Properties ± describing quantitative

data
 Numerical values of an observation around which
most numerical values of other observations in the
data set show a tendency to cluster or group
 Extent to which values are dispersed around the
central value called variation.
 Extent of departure of numerical values from
symmetrical distribution around the central value
called skew ness
aequisites of a measure of central
tendency
! It should be rigidly defined
! It should be based on all the observations
! Easy to understand and calculate
! Should have sampling stability
! Should not be unduly affected by extreme
observation
 MEASURES OF CENTRAL
TENDENCY
˜

 


 The Mode
 The Median



˜


 The Mean
=






 The Positively Skewed Distribution
 The Negatively Skewed Distribution
 Mode
 ˜ measure of central tendency
 Value that occurs most often
 Not affected by extreme values
 Used for either numerical or categorical data
 There may be no mode or several modes

Mode = 9 No Mode

Ô 1 2 3 4 5 6
Ô 1 2 3 4 5 6 7 8 9 1Ô 11 12 13 14

ode ± measure of location recognized by the
location of the most frequently occurring
value of a set of data

 Sales during 20 days period

 53,56,57,58,58,60,61,63,63,64,64,65,65,67,
68,71,71,71,71,74 (ascending order data)

Mode

ode for frequency distribution
Sales Volume (Class No. of Days (Frequency)
Interval)
53-56 2
57-60 4
61-64 5
65-68 4
69-72 4
72 and above 1
Frequency distribution of sales per day

ode: The Category or Score with the
Largest frequency(or %)

 The mode is always a category or score

 The mode is not necessarily the category
with the majority(more than 50% of the
cases)
 The mode is the only measure of central
tendency for nominal variables
 Some distributions are bimodal
Mode for grouped data,

0 = L + O
O  
2O
O 
O 
THE MEDIAN ± measuring
qualitative characters
 The median is a measure of central
tendency for variables which are at least
ordinal.

 The median represents the exact middle

of a distribution.

 It is the score that divides the

distribution into two equal parts
 Finding the Median in sorted data
³How satisfied are you with your health insurance?
Responses of 7 Individuals
very dissatisfied
very satisfied
somewhat satisfied
very dissatisfied
somewhat dissatisfied
somewhat satisfied
very satisfied

Total(N) 7
To locate the median
˜
















very dissatisfied
very dissatisfied
somewhat dissatisfied
somewhat satisfied ( = 
  )
somewhat satisfied
very satisfied
very satisfied
_________________________________________________
Summary :Locating the
edian with
N=Odd
 The median is the response associated with the
middle case.
 You find the middle case by :(N + 1) 2
 Since N= 7, the middle case is the (7 + 1)
2, or the 4th case
 The response associated with the 4th case is
³somewhat satisfied´. Therefore the median is:
Somewhat satisfied.
To locate the median (N=Even)









7.44, 1Ô.ÔÔ, 12.26, 12.61, 13.38, 14.11, 14.3Ô, 14.78

The median is located halfway between

the two middle cases. When the variable
is interval we can average the two middle
cases.
Median = 12.61 + 13.38 = 12.99
2
 Median
 aobust measure of central tendency
 Not affected by extreme values

Ô 1 2 3 4 5 6 7 8 9 1Ô Ô 1 2 3 4 5 6 7 8 9 1Ô 12 14

Median = 5 Median = 5
 In an ordered array, the median is the
³middle´ number
If n or N is odd, the median is the middle number
If n or N is even, the median is the average of the
two middle numbers

edian for grouped data
˜

˜

 

 
 



Ô
 

 


edian
 
 Ô
   
class
 Ô  Ô


Med = L + (VX
±


 Partition Values:
Quartiles, Deciles, and Percentiles
·uartiles ± Divide an ordered data set into 4
equal parts - 2nd ·uartile -
edian
Deciles ± Divide an ordered data set into 10
equal parts - 5th Decile -
edian

Percentiles ± Divide an ordered data set into

100 equal parts - 50th Percentile -
edian
·     ,
QÕ = L + i(VX
±
Õ


ï
    

ïÕ = L + i(VX
±
Õ


 
     

Õ = L + i(VX
±
Õ



_____________________________
Mean. The arithmetic average obtained by
adding up all the scores and dividing by the
total number of scores.
___________________________________________________________
Objectives of an ˜verage
 Determine one single value that may be
used to describe the character sticks of
entire series.
 Facilitate comparison at a particular point of
time
 Facilitate statistical inference
 Helps in decision making process
The Mean
_________________________________________________________________
Mean. The arithmetic average obtained by adding up all the scores and
dividing by the total number of scores.
_________________________________________________________________

