
(v) Root mean square deviation: The root mean square deviation of n observations x1,x 2 ,....xn about any constant
A is denoted by ‘s’ and it is given by

 A  x1 2   A  x2 2  .....   A  xn 2 1
n

A  x 
2
s  i .
n n i 1

n n
1
 fi  xi  A  , where N  f .
2
In case of frequency distribution xi | fi , i  1,2,.....,n, we have s  i
N i 1 i1

s2 is called Mean square deviation.

2. Various formulae for Variance:

n
1
x  x
2
(a) 2  i
n i1
n n 2
1 1 x

n  x   2x  n  x  n n
i1
2
i
i1
i

n
1

n x
i1
2
i i  2x  x   x 2

n
1
 2 
n x i 1
2
i  x2

(b) The shifting of origin doesn’t change value of the variance i.e variance is independent of shifting the origin.

Proof: Let 2x be the variance of n observations x1 , x 2 .....xn .

n
1
 x  x
2
Then, 2x  2
i
n i 1
Let yi  xi  k , for all i  1, 2,...., n and k is a constant .
Let y be the variance of n observations y1 , y 2 .....yn .
2

n
1
 y   y
2
 2y  2
i
n i 1
n
1
x  k  x  k 2 2
 i
n i 1
n n n
k2
1   x 
1 2k

n 
i 1
xi2 
n  i1
xi 
n i1
2
 2kx  k 2

n
1
 x  x 2 2
 i  2x
n i1

(c) The variance is depending on change of scale

Proof: Let 2x be the variance of n observations x1 , x 2 .....xn .
n
1
 x  x
2
Then, 2x  2
i
n i 1
Let yi  hx i , for all i  1, 2,...., n and h is a non zero constant .
Let y be the variance of n observations y1 , y 2 .....yn .
2

1
n
1
n 1 n 
 yi2   y    hxi   hx   x   x
2 2 2 2
 2y   h2  2
  h2 2x
n n n i 
i 1 i 1  i 1 

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
(d) Relation between s and  :
n
1
x  A
2
s2  i
n i 1
n
1
x  x  x  A
2
 i
n i 1
n
1
  x  x  d
2
 i , where d  x  A
n i1

  x  x  
n
1
 d2  2d  xi  x 
2
 i
n i 1

1
n 1 n 1
n 
x  x   x 
2
  d2  2d  xi 
n
i n n
i 1  i1 i 1

   d  2d  x  x     d2
2 2 2

 s2 = σ2 + d2 where, d2   x  A 
2

3. Coefficient of Dispersion:
Whenever we want to compare, the variability of two series which differ widely in their averages or which are
measured in different units, we do not merely calculate the measure of dispersion but we calculate the coefficients
of dispersion. The coefficient of dispersion (C.D) based on different measures of dispersion are as follows:

A B
(i) Based upon the range: C.D = ,where A and B are greatest and the smallest items in the series.
A B
Q  Q1
(ii) Based upon Quartile deviation: C.D  3
Q3  Q1

Mean deviation
(iii) Based upon Mean deviation: C.D 
Average from which it is calculated


(iv) Based upon standard deviation: C.D =
x

4. Coefficient of variation (C.V.): The coefficient of variation is given by C.V   100
x

Examples:

1. Find the mean deviation about the mean for 38, 70, 48, 40, 42, 55, 63, 46, 54, 44
Sol: Given data is 38, 70, 48, 40, 42, 55, 63, 46, 54, 44.
38+70  48+40  42+55+63+46+54+44 500
The arithmetic mean of the given data is, x    50
10 10
Then the absolute values |xi  x| are 12, 20, 2, 10, 8, 5, 13, 4, 4, 6
10

|xi  x| 84
i 1
 Mean deviation about the mean =   8.4
10 10

2. Find the mean deviation about the median for 13, 17, 16, 11, 13, 10, 16, 11, 18, 12, 17
Sol. Given data is 13, 17, 16, 11, 13, 10, 16, 11, 18, 12, 17.
On keeping the observations in ascending order 10, 11, 11, 12, 13, 13, 16, 16, 17, 17, 18
The median (b) of the given 11 observations is 13.
Then the absolute values |xi  b|are 3, 2, 2, 1, 0, 0, 3, 3, 4, 4, 5.

