CHAPTER

22 Statistics

ARITHMETIC MEAN n th

(b) If n is even, then there are two middle terms namely  
(i) Arithmetic Mean for Unclassified (Ungrouped or Raw) 2
Data: If there are n observations, x1, x2, x3, ... , xn, then their  n 
and  + 1 th terms, median is mean of these terms.
arithmetic mean 2 
n
Median for Classified (Grouped) Data or Grouped
x1 + x2 + ... + xn ∑x i
=A or x = i =1 Frequency Distribution
n n
For a continuous distribution, median
(ii) Arithmetic Mean for Discrete Frequency Distribution
N
or Ungrouped Frequency Distribution: Let f1, f2,..., fn be −C
corresponding frequencies of x1, x2,...,xn. Then, arithmetic l+ 2
Md = ×h
mean f
n
where, l = lower limit of the median class

x1 f1 + x2 f 2 + ... + xn f n ∑x f i i f = frequency of the median class

=A = i =1
n n
f1 + f 2 + ... + f n
∑ fi N = total frequency = ∑f
i =1
i
i =1

C = cumulative frequency of the class just before the

(iii) Arithmetic Mean for Classified (Grouped) Data or median class
Grouped Frequency Distribution: For a classified data,
we take the class marks x1, x2,... , xn of the classes, then h = length of the median class
arithmetic mean by Mode for Classified (Grouped) Distribution or
n

∑x f i i
Grouped Frequency Distribution
A= i =1 f 0 − f1
n Mo =
l+ ×h
∑f
i =1
i
2 f 0 − f1 − f 2
where, l = lower limit of the modal class
Combined Mean: If A1, A2,..., Ar are means of n1, n2,...,nr
f0 = frequency of the modal class
observations respectively, then arithmetic mean of the combined
group is called the combined mean of the observation. f = frequency of the pre-modal class
r f = frequency of the post-modal class
n1 A1 + n2 A2 + ... + nr Ar ∑n A i i h = length of the class interval
=A = i =1
r
n1 + n2 + ... + nr Relation Between Mean, Median and Mode
∑ni =1
i

(i) Mean – Mode = 3 (Mean – Median)

MEDIAN (ii) Mode = 3 Median – 2 Mean

Median for Simple Distribution or Raw Data MEAN DEVIATION (MD)

Firstly, arrange the data in ascending or descending order and then n
find the number of observations n. ∑ x −x
i

 n +1 (i) For simple (raw) distribution, δ = i =1

(a) If n is odd, then   th term is the median. n
 2 
where, n = number of terms, x = A or Md or Mo
 n Standard Deviation of the Combined Series
∑f i xi − x
(ii) For unclassified frequency distribution, δ = i =1 If n1, n2 are the sizes, X 1 , X 2 are the means and s1, s2, are the
n
standard deviation of the series, then the standard deviation of the
∑f i
i =1 combined series is
n

∑f i xi − x n1 (σ12 + d12 ) + n2 ( σ 22 + d 22 )
(iii) For classified distribution, δ = i =1 σ=
n n1 + n2
∑f
i =1
i

where, d1 =X 1 − X ,d 2 =X2 − X
where, xi is the class mark of the interval.
n1 X 1 + n2 X 2
STANDARD DEVIATION AND VARIANCE and X =
n1 + n2
(i) For simple distribution
n

∑ (x − x ) 2
2
IMPORTANT POINTS TO BE REMEMBERED
i
1 n
 n 
=σ i =1
= n∑ xi 2 −  ∑ xi  (i) The ratio of SD (s) and the AM ( x ) is called the coefficient
n n
= i 1= i 1 
σ
where, n is a number of observations and x is mean. of standard deviation  
x
(ii) For discrete frequency distribution σ
(ii) The percentage form of coefficient of SD i.e.   × 100 is
n
called coefficient of variation. x
∑ f (x − x )
1 i n
2

 n 
2

=σ =
i =1
N ∑ f i xi 2 −  ∑ f i xi  (iii) The distribution for which the coefficient of variation is less
N N
=i 1 = i 1 
is more consistent.
(iii) For continuous frequency distribution
n n2 − 1
(iv) Standard deviation of first n natural numbers is .
∑ f (x − x )
i i
2
12
σ= i =1

N (v) Standard deviation is independent of change of origin, but it

where, xi is class mark of the interval. depends on change of scale.

P Statistics 41
W