´
Ô
Y = raw scores of the variable y
__
Y = the mean of y

Y = the sum of all the y scores

N = the number of observations

Crime aate in Indian Cities:Finding the
ean

 




 !!!
 "#$%

 "&$#

 %"$#
  %'$(

 "(
  )$*
  (&$)
 )&$&
  "$&
 
" $&
 

 %$)
 '$'
 )!$'
 "$#
  #$&




%#"$)

392.4
 Ë Ë Ë Ë 26.16
Ô 15
 Sample statistic ± a numerical value used as
a summary measure using data of the
sample for estimation or hypothesis testing

 Population parameter - a numerical value

used as a summary measure using data of
the population
Mean (Arithmetic Mean)

ean (arithmetic mean) of data values

Sample mean
 j
j

 Ë1 1
 2
L
 
Ë Ë
 
Population mean
Ô
 
  j

 Ë1 1
 2
L
 Ô
·Ë Ë
Ô Ô
˜rithmetic mean of ungrouped raw
data
 Direct method
 Indirect method (short cut method)
 Finding the mean in a frequency distribution
When data are arranged in a frequency distribution, we must
give each score its proper weight by multiplying it by its
frequency. We use the following formula to calculate the mean:

__
Y =ȈY


where
__
Y = the mean
Y = a score multiplied by its frequency
Ȉ Y = the sum of all the  Y¶s
 = the total number of cases in the distribution
Calculating the
ean from a
Frequency Distribution
>
 
  
   
   
  
  
  
  
  
  
  
  

  

O 3526
Ë Ë Ë 2 .6
Ô 1339
Î

 




 Used when values are grouped by frequency or

relative importance

Example: Sample of
26 Repair Projects
Weighted Mean Days
Days to Frequency to Complete:
Complete
5 4 XW Ë
w x
i i
Ë
(4
5) (12
6) (8
7) (2
8)
6 12 w i 4 12 8 2
7 8 164
Ë Ë 6.31 days
8 2 26
Indirect method
 The human resource manager at a city
hospital began a study of the overtime hours
of the registered nurses. Fifteen nurses were
selected at random and following overtime
hours were recorded during a month:
 13 13 12 15 17 15 5 12 6 7 12 10 9 13 12
 5 9 6 10 5 6 9 6 9 12
˜rithmetic mean of grouped (classified) data
Direct & Step deviation method)
The following distribution gives the pattern of overtime work
done by 100 employees of a company. Calculate the average
overtime work done per employee

  

 
 



  
     
     
     
     
     
     

ƒeometric Mean
Geometric
ean of a set of  numbers is
defined as the  root of the product of
the  numbers and is used to average
percents, indexes, and relatives.
 The formula is: ( È 0)

 ƒ Ë  1   2 L  

ore directly measures the change over
more than one period
 Geometric
ean w ˜rithmetic
ean
30
aelationship between
ean,

edian and
ode

0 = 3
edian ± 2
ean
Oa


ean ±
ode = 3 (
ean ±
edian)
The Shape of Distributions
 Distributions can be either symmetrical
or skewed, depending on whether there
are more frequencies at one end of the
distribution than the other.

 
Symmetrical
Distributions
 A distribution is symmetrical if the
frequencies at the right and left tails of
the distribution are identical, so that if it
is divided into two halves, each will be the
mirror image of the other.
 In a unimodal symmetrical distribution
the mean, median, and mode are
identical.
1.4. Shape of a Distribution

 Describeshow data is distributed

 Symmetric or skewed

Left-Skewed Symmetric Right-Skewed

Mean < Median < Mode Mean = Median = Mode Mode < Median < Mean
ÿ

 

 ÿ

 
 
 Choosing a
easure of Central Tendency

 IF variable is Nominal..
±
ode
 IF variable is Ordinal...
±
ode or
edian(or both)
 IF variable is Interval-aatio and distribution is
Symmetrical«
±
ode,
edian or
ean
 IF variable is Interval-aatio and distribution is
Skewed«
±
ode or
edian
Calculate the mean, median and mode for the
following data pertaining to marks in statistics.
There are 8Ô students in class and the test is of
14Ô marks.
Marks more than No. of Students
Ô 8Ô
2Ô 76
4Ô 5Ô
6Ô 28
8Ô 18
1ÔÔ 9
12Ô 3