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
 11

|xi  b| 27
i 1
Mean deviation about the median    2.45
11 11
3. Find the mean deviation about the mean for the following distribution.
xi 10 11 12 13
fi 3 12 18 12

Sol. Construct the following data to find mean deviation about mean
xi 10 11 12 13
fi 3 12 18 12
xi f i 30 132 216 156

|xi  x| 1.87 0.87 0.13 1.13

f i |xi  x| 5.61 10.44 2.34 13.56

Arithmetic mean, x 
 fi xi  534  11.87
 fi 45
4

 fi |xi  x| 45  31.95  0.71

1 1
Mean about the mean 
 fi i 1

4. Find the mean deviation about the median for the following frequency distribution.
xi 5 7 9 10 12 15
fi 8 6 2 2 2 6

Sol. We form the following table from the given data

xi fi c.f xi  M f i xi  M
5 8 8 2 16
7 6 14 0 0
9 2 16 2 4
10 2 18 3 6
12 2 20 5 10
15 6 26 8 48
N  26 84
N  26 
    13 (Value in the ‘c.f column’ which is just greater than 13 is 14). Median M  7
2  2 

Mean deviation about median M .D 

 fi xi  M  84  3.23
N 26

5. Find the mean deviation about the mean for the following data:
Marks obtained 0-10 10-20 20-30 30-40 40-50
No. of students 5 8 15 16 6

Sol. We take the assumed mean A  25 , Here, C  10 . Hence, we form the following table
xi  25
frequency  f i 
Class Midpoint
di  f i di xi  x f i xi  x
interval xi 10

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
 0 – 10 5 5 2 10 22 110
10 – 20 8 15 1 8 12 96
20 – 30 15 25A 0 0 2 30
30 – 40 16 35 1 16 8 128
40 – 50 6 45 2 12 18 108

 fi  50  N  fi di  10  fi xi  x  472
Here, N  50 . Mean x  A  C 

  fi di   25  10 10   25  2  27
   50 
 N 
5

 fi xi  x  50 472  9.44
1 1
 Mean deviation about the mean M.D 
N i 1

6. Find the mean deviation from the median for the following data
Age (years) 20-25 25-30 30-35 35-40 40-45 45-50 50-55 55-60
No. of workers  f i  120 125 175 160 150 140 100 30

Sol. We form the following table for the given data:

Class frequency Cumulative Mid point

xi  med. f i xi  med.
interval
 fi  frequency
c.f.
xi

20 – 25 120 120 22.5 15 1800

25 – 30 125 245 27.5 10 1250
30 – 35 175 420 32.5 5 875
35 – 40 160 580 37.5 0 0
40 – 45 150 730 42.5 5 750
45 – 50 140 870 47.5 10 1400
50 – 55 100 970 52.5 15 1500
55 – 60 30 1000 57.5 20 600

8
N = 1000  fi xi  med.  8175
i1
N 1000
Here th observation   500th observation lies in the class interval 35  40. This is the
2 2
median class.
 500  420 
 Median  L   N /2  p.c. f  / f  i  35    5
 160 
400
 35   35  2.5  37.50
160
8

 fi xi  med.  1000 8175  8.175

1 1
 Mean deviation about median 
N i 1

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.
Some Important results to be remembered:
nx  xi  x '
1. The A.M. of n numbers x1, x 2 ,x3 ,......, xn is x . If xi is replaced by x ' , then new average is .
n

2. Let x, 2 ,  are Mean, Variance and standard deviation of x1, x 2 ,x3 ,......, xn then,

Mean ( x ) Variance ( 2 ) Standard Deviation (  )

x1  k,x 2  k,x 3  k,......, xn  k x k 2 

k.x1,k.x 2 ,k.x 3 ,......,k.xn (k  0) k.x k 2 .2 k .

a. x1  b,a. x 2  b,a.x 3  b,......,a. xn  b ax  b a2  2 a .

3. Sum of deviations of variants from their A.M. is always zero. i.e.  x  x  0,  f x  x  0 .

i i i

 x  A
2
4. Sum of squares of deviations of variants is minimum when taken about the mean. i.e. if i is

minimum then A  x .

5. A.M. of a combined series: Let x and y are the A.M.s of two series of sizes m and n and X be the mean of

m.x  n.y
the combined data then X  .
mn
6. G.M. of a combined series: Let G1 and G2 are G.M.s of two series of sizes m and n and if G is geometric

m.logG1  n.logG2
mean of combined series, then logG  .
mn

7. Variance of a combined series: If m,n are the sizes, x, y the means and 1, 2 the standard deviation of two

1 m.x  n.y
series, then 2  [m.(12  d12 )  n.(22  d22 )] , where d1  x  X , d2  y  X and X  .
mn mn

n2  1
8. The variance of first n natural numbers is .
12

 xi  a  n and  x  a
n n
 na n,a  1 then the standard deviation of ‘n’ observations x1,x 2 ,,xn
2
9. If i
i1 i1

is a 1 .
10. If the distribution is moderately asymmetrical Mean, Median and Mode obey the empirical relation
Mode = 3.Median – 2.Mean.

KUKATPALLY CENTRE: # 22-97, Plot No.1, Opp. Patel Kunta Huda Park, Vijaynagar Colony, Hyderabad - 500 072. Ph.: 040-64601123
MIYAPUR CENTRE: Above Sai Motors Maruthi Showroom, Allwyn X Road, Miyapur, Hyderabad.