Chapter-1 Basic statistics - I

 MEASURE OF DISPERSION
 Quartiles, Deciles & Percentiles
 Quartile & Mean Deviation
 Standard Deviation
 Variance

 Quartiles, Deciles & Percentiles:

 For individual observation:

 For Quartiles Qk , where k = 1,2,3

 For Deciles Dk , where k = 1, 2,…,9
 For Percentiles Pk , where k = 1,2,…,99

 Rearrange the data in increasing order. Let n be the number of

observations.
nk
 For Qk , Find N= 4 ; k = 1,2,3
nk
 For Dk , Find N= 10 ; k = 1,2,…,9
nk
 For Pk , Find N= 100 ; k = 1,2,…,99

 If N is not an integer then find the value at the place of next integer
from N( greater than N)

 If N is an integer then find the mean value at the N th and ( N +1 )th places.
N +( N +1)
Q=
2

 For simple frequency distribution:

The value of the variable corresponding to the cumulative frequency just

greater than (N the total of frequencies)

NK
=Q k where k = 1,2,3
4
NK
=D k where k = 1,2,…,9
10
NK
=Pk where k = 1,2,…,99
100
 Chapter-1 Basic statistics - I

 For grouped frequency distribution:

Qk =L+
h
f [ kN
4 ]
−F ; k = 1,2,3

Dk = L+
h
f [ kN
10 ]
−F ; k = 1,2,…,9

Qk =L+
h
f [ kN
100 ]
−F ; k = 1,2,…,99

 Measure of Dispersion:

Range = largest value – smallest value or

Range = upper limit of the highest value – lower limit of the smallest
class
 Quartile Deviation:
 Interquartile range = Q3−Q1
Q3−Q1
 Quartile Deviation (Q.D) =
2
Which is called semi-quartile range.
 Co-efficient of inter quartile range or co-efficient of quartile
Q3−Q1
deviation = Q +Q
3 1

(Relative measure for comparative studies of variability of two

distributions)
 If two distributions have too much variability then, we can have
Q.D.
co-efficient of variation = Median ×100

 Mean Deviation(M.D):

The range and quartile deviations, as we have seen, are not based on
all observations, to remove such obstructions. We shall now see the
mean deviation also called as “Average Deviation” or “first moment
of dispersion”.
 Let x 1 , x 2 ,… , x n be the n – observations and let A be the
mean ( μ ¿ or Median (M) or Mode (Z).
n
1
M . D= ∑ |x i− A|
n i=1
M.D
Co-efficient of M.D ¿ A
 For simple frequency distribution,
 Chapter-1 Basic statistics - I

n n
 M . D= 1 ∑ f i|x i− A| where, n=∑ f i
n i=1 i=1
M.D
 Co-efficient of M.D ¿ A
 For Grouped frequency distribution:
for grouped frequency distribution x i will be taken as class-Mark
NOTE:
n
1
1. Mean deviation about Mean = ∑|x −μ|
n i=1 i
M.D
Co-efficient of mean deviation about mean = μ
n
1
2. Mean deviation about Median = ∑ |x i−M |
n i=1
M.D
Co-efficient of mean deviation about median = M
n
1
3. Mean deviation about Mode = ∑ |x i−mode|
n i=1
M.D
Co-efficient of mean deviation about mode = mode
n
1
4. Mean deviation about Mean = ∑ f i| xi −μ|
n i=1
M.D
Co-efficient of mean deviation about mean = μ
n
1
5. Mean deviation about Median = ∑ f |x −M|
n i=1 i i
M.D
Co-efficient of mean deviation about median = M
n
1
6. Mean deviation about Mode = ∑ f |x −mode|
n i=1 i i
M.D
Co-efficient of mean deviation about mode = mode
7. It is important to note that mean deviation about median is least compared
to the mean deviation mean or mode.
8. The sum of the deviations is always zero.
n

ie, ∑ [ x i−μ ]=0

i=1

 Standard Deviation(S.D):
 Chapter-1 Basic statistics - I

Standard Deviation is the positive square root of the mean of the squares of
the deviations of the given values from their mean. It is denoted by the
Greek letter σ .
 For individual observations:

√ Σ ( x−μ )
2 Σ xi
σ= where, μ=
n n
 For frequency distributions:

√
2
Σf ( x−μ ) Σ fx
σ= where, μ=
Σf Σf

 Chapter-1 Basic statistics - I

 Standard Deviation by step-deviation method:

√ ( )
Σ f d 2 Σ fd x i− A
2
σ= − where, d i= (A = Assumed mean)
n n h

 When we have ‘n’ data values that are:

 Population: Divide by ‘n’ when calculate by standard deviation

and variance.
 A Sample: Divide by n – 1(instead of n) when calculating
standard deviation and variance.
 Variance: the variance is the square of the standard deviation
and denoted by σ 2 .
2
Σ ( x−μ ) Σx
1. σ =¿2
where, μ= i (For individual observations)
n n
2
Σf ( x−μ ) Σ fx
2. σ 2= where, μ= Σf (For simple frequency distribution)
Σf

( ) x −A
2 2
2 Σf d Σ fd
3. σ = − where, d i= i (A = Assumed mean)
n n h
(For Step-Deviation Method)
 Co-efficient of variance:
The Standard Deviation is an absolute measure of
dispersion. The co-efficient of variation is a relative
measure of dispersion and is denoted by co-efficient of
variance (C.V)
σ
C . V = ×100
μ